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>

package/libesmtp: requires DES support in openssl

Enable DES in openssl to fix build failure raised since bump to version
1.1.0 in commit b72401692e1ea73a37fcf4c0b421b654e8a2100f

Fixes:
 - http://autobuild.buildroot.org/results/52bce321552d23c10617c4ba4dc5b290e1d43dd5

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

package/s390-tools: properly fix build with libcurl

Update third patch to fix the following build failure:

/bin/sh: 1: curl-config: not found
make[2]: Leaving directory '/tmp/instance-2/output-1/build/s390-tools-2.17.0/libvmdump'
Error: libcurl is not built with the OpenSSL backend
make[2]: *** [Makefile:54: check-dep-libekmfweb] Error 1

Fixes:
 - http://autobuild.buildroot.org/results/b4ebe39e54f80b26bc81b249bfa643140c41d0bd

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

package/mutt: link with -latomic if needed

Link with -latomic if needed to avoid the following build failure with
openssl:

configure:10549: checking for X509_STORE_CTX_new in -lcrypto
configure:10574: /tmp/instance-0/output-1/host/bin/sparc-linux-gcc -o conftest -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64  -Os -g0  -static -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -I/tmp/instance-0/output-1/host/sparc-buildroot-linux-uclibc/sysroot/usr/include  -static -L/tmp/instance-0/output-1/host/sparc-buildroot-linux-uclibc/sysroot/usr/lib conftest.c -lcrypto -lz  >&5
/tmp/instance-0/output-1/host/opt/ext-toolchain/bin/../lib/gcc/sparc-buildroot-linux-uclibc/9.3.0/../../../../sparc-buildroot-linux-uclibc/bin/ld: /tmp/instance-0/output-1/host/sparc-buildroot-linux-uclibc/sysroot/usr/lib/libcrypto.a(x509cset.o): in function `X509_CRL_up_ref':
x509cset.c:(.text+0xfc): undefined reference to `__atomic_fetch_add_4'

Ideally, mutt should use pkg-config but upstream don't want to use it:
https://gitlab.com/muttmua/mutt/-/merge_requests/25

Fixes:
 - http://autobuild.buildroot.org/results/e357984853f8ca879156245717fadbeaa9b0dbae

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

package/xen: fix build with gcc 11

Fix the following build failure with gcc 11:

linux.c:165:50: error: argument 7 of type 'const xen_pfn_t[]' {aka 'const long long unsigned int[]'} declared as an ordinary array [-Werror=vla-parameter]
  165 |                                  const xen_pfn_t arr[/*num*/], int err[/*num*/])
      |                                  ~~~~~~~~~~~~~~~~^~~~~~~~~~~~
In file included from linux.c:29:
private.h:35:50: note: previously declared as a variable length array 'const xen_pfn_t[num]' {aka 'const long long unsigned int[num]'}
   35 |                                  const xen_pfn_t arr[num], int err[num]);
      |                                  ~~~~~~~~~~~~~~~~^~~~~~~~

Fixes:
 - http://autobuild.buildroot.org/results/f70d060cf50254694113f19f50a8ef96ef33dd1a

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

package/mpd: needs gcc >= 8

Since bump to version 0.22.3 in commit b6576a458ca0 (package/mpd: bump
to version 0.22.3), mpd needs gcc >= 8, as documented in their manual
[0], to avoid the following build failure with gcc 7.3.1:

/tmp/instance-7/output-1/host/opt/ext-toolchain/aarch64-linux-gnu/include/c++/7.3.1/bits/stl_tree.h:2091:28: error: no matching function for call to 'std::_Rb_tree<std::__cxx11::basic_string<char>, std::pair<const std::__cxx11::basic_string<char>, std::__cxx11::basic_string<char> >, std::_Select1st<std::pair<const std::__cxx11::basic_string<char>, std::__cxx11::basic_string<char> > >, std::less<std::__cxx11::basic_string<char> >, std::allocator<std::pair<const std::__cxx11::basic_string<char>, std::__cxx11::basic_string<char> > > >::_M_get_insert_unique_pos(std::pair<std::basic_string_view<char>, std::basic_string_view<char> >::first_type&)'
  = _M_get_insert_unique_pos(_KeyOfValue()(__v));
    ~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~

Fixes:
 - http://autobuild.buildroot.org/results/4888d99404cc4273349ab036035c5ff7e086b83e

[0] https://mpd.readthedocs.io/en/stable/user.html#compiling-from-source)

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
[yann.morin.1998@free.fr: reword commit log to reference the manual]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/qemu: disable vhost-user with uclibc-ng

Disable vhost-user on uclibc-ng as it raises build failures since bump
to version 6.0.0 in commit 6b86c9335fc3ff381878156c6243454d4b688df9 and
https://github.com/qemu/qemu/commit/810033be083040591fee2fc09f2f294670ed1930

The issue is that F_ADD_SEALS and F_SEAL_xxx variables are not defined on
uclibc-ng (even if MFD_ALLOW_SEALING is defined) resulting in the
following build failure:

../subprojects/libvhost-user/libvhost-user.c:1637:22: error: 'F_ADD_SEALS' undeclared (first use in this function)
 1637 |     ret = fcntl(*fd, F_ADD_SEALS, flags);
      |                      ^~~~~~~~~~~

Fixes:
 - http://autobuild.buildroot.org/results/9f393539931b51191cf7128e5c618c2398fc86c1

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
[yann.morin.1998@free.fr: use conditional block like for all other OPTS]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/ltp-testsuite: renumber patch

Commit 84968aa4954b308ed67adde305e0fc5c801edc9e forgot to renumber
remaining patch

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

package/libesmtp: security bump to version 1.1.0

After more than a decade, libESMTP version 1.0.6 is superceded. Despite
proving robust a little bitrot has occurred, especially regarding
OpenSSL support. The original application data APIs are prone to memory
leaks and are deprecated in favour of safer replacements. Version 1.1
updates libESMTP without breaking API and ABI compatibility and
provides a basis for future development.

In addition to updates to the codebase, documentation is modernised and
is more comprehensive.

All libESMTP users are encouraged to upgrade from version 1.0.6.

- Update license files
- Update indentation in hash file (two spaces)
- Switch to meson-package
- Handle threads and tls meson options
- libesmtp-config has been dropped:
  https://github.com/libesmtp/libESMTP/issues/8
- Fix CVE-2019-19977: libESMTP through 1.0.6 mishandles domain copying
  into a fixed-size buffer in ntlm_build_type_2 in ntlm/ntlmstruct.c, as
  demonstrated by a stack-based buffer over-read.

https://github.com/libesmtp/libESMTP/releases/tag/v1.1.0
https://libesmtp.github.io/changes-since-v1.0.6.html

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

package/libqrtr-glib: add gobject-introspection optional dependency

gobject-introspection is an optional dependency (enabled by default)
since the addition of the package in commit
ea64e05a1bfbe3b258a69013462f94d7db7a9c0b

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/libqmi: add gobject-introspection optional dependency

gobject-introspection is an optional dependency (enabled by default)
since version 1.26.0 and
https://gitlab.freedesktop.org/mobile-broadband/libqmi/-/commit/2aa0badc79595c92981793d18f05c99485156a6b

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

package/libebml: fix patch

Fix patch added by commit 1116a3060955c36c43a8bea5f1f00ea4effb4e48

Fixes:
 - http://autobuild.buildroot.org/results/997d0d6204ad7a5456f25392257c8d876a1289ca

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

docs/manual/migrating.txt: add section with general migrating tips

This is based on Yann's and Arnout's experience with migrating
Buildroot.

Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Cc: Yann E. MORIN <yann.morin.1998@free.fr>
[yann.morin.1998@free.fr: add "For more details, "]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/pixman: fix build failure due to gcc bug 101737

The pixman package exhibits gcc bug 101737 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_101737=y.
Also let's add PIXMAN_CFLAGS and pass the Codesourcery work around CFLAGS
to it for consistency like we do for the rest of the packages.

Fixes:
http://autobuild.buildroot.net/results/b20/b20869bbb48edb1f0a847ea9e2e1a0462d6350be/

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

toolchain: introduce BR2_TOOLCHAIN_HAS_GCC_BUG_101737

pixman package fails to build for the SH4 architecture with optimization
enabled with gcc up to 11.1.0:
http://autobuild.buildroot.net/results/b20/b20869bbb48edb1f0a847ea9e2e1a0462d6350be/

It's been reported upstream:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101737

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

package/pistache: fix build with gcc 11

Fix the following build failure with gcc 11:

In file included from /tmp/instance-3/output-1/build/pistache-f2f5a50fbfb5b8ef6cf1d3d2a9d442a8270e375d/src/../include/pistache/async.h:10,
                 from /tmp/instance-3/output-1/build/pistache-f2f5a50fbfb5b8ef6cf1d3d2a9d442a8270e375d/src/../include/pistache/client.h:9,
                 from /tmp/instance-3/output-1/build/pistache-f2f5a50fbfb5b8ef6cf1d3d2a9d442a8270e375d/src/client/client.cc:7:
/tmp/instance-3/output-1/build/pistache-f2f5a50fbfb5b8ef6cf1d3d2a9d442a8270e375d/src/../include/pistache/typeid.h:26:12: error: expected type-specifier before 'size_t'
   26 |   operator size_t() const { return reinterpret_cast<size_t>(id_); }
      |            ^~~~~~

Fixes:
 - http://autobuild.buildroot.org/results/2443559df8c2357476e4cbdbebb08280cbb80a3b

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

DEVELOPERS: add Giulio Benetti to libfuse3 package

Add myself as libfuse3 package maintainer.

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

package/x11r7/xlib_libxshmfence: disable on riscv32

xlib_libxshmfence unconditionally uses SYS_futex which raises the
following build failure on riscv32:

xshmfence_futex.h:58:17: error: 'SYS_futex' undeclared (first use in this function); did you mean 'sys_futex'?
   58 |  return syscall(SYS_futex, addr1, op, val1, timeout, addr2, val3);
      |                 ^~~~~~~~~
      |                 sys_futex

Fixes:
 - http://autobuild.buildroot.org/results/b3523e35fde0fac04b96a6278cbc6ffdfe56f7d1

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

package/mupdf: fix CVE-2021-37220

MuPDF through 1.18.1 has an out-of-bounds write because the cached color
converter does not properly consider the maximum key size of a hash
table. This can, for example, be seen with crafted "mutool draw" input.

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

package/mupdf: renumber second patch

Fix mistake in second patch number added by commit
d7d23710facd789c8986de055e30792af2591f2c

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

package/qpdf: fix CVE-2021-36978

QPDF 9.x through 9.1.1 and 10.x through 10.0.4 has a heap-based buffer
overflow in Pl_ASCII85Decoder::write (called from Pl_AES_PDF::flush and
Pl_AES_PDF::finish) when a certain downstream write fails.

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

package/uhd: fix build with gcc 11

Fix the following build failure with gcc 11:

In file included from /tmp/instance-4/output-1/build/uhd-3.15.0.0/host/lib/usrp/dboard/e3xx/e3xx_ad9361_iface.hpp:14,
                 from /tmp/instance-4/output-1/build/uhd-3.15.0.0/host/lib/usrp/dboard/e3xx/e3xx_ad9361_iface.cpp:7:
/tmp/instance-4/output-1/build/uhd-3.15.0.0/host/lib/include/uhdlib/utils/rpc.hpp: In function 'bool {anonymous}::await_rpc_connected_state(std::shared_ptr<rpc::client>, std::chrono::milliseconds)':
/tmp/instance-4/output-1/build/uhd-3.15.0.0/host/lib/include/uhdlib/utils/rpc.hpp:28:27: error: 'sleep_for' is not a member of 'std::this_thread'
   28 |         std::this_thread::sleep_for(std::chrono::milliseconds(1));
      |                           ^~~~~~~~~

Fixes:
 - http://autobuild.buildroot.org/results/1591ff7fabb09e4eb79f3aa293fbb228c255c45b

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

package/network-manager: needs gcc >= 4.9

network-manager needs gcc >= 4.9 since bump to version 1.32.2 in commit
360d1aad84449c6a1993d8c7e469dc0cd35339a2 to avoid the following build
failure:

src/libnm-core-impl/nm-setting-bond.c: At top level:
src/libnm-core-impl/nm-setting-bond.c:149:5: error: array initialized from non-constant array expression
     NM_MAKE_STRV("stable", "bandwidth", "count");
     ^

Fixes:
 - http://autobuild.buildroot.org/results/dc270ead82707f89d47616a0c18629715c6ea1a5

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

boot/optee-os: use dedicated 3.13.0 patch directory

Existing patch is for 3.13.0 optee version but is located at the root
of optee-os folder, leading to error when applying it on more recent
version. Move it to a dedicated 3.13.0 folder.

Signed-off-by: Clément Léger <clement.leger@bootlin.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/libebml: fix build with gcc 11

Fix the following build failure with gcc 11:

/tmp/instance-0/output-1/build/libebml-1.4.2/src/EbmlString.cpp: In member function 'virtual filepos_t libebml::EbmlString::ReadData(libebml::IOCallback&, libebml::ScopeMode)':
/tmp/instance-0/output-1/build/libebml-1.4.2/src/EbmlString.cpp:147:41: error: 'numeric_limits' is not a member of 'std'
  147 |     auto Buffer = (GetSize() + 1 < std::numeric_limits<std::size_t>::max()) ? new (std::nothrow) char[GetSize() + 1] : nullptr;
      |                                         ^~~~~~~~~~~~~~
/tmp/instance-0/output-1/build/libebml-1.4.2/src/EbmlString.cpp:147:67: error: expected primary-expression before '>' token
  147 |     auto Buffer = (GetSize() + 1 < std::numeric_limits<std::size_t>::max()) ? new (std::nothrow) char[GetSize() + 1] : nullptr;
      |                                                                   ^
/tmp/instance-0/output-1/build/libebml-1.4.2/src/EbmlString.cpp:147:70: error: '::max' has not been declared; did you mean 'std::max'?
  147 |     auto Buffer = (GetSize() + 1 < std::numeric_limits<std::size_t>::max()) ? new (std::nothrow) char[GetSize() + 1] : nullptr;
      |                                                                      ^~~
      |                                                                      std::max

Fixes:
 - http://autobuild.buildroot.org/results/3a28507127eae47994ca62e2b656da32bd2fb1f8

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

package/usbguard: needs at least a crypto backend

Since its addition in commit fbff7d7289cc95db991184f890f4ca1fcf8a101e,
usbguard needs a crypto backend otherwise the build will fail on:

checking for sodium... no
checking for libcrypto... no
checking for powerpc-buildroot-linux-uclibc-libgcrypt-config... no
checking for libgcrypt-config... no
checking for LIBGCRYPT - version >= 1.5.0... no
configure: error: The selected crypto backend library is not available.

So select sodium as the default crypto backend as this is the default
choice in configure.ac.

Fixes:
 - http://autobuild.buildroot.org/results/fb9865e2c65ea1dd9a657e1181a6c8c49481fda7

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

package/luaossl: requires DES support in openssl

Enable DES in openssl to fix build failure raised since commit
a83d41867c8d69a77d5cd0a665aa216af5340359

Fixes:
 - http://autobuild.buildroot.org/results/3e043f992f81ba1e7ad7a2c64d73425e074301a2

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

package/php: drop mcrypt

mcrypt has been dropped from php since version 7.2.0 and
https://github.com/php/php-src/commit/5a17f632d4c7d936385485a0cb187172f33eb97a

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

package/shairport-sync: fix static build with libsndfile

Fix the following static build failure with libsndfile raise since
commit dbc6e9e9f385995eb10bbfb0bb08e7674100ce66:

/tmp/instance-3/output-1/host/lib/gcc/xtensa-buildroot-linux-uclibc/10.3.0/../../../../xtensa-buildroot-linux-uclibc/bin/ld: /tmp/instance-3/output-1/host/bin/../xtensa-buildroot-linux-uclibc/sysroot/usr/lib/libsndfile.a(libsndfile_la-flac.o): in function `flac_byterate':
flac.c:(.text+0xfc): undefined reference to `FLAC__StreamDecoderErrorStatusString'

Fixes:
 - http://autobuild.buildroot.org/results/92ed30a6855ca11800b779718822bcba4a69c9a3

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

package/shairport-sync: use --without-foo options

Use --without-foo options to disable unused features, as they have been
fixed since version 3.3.8

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

package/shairport-sync: bump to version 3.3.8

Version 3.3.8 brings many bug fixes and enhancements and is recommended
for all users.

https://github.com/mikebrady/shairport-sync/releases/tag/3.3.8

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

package/libuwsc: fix build with gcc 11

Fix the following build failure with gcc 11:

/tmp/instance-1/output-1/build/libuwsc-3.3.5/src/lua/uwsc_lua.c: In function 'uwsc_lua_on':
/tmp/instance-1/output-1/build/libuwsc-3.3.5/src/lua/uwsc_lua.c:177:5: error: this 'else' clause does not guard... [-Werror=misleading-indentation]
  177 |     else
      |     ^~~~
/tmp/instance-1/output-1/build/libuwsc-3.3.5/src/lua/uwsc_lua.c:180:9: note: ...this statement, but the latter is misleadingly indented as if it were guarded by the 'else'
  180 |         return 0;
      |         ^~~~~~

Fixes:
 - http://autobuild.buildroot.org/results/1b031667e6d184493ea55ea6630272af4e475375

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

package/sox: fix static build with libsndfile

Fix the following static build failure with libsndfile raised since
commit dbc6e9e9f385995eb10bbfb0bb08e7674100ce66:

configure:14720: checking for sf_open_virtual in -lsndfile
configure:14745: /tmp/instance-1/output-1/host/bin/xtensa-buildroot-linux-uclibc-gcc -o conftest -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -mlongcalls -mauto-litpools -Os -g0  -static -Wall -Wmissing-prototypes -Wstrict-prototypes  -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64  -static -Wl,--as-needed conftest.c -lsndfile    >&5
conftest.c:73:1: warning: function declaration isn't a prototype [-Wstrict-prototypes]
   73 | char sf_open_virtual ();
      | ^~~~
conftest.c:75:1: warning: function declaration isn't a prototype [-Wstrict-prototypes]
   75 | main ()
      | ^~~~
/tmp/instance-1/output-1/host/lib/gcc/xtensa-buildroot-linux-uclibc/10.3.0/../../../../xtensa-buildroot-linux-uclibc/bin/ld: /tmp/instance-1/output-1/host/xtensa-buildroot-linux-uclibc/sysroot/usr/lib/libsndfile.a(libsndfile_la-flac.o): in function `flac_byterate':
flac.c:(.text+0xfc): undefined reference to `FLAC__StreamDecoderErrorStatusString'

Fixes:
 - http://autobuild.buildroot.org/results/4bc58ed68b29642876bb02710d0cd4f31540de86

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

package/openzwave: fix example build

-std=c++11 has been wrongly removed from CFLAGS since commit
c5ca521e232eae31276e7a12d17f0cf9d9c0108f resulting in the following
build failure:

In file included from /home/buildroot/autobuild/run/instance-1/output-1/build/openzwave-62444b0f979c337d2091d77d89cf63c2ae9775cf/cpp/src/Driver.h:38:0,
                 from /home/buildroot/autobuild/run/instance-1/output-1/build/openzwave-62444b0f979c337d2091d77d89cf63c2ae9775cf/cpp/src/Manager.h:39,
                 from Main.cpp:37:
/home/buildroot/autobuild/run/instance-1/output-1/build/openzwave-62444b0f979c337d2091d77d89cf63c2ae9775cf/cpp/src/Node.h:566:32: error: 'std::shared_ptr' has not been declared
    void SetProductDetails(std::shared_ptr<Internal::ProductDescriptor> product);
                                ^

Fixes:
 - http://autobuild.buildroot.org/results/a7999ffe3b93391b9a20dcfa3ddd9d21daf0cace

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

package/system-config-printer: add host desktop-file-utils dependency

Fixes:
 - http://autobuild.buildroot.net/results/33f/33f00aa38d55e5d3acb7171428a7ef23532afa94

Signed-off-by: James Hilliard <james.hilliard1@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/desktop-file-utils: new package

Signed-off-by: James Hilliard <james.hilliard1@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/libfuse3: fix build failure on Microblaze due to wrong symver attribute detection

Add a patch to fix symver wrong detection. As explained into the patch
pending upstream[1] __has attribute() gives false positive when an
attribute is not supported[2] and this leads to assuming symver attribute
is present but it's not. This consequentely leads to build failure. I.e.
Microblaze doesn't really support it since it doesn't include elfos.h in
config.gcc that determines symver to be available.

[1]: https://github.com/libfuse/libfuse/pull/620
[2]: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101766

Fixes:
http://autobuild.buildroot.net/results/d6c/d6cfaf2aafaeda3c12d127f6a2d2e175b25e654f

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

package/syslog-ng: fix static build

Fix static build failure raised since bump to version 3.33.1 in commit
a9e405b07976815a5b7202a5fe50c8797e754178

Fixes:
 - http://autobuild.buildroot.org/results/ccd0a954f6e0676db67026f8b8c89823f3e5c510

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

package/php: opcache needs dynamic library

opcache needs dynamic library since version 8.0.0 and
https://github.com/php/php-src/commit/9a06876072b9ccb023d4a14426ccb587f10882f3
otherwise the build will fail on:

/tmp/instance-4/output-1/host/opt/ext-toolchain/bin/../lib/gcc/arm-buildroot-linux-uclibcgnueabi/9.3.0/../../../../arm-buildroot-linux-uclibcgnueabi/bin/ld: /tmp/instance-4/output-1/host/bin/../arm-buildroot-linux-uclibcgnueabi/sysroot/usr/lib/libc.a(__uClibc_main.os): in function `__uClibc_fini':
__uClibc_main.c:(.text+0x15c): undefined reference to `__fini_array_start'
/tmp/instance-4/output-1/host/opt/ext-toolchain/bin/../lib/gcc/arm-buildroot-linux-uclibcgnueabi/9.3.0/../../../../arm-buildroot-linux-uclibcgnueabi/bin/ld: __uClibc_main.c:(.text+0x160): undefined reference to `__fini_array_end'

Fixes:
 - http://autobuild.buildroot.org/results/526d2e3108ff2506482d1665bcc3ed083961430f

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

package/cegui: disable OpenGLES renderer

Support of OpenGLES generates tons of autobuilders failures
so let's disable it till upstream finds a solution for that.

Fixes:
 http://autobuild.buildroot.net/results/41d/41d10c09472fbb8481a72c4f8bed789bb15bb3c2//
 http://autobuild.buildroot.net/results/9a8/9a8122b4d3d0aa9d71cdf8d21359411b130741f8//
 http://autobuild.buildroot.net/results/7e5/7e5dd79b48abde09bf4633b196579da5561f9fa1//

and many more...

Signed-off-by: Bartosz Bilas <b.bilas@grinn-global.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/pdbg: Fix alignment error on autobuilder

The autobuilder fails when building for sh because of an alignment issue
in assembly that includes an external data file.

Fixes:
 http://autobuild.buildroot.net/results/457b88c4ba85edb42b3a36cce48bf90712f590bf/

This fix has been submitted upstream.

Signed-off-by: Joel Stanley <joel@jms.id.au>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/azure-iot-sdk-c: fix build with gcc 11

Fix the following build failure with gcc 11:

/tmp/instance-7/output-1/build/azure-iot-sdk-c-LTS_01_2021_Ref01/c-utility/src/hmac.c:211:43: error: argument 2 of type 'uint8_t *' {aka 'unsigned char *'} declared as a pointer [-Werror=array-parameter=]
  211 | int hmacResult(HMACContext *ctx, uint8_t *digest)
      |                                  ~~~~~~~~~^~~~~~
In file included from /tmp/instance-7/output-1/build/azure-iot-sdk-c-LTS_01_2021_Ref01/c-utility/src/hmac.c:13:
/tmp/instance-7/output-1/build/azure-iot-sdk-c-LTS_01_2021_Ref01/c-utility/inc/azure_c_shared_utility/sha.h:252:42: note: previously declared as an array 'uint8_t[64]' {aka 'unsigned char[64]'}
  252 | int hmacResult(HMACContext *ctx, uint8_t digest[USHAMaxHashSize]);
      |                                  ~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~

Fixes:
 - http://autobuild.buildroot.org/results/11a6eebd0deea679683a93a5c8355d00acdfe51e

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

package/libuhttpd: fix build with wolfssl

Fix the following build failure with wolfssl raised since bump to
version 3.12.1 in commit 4d85defa7103aadc74223868fb83ed0bed144f4d:

/tmp/instance-4/output-1/build/libuhttpd-3.12.1/src/ssl/openssl.c: In function 'ssl_context_new':
/tmp/instance-4/output-1/build/libuhttpd-3.12.1/src/ssl/openssl.c:174:33: error: 'tls13_ciphersuites' undeclared (first use in this function)
  174 |     SSL_CTX_set_ciphersuites(c, tls13_ciphersuites);
      |                                 ^~~~~~~~~~~~~~~~~~

Fixes:
 - http://autobuild.buildroot.org/results/25db1e999950bbee069ba7364a228184a6a77808

Note: we have a reference to a pull request on the 'ssl' repository,
because it is a git submodule of libuhttpd.

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
[yann.morin.1998@free.fr: explain the reference to ssl/pull/1]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/libodb-mysql: fix build with gcc 11

Fixes:
http://autobuild.buildroot.net/results/8f0/8f0e3b604ff9809e3b415036b8d99521db33c4db
http://autobuild.buildroot.net/results/d03/d038b79e719c37a93d635c6b61443033094d8716

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

package/python3: add patch to fix highly parallel builds

Highly parallel host-python3 builds sometimes fail with:

Exception in thread Thread-1:
Traceback (most recent call last):
  File "/tmp/instance-3/output-1/host/lib/python3.9/threading.py", line 973, in _bootstrap_inner
    self.run()
  File "/tmp/instance-3/output-1/host/lib/python3.9/concurrent/futures/process.py", line 317, in run
    result_item, is_broken, cause = self.wait_result_broken_or_wakeup()
  File "/tmp/instance-3/output-1/host/lib/python3.9/concurrent/futures/process.py", line 376, in wait_result_broken_or_wakeup
    worker_sentinels = [p.sentinel for p in self.processes.values()]
  File "/tmp/instance-3/output-1/host/lib/python3.9/concurrent/futures/process.py", line 376, in <listcomp>
    worker_sentinels = [p.sentinel for p in self.processes.values()]
RuntimeError: dictionary changed size during iteration

During the compile_all.py step of host-python3. This issue is reported
upstream at https://bugs.python.org/issue43498, and while not yet
fixed upstream, a PR was proposed with a possible fix for it. Seems
the PR seems reasonable, let's give it a chance and see if it improves
the situation.

Hopefully Fixes:

  http://autobuild.buildroot.net/results/ae6c4ab292589a4e4442dfb0a1286349a9bf4d29/

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

package/sysdig: fix C++ comment

Replace libjson by jsoncpp for C++ dependency which was wrongly added
by commit 74fc60a267341879ca71f8eaa64a80f5b5aaf78b

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

package/zstd: observe Buildroot optimisation flags

Use Buildroot's setting for optimization, zstd's build system
overrides CFLAGS (adds -O3), but MOREFLAGS can override again.
Quick tests show that using -O2 (like buildroot)
is actually a little faster than -O3 on x86_64 Atoms.

Disable the legacy format, these are just needed for
decompressing files created with pre-release version.

Signed-off-by: Norbert Lange <nolange79@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/zstd: fix dynamic build without threads

Dynamic build without threads is broken since bump to version 1.5.0 in
commit aad8bbf58842dd62a4911d2c671ad9598e26b459

Indeed, dynamic library is built with threads support since
https://github.com/facebook/zstd/commit/91465e23b2710de031a762874cafb417f8b7556e

To fix this build failure, add -nomt if needed however libzstd-nomt is a
"special target that builds a library in single-thread mode _and_
without zstdmt_compress.c". For an unknown reason, this target fails to
build and don't create any symlinks so replace libzstd-{mt,nomt} by
lib-{mt,nomt} even if this will have the side effect of building a
shared and a static version of the library.

Move the existing HAVE_THREAD setting in the same condition.

While at it, also replace "libzstd.a libzstd" by "lib"

Fixes:
 - http://autobuild.buildroot.org/results/e609601a0fc91c44d88a12c35b29ce937381462f

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
[Arnout: move HAVE_THREAD in the same condition]
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/libodb-boost: fix build with gcc 11

Fixes:
 - No autobuilder failures yet

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

package/libodb-pqsql: fix build with gcc 11

GCC 11 defaults to C++17. Fix the following build failure with gcc 11:

configure:13698: checking for libodb
configure:13772: /tmp/instance-1/output-1/host/bin/arm-buildroot-linux-gnueabihf-g++ -c -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64  -Os -g0 -D_FORTIFY_SOURCE=2 -D_REENTRANT  -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 conftest.cpp >&5
In file included from /tmp/instance-1/output-1/host/arm-buildroot-linux-gnueabihf/sysroot/usr/include/odb/exception.hxx:15,
                 from conftest.cpp:26:
/tmp/instance-1/output-1/host/arm-buildroot-linux-gnueabihf/sysroot/usr/include/odb/details/shared-ptr/base.hxx:38:49: error: ISO C++17 does not allow dynamic exception specifications
   38 | operator new (std::size_t, odb::details::share) throw (std::bad_alloc);
      |                                                 ^~~~~

Fixes:
 - http://autobuild.buildroot.org/results/73b06d7f923e5a7294377951f212b89b6cb0946a

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

package/s390-tools: fix build with binutils 2.36

Fix the following build failure with binutils 2.36:

/tmp/instance-16/output-1/host/bin/s390x-buildroot-linux-gnu-ld: Error: unable to disambiguate: -no-pie (did you mean --no-pie ?)
make[3]: *** [Makefile:113: data.o] Error 1

Fixes:
 - http://autobuild.buildroot.org/results/90100519d696c49441669117276fb8084796544b

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

package/bluez5_utils: hid2hci needs tools

Fixes:
http://autobuild.buildroot.net/results/9f91d530c22cad053d89a7bf54c4567d3012d05e
http://autobuild.buildroot.net/results/80b90e1b6fee99ed7bf4aed87a14311a12469993
http://autobuild.buildroot.net/results/444eaab22b140b26a730c7fdd4bada71a8d6fef2

Signed-off-by: Michael Nosthoff <buildroot@heine.tech>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/wireshark: link with -latomic if needed

Fix the following build failure raised since bump to version 3.4.7 in
commit bb75c4b541fac144b53b63248e235f22ba1d25ad:

/tmp/instance-5/output-1/host/opt/ext-toolchain/bin/../lib/gcc/sparc-buildroot-linux-uclibc/9.3.0/../../../../sparc-buildroot-linux-uclibc/bin/ld: ui/qt/CMakeFiles/qtui.dir/sequence_diagram.cpp.o: undefined reference to symbol '__atomic_compare_exchange_4@@LIBATOMIC_1.0'
/tmp/instance-5/output-1/host/opt/ext-toolchain/bin/../lib/gcc/sparc-buildroot-linux-uclibc/9.3.0/../../../../sparc-buildroot-linux-uclibc/bin/ld: /tmp/instance-5/output-1/host/sparc-buildroot-linux-uclibc/sysroot/lib/libatomic.so.1: error adding symbols: DSO missing from command line

Fixes:
 - http://autobuild.buildroot.org/results/6617ee0e0046a0452a1515b89e9c704b1c125ec4

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

package/gpsd: fix workaround for gcc bug 68485

Workaround for gcc bug 68485 doesn't work anymore since bump to version
3.21 in commit 5b3e72139952e0c953c8649cca55571cec5f3836 because CXXFLAGS
is appended after CFLAGS and does not contain -O0:

/tmp/instance-7/output-1/host/bin/microblazeel-linux-gcc -o rtcm2_json.os -c --sysroot=/tmp/instance-7/output-1/host/microblazeel-buildroot-linux-uclibc/sysroot -pthread -Wall -Wcast-align -Wextra -Wimplicit-fallthrough -Wmissing-declarations -Wmissing-prototypes -Wno-missing-field-initializers -Wno-uninitialized -Wpointer-arith -Wreturn-type -Wstrict-prototypes -Wvla -O0 -Os -g0 -pthread -fPIC -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -I/tmp/instance-7/output-1/host/microblazeel-buildroot-linux-uclibc/sysroot/usr/include/dbus-1.0 -I/tmp/instance-7/output-1/host/microblazeel-buildroot-linux-uclibc/sysroot/usr/lib/dbus-1.0/include rtcm2_json.c
during RTL pass: reload
rtcm2_json.c: In function ‘json_rtcm2_read’:
rtcm2_json.c:267:1: internal compiler error: in gen_reg_rtx, at emit-rtl.c:1155
  267 | }
      | ^

It should be noted that GPSD_CXXFLAGS was added 5 years ago with commit
b4c050e9afed7a8dc302e6e1014cc4ffb508590d but was not actually used in
that commit or any later.

Fixes:
 - http://autobuild.buildroot.org/results/fab33f25b08fa218af91640fdbd8c0dcf1d07228

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

package/pipewire: add libdrm dependency

The libcamera option requires libdrm, require and depend on it
when available for libcamera plugin.

Fixes:
 - http://autobuild.buildroot.net/results/e3bc1d9aa69af1d2cdeee7709df318fccbaa5476

Signed-off-by: James Hilliard <james.hilliard1@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/openzwave: set back -fPIC

Set back -fPIC in CFLAGS which was wrongly removed as a side effect of
commit c5ca521e232eae31276e7a12d17f0cf9d9c0108f.

Another side-effect of that commit was to remove
-DSYSCONFDIR="\"$(PREFIX)/etc/openzwave/\"" - however, we don't need
SYSCONFDIR to be set, since /etc/openzwave is checked first anyway (see
cpp/src/Options.cpp).

Fixes:
 - http://autobuild.buildroot.org/results/161f53574697016ab8bdb91aaaf4a2be5410ebb0

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

package/erlang: risc-v should only add risc-v

This restricts the Erlang RISC-V update from
0bab8c1089983a4b5f0f2acfe5cad77b02d6f654 to only enabling RISCV_64
rather than enabling builds on all platforms with toolchains that
include atomics.

This fixes build errors on SPARCv8 and other platforms that look like
this:

/tmp/ccD9Q9F7.s: Assembler messages:
/tmp/ccD9Q9F7.s:230: Error: Architecture mismatch on "cas [%g3],%g2,%g1".
/tmp/ccD9Q9F7.s:230: (Requires v9|v9a|v9b|v9c|v9d|v9e|v9v|v9m|m8; requested architecture is v8.)

Fixes:
http://autobuild.buildroot.net/results/a57/a5779e22c41eb63b8400409f26f96b924fdde1bc/

Signed-off-by: Frank Hunleth <fhunleth@troodon-software.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

Update for 2021.08-rc1

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

package/armadillo: allow to select between lapack or openblas

armadillo can use lapack or openblas as BLAS provider. LAPACK support is
optional.

This patch
- adds an _ARCH_SUPPORTS variable to check if one is available
- adds an option to choose lapack or openblas as BLAS provider

The choice is required since applications may potentially need lapack.

Signed-off-by: Gwenhael Goavec-Merou <gwenhael.goavec-merou@trabucayre.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

CHANGES: let's start by releasing -rc1, not -rc8

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

package/nfs-utils: Backport yet another printf fix

Previous fixes from daa5459b6a caused regression that even more builds
were broken. Backporting upstream fix 383d787d ("nfsdcltrack: Use
uint64_t instead of time_t") which fixes it.
It fixes bug:
nfsdcltrack.c: In function ‘cltrack_gracedone’:
nfsdcltrack.c:529:18: error: format ‘%llu’ expects argument of type ‘long long unsigned int’, but argument 4 has type ‘time_t’ {aka ‘long int’} [-Werror=format=]
  529 |  xlog(D_GENERAL, "%s: grace done. gracetime=%"PRIu64, __func__, gracetime);
      |                  ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~                  ~~~~~~~~~
      |                                                                 |
      |                                                                 time_t {aka long int}

Fixes:
- http://autobuild.buildroot.net/results/550ad9c74666775858887050ea83a8618797de79
- http://autobuild.buildroot.net/results/caf0a69121d49504db63910ce1df347802065cd9
- http://autobuild.buildroot.net/results/294d11f8c4330bcfd9412f47f21df22d4f2b6e82
- http://autobuild.buildroot.net/results/1cb6b19cbbc2b829a038a9a70b6734c084e81c93
- http://autobuild.buildroot.net/results/c0eb120f5d14798b4f88b90eaccd3bc1f657c00d
- http://autobuild.buildroot.net/results/a028299b612a0ca6d1a69757749a8b9e37094954
- http://autobuild.buildroot.net/results/996bb74445cfbde0e230b2185207d91d4ec46dfb
- http://autobuild.buildroot.net/results/624b6dc0793d0263f46831ed0f29fa265ff0e41f
- http://autobuild.buildroot.net/results/5eb2dd5d6bc0540443a52f0b9b9568af01892a23
- http://autobuild.buildroot.net/results/65bad834738d5c5b1176e8847777b3cfd95c4351
- http://autobuild.buildroot.net/results/3146237c356e5649163784fb6628fbbbc7c61153
- http://autobuild.buildroot.net/results/2c5e77600cbf5009e69f057e7a2dc164fc8b0466
- http://autobuild.buildroot.net/results/ec8ec029dc0d7d7d294f8ade81ee53ca2fc3c54c
- http://autobuild.buildroot.net/results/60ea3bdeddf71270468d6177ccf95a1dc6b68dac
- http://autobuild.buildroot.net/results/ff93a63cd084043db5f61ac5ea2bbbca4f0e6859
- http://autobuild.buildroot.net/results/cac3392373d05d78c299d07d8080ae4938773401
- http://autobuild.buildroot.net/results/438155f831d8a44ec6e0f7020e8783b48d2c8df2
- http://autobuild.buildroot.net/results/66bc1ca1a05e1676c4e65ef7364ba1415bdbc723
- http://autobuild.buildroot.net/results/5bacb3c38dc1012f9445a0c23e59336dcac5abce

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

package/rtl_433: force Release build

Force Release build to avoid the following build failure which is raised
since the addition of the package in commit
4634e84978a3e64e9050021d209a41cd88a1ad16:

/tmp/instance-3/output-1/host/opt/ext-toolchain/bin/../lib/gcc/i586-buildroot-linux-musl/9.3.0/../../../../i586-buildroot-linux-musl/bin/ld: cannot find -lasan
/tmp/instance-3/output-1/host/opt/ext-toolchain/bin/../lib/gcc/i586-buildroot-linux-musl/9.3.0/../../../../i586-buildroot-linux-musl/bin/ld: cannot find -lubsan

Fixes:
 - http://autobuild.buildroot.org/results/83274082879f9599bc106e12a253f4b9229ea00b

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

CHANGES: prepare for 2021.08-rc1

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

package/open62541: disable -Werror

Disable -Werror to avoid the following build failures raised since bump
to version 1.2.2 in commit b436b82411305839abc76bb438fe92661a067526:

/tmp/instance-0/output-1/build/open62541-v1.2.2/arch/network_tcp.c: In function 'connection_recv':
/tmp/instance-0/output-1/build/open62541-v1.2.2/arch/network_tcp.c:96:5: error: conversion to 'unsigned int' from 'int' may change the sign of the result [-Werror=sign-conversion]
   96 |     UA_fd_set(connection->sockfd, &fdset);
      |     ^~~~~~~~~

/tmp/instance-6/output-1/build/open62541-v1.2.2/plugins/ua_pubsub_udp.c: In function 'UA_PubSubChannelUDPMC_receive':
/tmp/instance-6/output-1/build/open62541-v1.2.2/plugins/ua_pubsub_udp.c:477:21: error: conversion to '__suseconds_t' {aka 'int'} from 'UA_UInt32' {aka 'unsigned int'} may change the sign of the result [-Werror=sign-conversion]
  477 |     tmptv.tv_usec = (long int)(timeout % 1000000);
      |                     ^

Fixes:
 - http://autobuild.buildroot.org/results/911811de81d8abb2a31feb8f27af1592641c6fbc
 - http://autobuild.buildroot.org/results/f0187b3f2d62e955fddeef4e90f84ba4fd642bd2

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

package/refpolicy: fix build with ftp

Fix build failure raised since commit
61a47bb1037ac20772f4e0106aab956adffeee12

Fixes:
 - http://autobuild.buildroot.org/results/bde34d81a04b5ea96722dc56249089dce7e65b74

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

package/fluxbox: fix build with gcc 11

Fix the following build failure with gcc 11:

util/fluxbox-remote.cc: In function 'int main(int, char**)':
util/fluxbox-remote.cc:76:32: error: ordered comparison of pointer with integer zero ('unsigned char*' and 'int')
   76 |             && text_prop.value > 0
      |                ~~~~~~~~~~~~~~~~^~~

Fixes:
 - http://autobuild.buildroot.org/results/6e718bb662aa4961ea264eb81e4b7e7b20485e30

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

package/ripgrep: add workaround for cargo bug #3349

To fix this bug we have to use an undocumented test variable to enable
the nightly target-applies-to-host feature so that the target linker
doesn't get used for host binaries like the build-script.

Fixes:
error: failed to run custom build command for `ripgrep v0.8.1 (/home/buildroot/buildroot/output/build/ripgrep-0.8.1)`

Caused by:
  process didn't exit successfully: `/home/buildroot/buildroot/output/build/ripgrep-0.8.1/target/release/build/ripgrep-59eeb7069534e1ef/build-script-build` (exit status: 1)
  --- stderr
  /home/buildroot/buildroot/output/build/ripgrep-0.8.1/target/release/build/ripgrep-59eeb7069534e1ef/build-script-build: /lib/x86_64-linux-gnu/libc.so.6: version `GLIBC_2.33' not found (required by /home/buildroot/buildroot/output/build/ripgrep-0.8.1/target/release/build/ripgrep-59eeb7069534e1ef/build-script-build)

Details:
https://github.com/rust-lang/cargo/issues/3349
https://github.com/rust-lang/cargo/pull/9322

Signed-off-by: James Hilliard <james.hilliard1@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/{rust, rust-bin}: bump to version 1.54.0

Signed-off-by: James Hilliard <james.hilliard1@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/tpm2-tools: add SELinux module

Support for tpm2-tools is added by the services/tpm2 module in the
SELinux refpolicy.

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

package/tpm2-totp: add SELinux module

Support for tpm2-totp is added by the services/tpm2 module in the
SELinux refpolicy.

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

package/tpm2-abrmd: add SELinux module

Support for tpm2-abrmd is added by the services/tpm2 module in the
SELinux refpolicy.

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

package/msgpack: bump to version 3.3.0

https://github.com/msgpack/msgpack-c/releases/tag/cpp-3.3.0

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

package/libjson: fix build with gcc 11

GCC 11 defaults to C++17. Fix the following build failure with gcc 11:

In file included from _internal/Source/JSONDefs.h:12,
                 from _internal/Source/JSONDebug.h:4,
                 from _internal/Source/JSONNode.h:4,
                 from _internal/Source/JSONNode.cpp:1:
_internal/Source/JSONDefs/GNU_C.h:58:28: error: ISO C++17 does not allow dynamic exception specifications
   58 |     #define json_throws(x) throw(x)
      |                            ^~~~~

Fixes:
 - http://autobuild.buildroot.org/results/1e66dff705bbb38e7e0f0e5864ce794b4345dcc6

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

package/opencv4: do not detect ccache

OpenCV-4's buildsystem will try to detect ccache and use it if
available. This may yield a system-installed ccache.

However, in Buildroot, ccache is entirely hidden away and handled in the
toolchain wrapper.

Forcibly disable detection of ccache.

Commit 505e7f47715c77d8586444a4fefe4a190b77aaad already applied this
change to opencv3.

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

package/libmodsecurity: fix build with libmaxminddb

Build with libmaxminddb is broken since bump to version 3.0.5 in commit
464d0be380c84ac7c3f1684e49153c3868280d7e because of
https://github.com/SpiderLabs/ModSecurity/commit/785958f9b5089b918c7d054cbcc2fe4a3c7b3788

So revert this commit until upstream answer to comment to
https://github.com/SpiderLabs/ModSecurity/issues/2131

Reverting this commit requires autoreconfiguring, which itself causes
lots of warnings as configure.ac queries git to know the version of
various parts of libmodsecurity. However, it turns out that those
versions are only used to be displayed in the output of the configure
script, which is quite useless. The only one that is referenced
elsewhere is LIBINJECTION_VERSION, but it's in fact a different thing:
it is defined by others/libinjection/src/libinjection_sqli.c.

The only variable that was AC_SUBST() and therefore visible elsewhere
was MSC_GIT_VERSION, but it is not used anywhere in the code base,
except in the configure script itself.

Note that one patch is 0001 and the other 0003, because there was
already a 0002 patch.

Fixes:
 - http://autobuild.buildroot.org/results/4c639fd967faa06f8ae362bacd38f3409c47267c

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

package/optee-client: add dependency on headers >= 4.3

Since upstream commit 5da4da99e0bb3d694e93046207ae4ce3b5545043, in
3.13.0, OP-TEE client unconditionally uses the MMC_IOC_MULTI_CMD
ioctl(). But this ioctl() was only introduced in kernel 4.3, so
earlier kernels did not have this definition.

This was clearly known when upstream started using this ioctl. From
the upstream commit 5da4da99e0bb3d694e93046207ae4ce3b5545043 commit
log:

    As MMC_IOC_MULTI_CMD is available since linux 4.3, we should be safe to
    use it by default on read / write block operations.

So let's propagate this dependency into Buildroot.

Fixes:

  http://autobuild.buildroot.net/results/bd270a6793df499d5f52959e3de17f8a92d88305/

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

package/optee-test: fix Config.in comment on dependencies

The comment did not take into account the dynamic library dependency,
fix that.

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

package/optee-client: add dependency on !BR2_STATIC_LIBS

Since OP-TEE client 3.13.0, <dlfcn.h> is unconditionally used, so we
need to add a !BR2_STATIC_LIBS dependency to optee-client and its
reverse dependencies.

Fixes:

  http://autobuild.buildroot.net/results/169cf630866cb84d711d68877bb83c897e1d2814/

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

package/irqbalance: bump to version 1.8.0

- Drop first patch (already in version)
- Update indentation in hash file (two spaces)

https://github.com/Irqbalance/irqbalance/releases/tag/v1.8.0

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

package/irqbalance: use pkg-config to find numa

Use pkg-config to find numa to avoid the following build failure when
checking for numa_available:

configure:9667: checking for numa_available in -lnuma
configure:9692: /tmp/instance-7/output-1/host/bin/microblazeel-linux-gcc -o conftest -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64  -Os   -static -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64  -static conftest.c -lnuma   >&5
/tmp/instance-7/output-1/host/opt/ext-toolchain/bin/../lib/gcc/microblazeel-buildroot-linux-uclibc/9.3.0/../../../../microblazeel-buildroot-linux-uclibc/bin/ld: /tmp/instance-7/output-1/host/microblazeel-buildroot-linux-uclibc/sysroot/usr/lib/libnuma.a(libnuma.o): in function `numa_node_to_cpus_v1':
(.text+0x2a80): undefined reference to `__atomic_fetch_and_1'
/tmp/instance-7/output-1/host/opt/ext-toolchain/bin/../lib/gcc/microblazeel-buildroot-linux-uclibc/9.3.0/../../../../microblazeel-buildroot-linux-uclibc/bin/ld: /tmp/instance-7/output-1/host/microblazeel-buildroot-linux-uclibc/sysroot/usr/lib/libnuma.a(libnuma.o): in function `numa_node_to_cpus_v2':
(.text+0x2ddc): undefined reference to `__atomic_fetch_and_1'
collect2: error: ld returned 1 exit status

Fixes:
 - http://autobuild.buildroot.org/results/577a63432fba2f9ae1ed2c6c2a77c5ce54ac5521

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

package/libodb: fix build with gcc 11

GCC 11 defaults to C++17. Fix the following build failure with gcc 11:

In file included from details/shared-ptr/base.cxx:5:
../odb/details/shared-ptr/base.hxx:38:49: error: ISO C++17 does not allow dynamic exception specifications
   38 | operator new (std::size_t, odb::details::share) throw (std::bad_alloc);
      |                                                 ^~~~~

Fixes:
 - http://autobuild.buildroot.org/results/cfd5f92f0aa923815edba5fbfcd5b7b312d9d40e

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

package/kodi: enable BlueZ Audio and HID plugins

When Bluetooth support is enabled ensure to enable the Audio and
HID plugins.

Signed-off-by: Michael Nosthoff <buildroot@heine.tech>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/supertuxkart: enable BlueZ HID Plugin

When BlueZ is enabled ensure to select the BlueZ HID Plugin for wiimote support.

Signed-off-by: Michael Nosthoff <buildroot@heine.tech>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/pipewire: enable BlueZ Audio Plugins

When BlueZ is enabled ensure to select the BlueZ Audio Plugins.

Signed-off-by: Michael Nosthoff <buildroot@heine.tech>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/pulseaudio: enable BlueZ Audio Plugins

When BlueZ is enabled ensure to select the BlueZ Audio Plugins.

Signed-off-by: Michael Nosthoff <buildroot@heine.tech>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/gstreamer1/gst1-plugins-bad: enable BlueZ Audio

When enabling the BlueZ plugin select the BlueZ Audio Plugins.

Signed-off-by: Michael Nosthoff <buildroot@heine.tech>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/brltty: enable bluez HID plugin

brltty builds with bluetooth support when bluez is enabled.
Ensure the HID Plugin is enabled.

Signed-off-by: Michael Nosthoff <buildroot@heine.tech>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/bluez-alsa: select BlueZ audio plugins

The A2DP/AVCRP plugins now have to be explicitly selected.

Signed-off-by: Michael Nosthoff <buildroot@heine.tech>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/cwiid: select BlueZ HID plugin

the HID plugin now needs to be explictly enabled.

Signed-off-by: Michael Nosthoff <buildroot@heine.tech>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/bluez5_utils: drop default y

as discussed on the mailinglist this drops the default y
entry for the now optional plugins and tools.
This might break backward compatibility but gives new users
a smaller initial footprint.

Further add a select for the options that required the HID
plugin at runtime.

Signed-off-by: Michael Nosthoff <buildroot@heine.tech>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/bluez5_utils: make tools optional

The tools are currently always build and installed. In many cases
those tools are not needed in a production system. So make this
step optional. This saves up to 2MB on the target.

Further tweak the deprecated option. It has effects when the --enable-tools
is set as well as when --enable-client is set. So explain this in the
options description and update the install step for gatttool to only be
added when client is enabled.

Signed-off-by: Michael Nosthoff <buildroot@heine.tech>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/bird: needs autoreconf

Since the bump to version 2.0.8 in commit
f5906644b45269b1a5fe109cb64fad837e4fbff2, bird uses a home-grown "GNU Autoconf
2.69e" instead of standard autoconf 2.69 to generate the configure
script. This fork seems to generate incorrect code: it uses

ac_test_CFLAGS=${CFLAGS+y}

to remember if CFLAGS was saved, while standard autoconf uses

ac_test_CFLAGS=${CFLAGS+set}

The configure.ac code itself, however, uses

if test "$ac_test_CFLAGS" != set ; then

to check if CFLAGS was overridden.

Thus, Buildroot's CFLAGS are not taken into account and this leads to
the following build failure:

checking for glob.h... no
configure: error: glob.h not found.

This build failure is raised because -g0 and -g are both passed to gcc:

configure:5207: checking for glob.h
configure:5230: /data/buildroot-autobuilder/instance-0/output-1/host/bin/mips64el-linux-gcc -o conftest -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64  -Os -g0  -pthread -fno-strict-aliasing -fno-strict-overflow -flto -Wall -Wextra -Wstrict-prototypes -Wno-parentheses -Wno-pointer-sign -Wno-missing-field-initializers -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64  -pthread -flto=4 -g conftest.c  >&5
/data/buildroot-autobuilder/instance-0/output-1/host/opt/ext-toolchain/bin/../lib/gcc/mips64el-buildroot-linux-uclibc/8.4.0/../../../../mips64el-buildroot-linux-uclibc/bin/ld: /tmp/ccDboxph.ltrans0.ltrans.o:(.debug_info+0x2a): undefined reference to `conftest.c.0943dc99'
/data/buildroot-autobuilder/instance-0/output-1/host/opt/ext-toolchain/bin/../lib/gcc/mips64el-buildroot-linux-uclibc/8.4.0/../../../../mips64el-buildroot-linux-uclibc/bin/ld: /tmp/ccDboxph.ltrans0.ltrans.o:(.debug_info+0x2f): undefined reference to `conftest.c.0943dc99'

-g0 is passed by buildroot and -g is passed by bird if
$ac_test_CFLAGS" != set (since version 2.0.7 and
https://gitlab.nic.cz/labs/bird/-/commit/cc95b4594ac924b40325a4f1adcae5312179db40)

To fix this, we can simply run autoreconf and generate a correct
configure script.

Fixes:
 - http://autobuild.buildroot.org/results/1d2acc9f6b8830adc8b62d6b2e55837abae561a9

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

package/binutils: remove version 2.34

Now that we have 2.35, 2.36 and 2.37, with 2.36 as the default, we can
remove 2.34.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/binutils: default to version 2.36.x

Now that we have added version 2.37, it's time to use the 2.36.x
series as the default version.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
[Arnout: also update default in binutils.mk]
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/binutils: add version 2.37

State of the patches:

- 0001-sh-conf.patch
  Refreshed

- 0002-poison-system-directories.patch
  Refreshed, but needed some adaptations as the bfd_boolean type no
  longer exists, and the standard "bool" type is now used instead.

- 0003-or1k-Fix-issue-with-plt-link-failure-for-local-calls.patch
  Drop, present in 2.37, merged upstream as
  a76ef689b60405e494cb99e198acf3c82f467f7d

- 0004-or1k-Implement-relocation-R_OR1K_GOT_AHI16-for-gotha.patch
  Drop, present in 2.37, merged upstream as
  0b3e14c90283c5d234884d0ebe8510bc3c9bc687

- 0005-or1k-Avoid-R_OR1K_GOT16-overflow-failures-in-presenc.patch
  Drop, present in 2.37, merged upstream as
  3c3de29b048bca6b4aa4235c647b9328e71801b6

- 0006-or1k-Support-large-plt_relocs-when-generating-plt-en.patch
  Drop, present in 2.37, merged upstream as
  284a1309021a0ef4c29f198470d95652f02b13f0

- 0007-bfd-elf32-or1k-fix-building-with-gcc-version-5.patch
  Refreshed

- 0008-or1k-fix-pc-relative-relocation-against-dynamic-on-P.patch
  Refreshed

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/mesa3d: DRI drivers need either glx, egl or gbm

Fixes build error:
../output-1/build/mesa3d-21.1.5/meson.build:520:4:
 ERROR: Problem encountered: building dri drivers require at least one windowing system

OSMesa is not seen as a valid windowing system for DRI drivers anymore:
https://cgit.freedesktop.org/mesa/mesa/commit/meson.build?h=21.1&id=ee802372180a2b4460cc7abb53438e45c6b6f1e4

Commit 15a2f9b819806d38a7d8172a20f80130b1d60e63 blindly renamed
BR2_PACKAGE_MESA3D_OSMESA_CLASSIC to BR2_PACKAGE_MESA3D_OSMESA_GALLIUM
without checking for this particular case.

Fixes:
http://autobuild.buildroot.net/results/46e3a66ddc13a064e512aad0f3afd8655889ab9d/

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