package/libmaxminddb: bump to version 1.4.3

Update indentation in hash file (two spaces)

https://github.com/maxmind/libmaxminddb/releases/tag/1.4.3

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

configs/olimex_a20_olinuxino_lime*: bump Linux and U-Boot versions

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

package/whois: bump to version 5.5.7

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

package/minissdpd: bump to version 1.5.20200928

Update indentation in hash file (two spaces)

http://miniupnp.free.fr/files/changelog.php?file=minissdpd-1.5.20200928.tar.gz

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

package/jasper: bump to version 2.0.22

Signed-off-by: Michael Vetter <jubalh@iodoru.org>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/btrfs-progs: bump to version 5.7

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

support/scripts/generate-gitlab-ci-yml: rework generation of pipelines

Currently, we handle three kinds of tests: basic, defconfig, and
runtime, and we treat them totally independently ones from the others.

Except for the basic tests that are ignored when defconfig or runtime
tests are explicitly requested.

The basic tests are also run systematically on all our reference
branches: master, next (when it exists), and the maintenance branches:
YYYY.MM.x.

Furthermore, we can see that the conditions to run each set of tests
are very similar, with only the explicit queries differing by name.

Rework the script so that the conditions are expressed only once, and
each set of tests is decided for each condition. This makes it easier
to decide what tests should run under what conditions.

Using GitLab-CI's schedules, with a variable expressing the actual test
to run, would seem the obvious choice to trigger the pipelines. However,
a schedule is configured for a specific branch, which means we would
need one schedule per branch we want to build per test cases we want to
run, *and* that we update those schedules when we add/remove branches
(e.g. when we open/close 'next', or a maintenance branch). This is not
very nice, as it requires some manual tweaking and twiddling on the web
UI.

Instead, we resort to using triggers, that will be triggered from a
cronjob on some server. Using a cronjiob allows us to more easily manage
the branches we want to test and test cases we want to run, to more
easily spread the load over the week, etc...

Note: triggering a pipeline can be done with a simple curl invocation:

    $ curl -X POST \
        -F "token=${YOUR_TOKEN}" \
        -F "ref=${BRANCH_TO_TEST}" \
        -F "variables[BR_SCHEDULE_JOBS]=${TEST_TO_RUN}" \
        "https://gitlab.com/api/v4/projects/${YOUR_PROJECT_ID}/trigger/pipeline"

Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Cc: Romain Naour <romain.naour@gmail.com>
Cc: Arnout Vandecappelle <arnout@mind.be>
Cc: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

DEVELOPERS: add Antoine Ténart for libselinux and refpolicy

Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

support/testing/tests/core/test_selinux: new tests for the packages SELinux functionalities

Add tests to ensure the packages SELinux functionalities (being able to
select an extra SELinux module in the refpolicy, and being able to
provide a custom SELinux module) are working as expected.

We use a BR2_EXTERNAL folder, provided in the tests, to use a custom
SELinux enabled package.

Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

support/testing/tests/core/test_selinux: new test for BR2_PACKAGE_REFPOLICY_CUSTOM_GIT

Add a test for BR2_PACKAGE_REFPOLICY_CUSTOM_GIT (which allows to select
a custom location for the SELinux refpolicy). The test uses the official
refpolicy as a test (we only want to test the functionality is working,
not that another refpolicy is correctly building; that is an user
problematic).

Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

support/testing/tests/core/test_selinux: new test for BR2_REFPOLICY_EXTRA_MODULES_DIRS

Add a test for the BR2_REFPOLICY_EXTRA_MODULES_DIRS functionality (which
allows to provide custom SELinux modules).

Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

support/testing/tests/core/test_selinux.py: new test for BR2_REFPOLICY_EXTRA_MODULES

This patch adds a test for the BR2_REFPOLICY_EXTRA_MODULES
functionality (which allows to select extra modules within the SELinux
refpolicy using Kconfig).

Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

support/testing/tests/init/test_systemd_selinux: new SELinuxSystemdSquashfs test

Add a test called 'SELinuxSystemdSquashfs' which will perform the same
tests as the Ext4 version, but using a Squashfs filesystem. Thanks to
this, we'll have a test on a real only filesystem.

Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

support/testing/tests/init/test_systemd_selinux: new SELinuxSystemdExt4 test

This adds a test called 'SELinuxSystemdExt4'. This test will build an
SELinux enabled image with systemd, boot it, and perform a few runtime
tests to check SELinux related capabilities.

Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/refpolicy: test REFPOLICY_EXTRA_MODULES_DIR differently

REFPOLICY_EXTRA_MODULES_DIRS contains
$(PACKAGES_SELINUX_EXTRA_MODULES_DIRS) which is filled in by
package/pkg-generic.mk with the list of packages that have a selinux/
sub-directory. Due to how variable expansion works, if there is an
ifeq/ifneq test of REFPOLICY_EXTRA_MODULES_DIRS, it will only see the
value of REFPOLICY_EXTRA_MODULES_DIRS with the list of packages
*before* refpolicy in alphabetic ordering. This means that packages
after refpolicy in alphabetic ordering would not be taken into
account.

To fix this, we switch to an $(if ...) test, which allows the variable
to really be evaluated during the refpolicy build. This makes sures
the expansion is correct.

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

package/sentry-cli: pass HOST_LDFLAGS

Pass HOST_LDFLAGS to avoid the following build failure due to missing
RPATH:

*** ERROR: package host-sentry-cli installs executables without proper RPATH:
***   /srv/storage/autobuild/run/instance-3/output-1/host/bin/sentry-cli

Fixes:
 - http://autobuild.buildroot.org/results/d52638baefec961ff693aecf833af7adfc482fe4

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

package/mraa: drop C++ dependency

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

package/mraa: renumber patch

Commit fd5376a39d079cc5aa26d67788a0c32d234f2f96 wrongly named the patch
as 0002-x.patch instead of 0001-x.patch

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

boot/barebox: fix target bareboxenv command compile

The buildroot custom bareboxenv compile command misses the additional
include path 'scripts/include' to gain access to the local copy of the
kernel header files (which leads to compile error when using an older
toolchain).

This could be fixed by enhancing the custom bareboxenv compile command
(see [1]) or by using the barebox build system by simply enabling the
CONFIG_BAREBOXENV_TARGET option (available since April 2012, see [2])
instead (as suggested by Yann E. MORIN).

Fixes (with BR2_TARGET_BAREBOX_BAREBOXENV enabled):

  build/barebox-2019.12.0/scripts/bareboxenv.c:100:10: fatal error: linux/list.h: No such file or directory

[1] http://lists.busybox.net/pipermail/buildroot/2020-January/270942.html
[2] https://git.pengutronix.de/cgit/barebox/commit/?id=afb03d7a554a2911a3742e316f011319fcb416f1

Note: a user who would previously provide a barebox config file which
had CONFIG_BAREBOXENV_TARGET=y, but a Buildroot config file which did
not have BR2_TARGET_BAREBOX_BAREBOXENV=y, would have bareboxenv-target
built, but it would not be installed in the target. Now, and unset
BR2_TARGET_BAREBOX_BAREBOXENV will not even build it, but his is not a
regression: it was anyway previously not installed.

Reported-by: Frederick Gotham <cauldwell.thomas@gmail.com>
Signed-off-by: Peter Seiderer <ps.report@gmx.net>
[yann.morin.1998@free.fr:
  - also explicitly disable it when not selected
  - rewrap commit log
]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/ruby: update to 2.7.1

- add patch to recognize coroutines correctly
- remove unneeded verconf.h removal hack

License files reformatted / converted to markdown.  Update the LEGAL /
COPYING hashes to match.

Signed-off-by: Waldemar Brodkorb <wbx@openadk.org>
[Peter: fix license hashes]
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/hiawatha: bump to version 10.11

- Default value of MinTLSversion set to 1.2.
- mbed TLS updated to 2.23.0.
- Small bugfixes.

- Update indentation in hash file (two spaces)

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

package/links: fix hash

Hash was not updated by commit d133d33586c9a06d837d5db42a6f9046156034c6

Fixes:
 - http://autobuild.buildroot.org/results/c8333bb3c1abda67dadda203beb9b69fb0e1a080

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

package/xscreensaver: bump to version 5.44

- Update indentation in hash file (two spaces)

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

package/{mesa3d, mesa3d-headers}: bump version to 20.1.9

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/vlc: fix build with live555 and without openssl

Pass -DNO_OPENSSL to avoid a build failure with live555 but without
openssl

Fixes:
 - http://autobuild.buildroot.org/results/70ca93aa5c9488a4657c7bcafa40bfb2e974a5b3

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

package/live555: fix static build with openssl

Use pkg-config to retrieve openssl static dependencies such as -lz or
-latomic.

Moreover, don't pass the openssl libraries to LIBS_FOR_LIVEMEDIA_LIB
when linking statically with ar or the build will break on:

/home/fabrice/buildroot/output/per-package/live555/host/bin/arm-linux-ar cr libliveMedia.a  \
Media.o MediaSource.o FramedSource.o FramedFileSource.o FramedFilter.o ByteStreamFileSource.o ByteStreamMultiFileSource.o ByteStreamMemoryBufferSource.o BasicUDPSource.o DeviceSource.o AudioInputDevice.o WAVAudioFileSource.o MPEG1or2Demux.o MPEG1or2DemuxedElementaryStream.o MPEGVideoStreamFramer.o MPEG1or2VideoStreamFramer.o MPEG1or2VideoStreamDiscreteFramer.o MPEG4VideoStreamFramer.o MPEG4VideoStreamDiscreteFramer.o H264or5VideoStreamFramer.o H264or5VideoStreamDiscreteFramer.o H264VideoStreamFramer.o H264VideoStreamDiscreteFramer.o H265VideoStreamFramer.o H265VideoStreamDiscreteFramer.o MPEGVideoStreamParser.o MPEG1or2AudioStreamFramer.o MPEG1or2AudioRTPSource.o MPEG4LATMAudioRTPSource.o MPEG4ESVideoRTPSource.o MPEG4GenericRTPSource.o MP3FileSource.o MP3Transcoder.o MP3ADU.o MP3ADUdescriptor.o MP3ADUinterleaving.o MP3ADUTranscoder.o MP3StreamState.o MP3Internals.o MP3InternalsHuffman.o MP3InternalsHuffmanTable.o MP3ADURTPSource.o MPEG1or2VideoRTPSource.o MPEG2TransportStrea
 mMultiplexor.o MPEG2TransportStreamFromPESSource.o MPEG2TransportStreamFromESSource.o MPEG2TransportStreamFramer.o MPEG2TransportStreamAccumulator.o ADTSAudioFileSource.o ADTSAudioStreamDiscreteFramer.o JPEGVideoSource.o JPEGVideoRTPSource.o JPEG2000VideoRTPSource.o H263plusVideoRTPSource.o H263plusVideoStreamFramer.o H263plusVideoStreamParser.o AC3AudioStreamFramer.o AC3AudioRTPSource.o DVVideoStreamFramer.o DVVideoRTPSource.o AMRAudioSource.o AMRAudioFileSource.o InputFile.o StreamReplicator.o MediaSink.o FileSink.o BasicUDPSink.o AMRAudioFileSink.o H264or5VideoFileSink.o H264VideoFileSink.o H265VideoFileSink.o OggFileSink.o MPEG1or2AudioRTPSink.o MP3ADURTPSink.o MPEG1or2VideoRTPSink.o MPEG4LATMAudioRTPSink.o MPEG4GenericRTPSink.o MPEG4ESVideoRTPSink.o JPEGVideoRTPSink.o JPEG2000VideoRTPSink.o H263plusVideoRTPSink.o H264or5VideoRTPSink.o H264VideoRTPSink.o H265VideoRTPSink.o DVVideoRTPSink.o AC3AudioRTPSink.o VorbisAudioRTPSink.o TheoraVideoRTPSink.o VP8VideoRTPSink.o VP9V
 ideoRTPSink.o GSMAudioRTPSink.o SimpleRTPSink.o AMRAudioRTPSink.o T140TextRTPSink.o OutputFile.o RawVideoRTPSink.o uLawAudioFilter.o RTPSource.o MultiFramedRTPSource.o SimpleRTPSource.o H261VideoRTPSource.o H264VideoRTPSource.o H265VideoRTPSource.o QCELPAudioRTPSource.o AMRAudioRTPSource.o VorbisAudioRTPSource.o TheoraVideoRTPSource.o VP8VideoRTPSource.o VP9VideoRTPSource.o RawVideoRTPSource.o RTPSink.o MultiFramedRTPSink.o AudioRTPSink.o VideoRTPSink.o TextRTPSink.o RTPInterface.o RTCP.o rtcp_from_spec.o GenericMediaServer.o RTSPServer.o RTSPServerRegister.o RTSPClient.o RTSPCommon.o RTSPRegisterSender.o SIPClient.o MediaSession.o ServerMediaSession.o PassiveServerMediaSubsession.o OnDemandServerMediaSubsession.o FileServerMediaSubsession.o MPEG4VideoFileServerMediaSubsession.o H264VideoFileServerMediaSubsession.o H265VideoFileServerMediaSubsession.o H263plusVideoFileServerMediaSubsession.o WAVAudioFileServerMediaSubsession.o AMRAudioFileServerMediaSubsession.o MP3AudioFile
 ServerMediaSubsession.o MPEG1or2VideoFileServerMediaSubsession.o MPEG1or2FileServerDemux.o MPEG1or2DemuxedServerMediaSubsession.o MPEG2TransportFileServerMediaSubsession.o ADTSAudioFileServerMediaSubsession.o DVVideoFileServerMediaSubsession.o AC3AudioFileServerMediaSubsession.o MPEG2TransportUDPServerMediaSubsession.o ProxyServerMediaSession.o QuickTimeFileSink.o QuickTimeGenericRTPSource.o AVIFileSink.o MPEG2IndexFromTransportStream.o MPEG2TransportStreamIndexFile.o MPEG2TransportStreamTrickModeFilter.o MatroskaFile.o MatroskaFileParser.o EBMLNumber.o MatroskaDemuxedTrack.o MatroskaFileServerDemux.o MatroskaFileServerMediaSubsession.o MP3AudioMatroskaFileServerMediaSubsession.o OggFile.o OggFileParser.o OggDemuxedTrack.o OggFileServerDemux.o OggFileServerMediaSubsession.o MPEG2TransportStreamDemux.o MPEG2TransportStreamDemuxedTrack.o MPEG2TransportStreamParser.o MPEG2TransportStreamParser_PAT.o MPEG2TransportStreamParser_PMT.o MPEG2TransportStreamParser_STREAM.o HLSSegment
 er.o TLSState.o MIKEY.o SRTPCryptographicContext.o HMAC_SHA1.o BitVector.o StreamParser.o DigestAuthentication.o ourMD5.o Base64.o Locale.o -L/home/fabrice/buildroot/output/per-package/live555/host/bin/../arm-buildroot-linux-uclibcgnueabi/sysroot/usr/lib -lssl -L/home/fabrice/buildroot/output/per-package/live555/host/bin/../arm-buildroot-linux-uclibcgnueabi/sysroot/usr/lib -lz -latomic -lpthread -lcrypto -lz -latomic -lpthread
/home/fabrice/buildroot/output/per-package/live555/host/bin/arm-linux-ar: invalid option -- 'L'

Fixes:
 - http://autobuild.buildroot.org/results/1299a05cfaabccc45e2878e969b443a973ec4c9e

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Reviewed-by: Michael Nosthoff<buildroot@heine.tech>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/ti-utils: fix build with gcc 10

Fixes:
 - http://autobuild.buildroot.org/results/e96c70910544085dc5299fa59f80ab9c5922fde2

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

package/libuhttpd: bump version to 3.3.1

Signed-off-by: Jianhui Zhao <zhaojh329@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/localedef: bump to glibc-2.31-54

resync the version with glibc package.

Signed-off-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/supertux: build squirrel builtin library with fPIC

Ensure that squirrel is compiled with -fPIC to allow linking the static
libraries with dynamically linked programs. This is not a requirement
for most architectures but is mandatory for ARM.

Fixes:
http://autobuild.buildroot.org/results/46e8f5e622ce450a89bc6d70f4bfd38182557901
http://autobuild.buildroot.org/results/a43720492d817e4555d728546da9114e3ccba952

Signed-off-by: Romain Naour <romain.naour@gmail.com>
Cc: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

Revert "package/supertux: fix build with RELRO"

This reverts commit 80be040817e008a021219bf4423df7550723d912 because
libsquirrel.so.0 and libsqstdlib.so.0 are missing in TARGET_DIR.

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

package/x11r7/xdriver-xf86-video-ati: Fix link failure with gcc 10

Backport an upstream patch.

Signed-off-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/elfutils: needs threads

threads is a mandatory dependency since version 0.178 and
https://sourceware.org/git/?p=elfutils.git;a=commitdiff;h=e9d4067a15eb100c0598a9895567aac6e9ed6a10

Fixes:
 - http://autobuild.buildroot.org/results/1c818a50530816ba80864efaacb724cb4bdeebf9

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

package/makedumpfile: add threads dependency in comment

Commit adb64a97e79bcb4e28450dd858654003ec8706c3 forgot to add threads
dependency in comment

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

DEVELOPERS: add Heiko Thiery as contact for altera-stapl

Signed-off-by: Heiko Thiery <heiko.thiery@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/altera-stapl: change to tagged version 0.3.3

Signed-off-by: Heiko Thiery <heiko.thiery@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/xfsprogs: bump to version 5.8.0

- Update hash of GPL-2.0 file because spdx tags have been updated by
  https://git.kernel.org/pub/scm/fs/xfs/xfsprogs-dev.git/commit/?id=c7498b6911d03f2bd7f74e9f9862d8d5fbd5b5da
- Drop third patch (already in version)
- Update indentation in hash file (two spaces)

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

package/civetweb: bump to version 1.13

https://github.com/civetweb/civetweb/releases/tag/v1.13

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

package/memcached: bump to version 1.6.7

- Drop patches (already in version) and so autoreconf
- Update indentation in hash file (two spaces)

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

package/links: bump to version 2.21

Update indentation in hash file (two spaces)

http://links.twibright.com/download/ChangeLog

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

package/gcc: remove gcc 7.5.0

gcc 10.2 is around, gcc 9.3 is the default version, so drop
7.5 in order to reduce the gcc choice.

See:
https://gcc.gnu.org/legacy-ml/gcc/2019-11/msg00099.html

The last defconfig using gcc 7 (roseapplepi_defconfig) has been
updated by 9cd065438020582011f393bd71308717b07c5a71.

Signed-off-by: Romain Naour <romain.naour@gmail.com>
[Peter: add Config.in.legacy entry]
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/paho-mqtt-c: bump to version 1.3.6

Update LICENSE hash, EDL version has been fixed with
https://github.com/eclipse/paho.mqtt.c/commit/34ec96cac554c4ae1527b92433730233f1bdca40

https://github.com/eclipse/paho.mqtt.c/milestone/11?closed=1

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

package/fluidsynth: bump to version 2.1.5

Drop patch (already in version)

https://github.com/FluidSynth/fluidsynth/releases/tag/v2.1.5

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

package/numactl: bump to version 2.0.14

Also update hash file formatting (2 spaces)

Signed-off-by: Pierre-Jean Texier <pjtexier@koncepto.io>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/php: security bump to version 7.4.11

- Fix CVE-2020-7069: In PHP versions 7.2.x below 7.2.34, 7.3.x below
  7.3.23 and 7.4.x below 7.4.11, when AES-CCM mode is used with
  openssl_encrypt() function with 12 bytes IV, only first 7 bytes of the
  IV is actually used. This can lead to both decreased security and
  incorrect encryption data.
- Fix CVE-2020-7070: In PHP versions 7.2.x below 7.2.34, 7.3.x below
  7.3.23 and 7.4.x below 7.4.11, when PHP is processing incoming HTTP
  cookie values, the cookie names are url-decoded. This may lead to
  cookies with prefixes like __Host confused with cookies that decode to
  such prefix, thus leading to an attacker being able to forge cookie
  which is supposed to be secure. See also CVE-2020-8184 for more
  information.

https://www.php.net/ChangeLog-7.php#7.4.11

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

package/python-cycler: drop redundant python|python3 dependencies

All the python packages are inside a

if BR2_PACKAGE_PYTHON || BR2_PACKAGE_PYTHON3

conditional, so no need to repeat it in the Config.in.

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

package/libmbim: bump to version 1.24.4

https://lists.freedesktop.org/archives/libmbim-devel/2020-October/001127.html

Signed-off-by: Aleksander Morgado <aleksander@aleksander.es>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/powerpc-utils: bump to version 1.3.8

Set --disable-werror which has been added with
https://github.com/ibm-power-utilities/powerpc-utils/commit/3d8a7f5e2e3b735570ab99bb3a340ff7e0c14841

https://github.com/ibm-power-utilities/powerpc-utils/blob/v1.3.8/Changelog

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

support/scripts/setlocalversion: fix/improve Mercurial output

Commit 9e4ffdc8cfdf4c73f4fa8c66259a5aadaee4ae88 modified the output of
'setlocalversion' so that the Buildroot version tag is included in the
output, the version part was added in Makefile.

Due to differences in behavior of the used git and Mercurial commands, this
caused different output for the Mercurial case, in BR2_VERSION_FULL and thus
/etc/os-release and 'make print-version'. Assuming the official Buildroot
releases are tagged and no project-specific tags are present, the output
after commit 9e4ffdc8cfdf4c73f4fa8c66259a5aadaee4ae88 is:
    -hg<commit>
whereas it is expected to be something like:
    2020.02.6-hg<commit>

Change the Mercurial case in setlocalversion to behave similar to git,
looking up the latest tag if the current revision is not itself tagged.

The number of commits after the latest tag is not added, unlike in git, as
this value is not commonly present in Mercurial output, and its added value
can be disputed in this context. Even one commit could bring a huge change
to the sources, so in order to interpret the number one has to look at the
repository anyhow, in which case the commit ID can just be used.

Signed-off-by: Thomas De Schampheleire <thomas.de_schampheleire@nokia.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/lua: split hash files, add license hash for 5.3 / 5.4

The content of the license file (doc/readme.html) differs between lua 5.3
and 5.4, so we cannot use a shared .hash file for all versions.

Signed-off-by: Francois Perrad <francois.perrad@gadz.org>
[Peter: extend commit message]
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/librelp: needs threads

Fixes:
 - http://autobuild.buildroot.org/results/e09f6eb989388cec2dab06391a8b6580aa0ff118

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

board/qemu/riscv64-virt: update qemu cmdline for Qemu >= 5.1

Since 52f188140cd28e90103edf67db6c2cabb979f5d6 (qemu version bump to 5.1),
the image generated by qemu_riscv64_virt_defconfig doesn't boot anyore with
the following error:

rom: requested regions overlap (rom phdr #0: [...]/images//fw_jump.elf. free=0x000000008000e240, addr=0x0000000080000000)
qemu-system-riscv64: rom check and register reset failed

Update the qemu command line as described in the Qemu wiki for riscv64 [1]

Fixes:
https://gitlab.com/jugurthaB/buildroot/-/jobs/686104707

[1] https://wiki.qemu.org/Documentation/Platforms/RISCV#Booting_64-bit_OpenEmbedded_Images

Signed-off-by: Romain Naour <romain.naour@gmail.com>
Cc: Jugurtha BELKALEM <jugurtha.belkalem@smile.fr>
Cc: Alistair Francis <alistair.francis@wdc.com>
Cc: Mark Corbin <mark@dibsco.co.uk>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

configs/qemu_arm_versatile_defconfig: increase SD card image size to 64MiB

Since Qemu 5.1, this defconfig doesn't boot due to the to small SD card image size (60MB).

qemu-system-arm: sd_init failed: Invalid SD card size: 60 MiB
SD card size has to be a power of 2, e.g. 64 MiB.
You can resize disk images with 'qemu-img resize <imagefile> <new-size>'
(note that this will lose data if you make the image smaller than it currently is).
qemu-system-arm: sd_init failed

From [1]:
"While the possibility to use small SD card images has been seen as
a feature, it became a bug with CVE-2020-13253, where the guest is
able to do OOB read/write accesses past the image size end."

Fixes:
https://gitlab.com/kubu93/buildroot/-/jobs/766482935

[1] https://git.qemu.org/?p=qemu.git;a=commitdiff;h=a9bcedd15a5834ca9ae6c3a97933e85ac7edbd36

Signed-off-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/python-fire.mk: drop test_components_py3.py file for python 2.x to fix pyfile issue

Fixes:
http://autobuild.buildroot.net/results/72e0cc78194a1b93bf26a50742e59a1e93bde1d1/

fire has conditional logic to load test_components_py3.py when running under
Python 3.x:

if six.PY3:
  from fire import test_components_py3 as py3

pycompile unfortunately errors out on it:

../scripts/pycompile.py ..

error:   File "/usr/lib/python2.7/site-packages/fire/test_components_py3.py", line 18
    def identity(arg1, arg2: int, arg3=10, arg4: int = 20, *arg5,
                           ^
SyntaxError: invalid syntax

As a workaround, simply drop the unusable _py3 file from TARGET_DIR if
building for python 2.x.

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

package/python-aenum: drop test_v3.py file for python 2.x to fix pycompile issue

Fixes:
http://autobuild.buildroot.net/results/4ca459d54545c0e20b0f0cdc63bd81844ecd7f36/

aenum has conditional logic to load python 3.x code located in test_v3.py:

if pyver >= 3.0:
    from aenum.test_v3 import TestEnumV3, TestOrderV3, TestNamedTupleV3

And contains logic in setup.py to drop that file during setup.py install if
building for python 2.x:

py3_only = ('aenum/test_v3.py', )
..
if __name__ == '__main__':
    if 'install' in sys.argv:
        import os, sys
..
        if sys.version_info[0] != 3:
            for file in py3_only:
                try:
                    os.unlink(file)

But this doesn't work in Buildroot as pkg-python.dk first does setup.py
build (which copies test_v3.py to the build directory) before setup.py
install, so test_v3.py gets installed, leading to errors from pycompile:

error:   File "/usr/lib/python2.7/site-packages/aenum/test_v3.py", line 12
    class MagicAutoNumberEnum(Enum, settings=AutoNumber):
                                            ^
SyntaxError: invalid syntax

As a workaround, add a hook to drop it from the target directory when
building for python 2.x.

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

package/go: fix GO_ARCH for s390x arch

Fixes:
- http://autobuild.buildroot.net/results/82c440825290804c2fd6a92d916c93d2934e0a95
- http://autobuild.buildroot.net/results/e8f8e232fb6bf6167f25c82026f4e2321cca9c26
- http://autobuild.buildroot.net/results/7d4137c439ade81ab7bf141ad6504a69746d6dbc

Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/xfsprogs: fix build with gcc 10

Fixes:
 - http://autobuild.buildroot.org/results/18455090605f3691282778b6a64a624b109037d6

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

package/freetype: unconditionally disable harfbuzz

Commit 939e714393e9d7f60f3a198d831608b08a25662e added an optional
harfbuzz dependency to freetype but this creates a circular dependency
so unconditionally disable it

Fixes:
 - http://autobuild.buildroot.org/results/3cc4ce3207a253186a9c4f8f5151ea0fc0854a28

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

package/i2c-tools: bump to version 4.2

Use upstream provided hash.

Update README hash because of unrelated content changes.

Format hashes with two spaces delimiter.

Cc: Angelo Compagnucci <angelo.compagnucci@gmail.com>
Cc: Matt Weber <matthew.weber@rockwellcollins.com>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

{linux, linux-headers}: bump 4.{4, 9, 14, 19}.x / 5.{4, 8}.x series

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

package/python3: bump to version 3.8.6

Contains a number of bugfixes. For details, see the changelog:

https://docs.python.org/release/3.8.6/whatsnew/changelog.html#changelog

Update the license hash for the addition of a note stating that the examples
and documentation is now dual licensed under the PSF and a Zero-Clause BSD
license since:

https://github.com/python/cpython/commit/9fef7c54a0adfade5ec94259d97f22e05fe9e3e3

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

package/netatalk: fix build with gcc 10

Fixes:
 - http://autobuild.buildroot.org/results/c3381fd3072901871fdd491dc3d4f66e403c7a70

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

package/freetype: add brotli optional dependency

brotli is an optional dependency (enabled by default) since version
2.10.2 and
https://git.savannah.gnu.org/cgit/freetype/freetype2.git/commit/?id=9443a1bc3e6cfa315db976ff80f6c5e91b12b387

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

package/freetype: add harfbuzz optional dependency

harfbuzz is an optional dependency (enabled by default) since version
2.5.3 and
https://git.savannah.gnu.org/cgit/freetype/freetype2.git/commit/?id=86026a47b345a8c254dd5e6be77bf116737cdafb

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

package/freetype: drop libpng workaround

libpng workaround which has been added with commit
f7313cadf2ca9e0195a873f62f6b6c38a58a3630 is not needed since version
2.5.3 and
https://git.savannah.gnu.org/cgit/freetype/freetype2.git/commit/?id=00c79ed9680a0d7a367c6914adc7485391299542

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

boot/barebox: bump version to 2020.09.0

Signed-off-by: Bartosz Bilas <b.bilas@grinn-global.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/cppzmq: bump to version 4.7.0

https://github.com/zeromq/cppzmq/releases/tag/v4.7.0

Update indentation in hash file (two spaces)

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

package/nodejs: security bump to version 12.18.4

Fixes the following security issues:

- CVE-2020-8201: HTTP Request Smuggling due to CR-to-Hyphen conversion

  Affected Node.js versions converted carriage returns in HTTP request
  headers to a hyphen before parsing.  This can lead to HTTP Request
  Smuggling as it is a non-standard interpretation of the header.

  Impacts:
    All versions of the 14.x and 12.x releases line

- CVE-2020-8252: fs.realpath.native may cause buffer overflow

  libuv's realpath implementation incorrectly determined the buffer size
  which can result in a buffer overflow if the resolved path is longer than
  256 bytes.

  Impacts:
    All versions of the 10.x release line
    All versions of the 12.x release line

For more details, see the advisory:
https://nodejs.org/en/blog/vulnerability/september-2020-security-releases/

Adjust license hash for the addition of the BSD-3c licensed highlight.js:
https://github.com/nodejs/node/commit/6f8b7a85d239129273948386c34775810f2dc4a3

Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/elfutils: needs host-pkgconf

host-pkgconf is a mandatory dependency since version 0.180 and
https://sourceware.org/git/?p=elfutils.git;a=commit;h=f7f0cdc59a13780938ae3f578955737a75e60ea9

Fixes:
 - http://autobuild.buildroot.org/results/123bdb3e7244800e1717f968c1d371e51f7fff2b

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

package/sentry-cli: needs host-zlib

Fixes:
 - http://autobuild.buildroot.org/results/bca6f0ef30bc8fdfa98f48c96de368dc4644a1f5

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

package/gcc: drop leftovers from PowerPC conditional patching

We used to have a conditional patch applied on PowerPC soft-float, but
this logic was dropped in commit
0c82f3f63563c4d4d7e034d0e9e4ceb766d92322 ("package/gcc: remove powerpc
conditional patching logic"). However, we still have some related
leftovers in the calculation of the hashes for ccache, which can now
be dropped.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Cc: Romain Naour <romain.naour@gmail.com>
Cc: Arnout Vandecappelle <arnout@mind.be>
Acked-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/keyutils: fix build without C++

Create cxx.stamp to disable the check that keyutils.h has a valid C++
syntax that has been added with
https://git.kernel.org/pub/scm/linux/kernel/git/dhowells/keyutils.git/commit/?id=991c272062a4300bbd5fe22dce17ab2fb5126e3f

Fixes:
 - http://autobuild.buildroot.org/results/53ec9ca02798a88cd9ab773af10fc8f44ce9d1a2

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

package/stellarium: bump version to 0.20.3

Removed md5 hash.

Release notes:
http://stellarium.org/release/2020/09/27/stellarium-0.20.3.html

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/elfutils: fix host build

Commit 52c5c6050c5359288ca48c85024aebd169fab258 disabled debuginfod on
target but not on host resulting in the following build failure:

configure: error: dependencies not found, use --disable-libdebuginfod to disable or --enable-libdebuginfod=dummy to build a (bootstrap) dummy library.

Fixes:
 - http://autobuild.buildroot.org/results/97713c7e281313826067af559337832f22c3cb09

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

package/gcc: help libbacktrace detection of sync builtins

The logic in libbacktrace/configure.ac to detect if __sync builtins
are available assumes they are as soon as target_subdir is not
empty, i.e when cross-compiling. However, some platforms do not have
__sync builtins, so help the configure script a bit.

"libbacktrace_cv_sys_sync=no" is lost when it is added to
HOST_GCC_COMMON_CONF_ENV because the environment is not exported
when executing the libbacktrace configure script.

Use target_configargs to force "libbacktrace_cv_sys_sync=no" when
executiong the libbacktrace configure script.

Fixes:
https://gitlab.com/bootlin/toolchains-builder/-/jobs/729359681

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
[Romain: use target_configargs="libbacktrace_cv_sys_sync=no"]
Signed-off-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/ruby: drop useless SuperH CFLAGS tweaks

We dropped the CodeBench 2012.09 toolchain for SuperH a long time,
and with recent toolchains, Ruby at -Os builds perfectly fine, so
there's no reason to keep this work-around.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
[yann.morin.1998@free.fr: slight rewording in the commit log]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/ipmiutil: bump to version 3.1.7

- Drop patch (not needed since
  https://github.com/arcress0/ipmiutil/commit/32b6e5634d773150338b60585e9c258cd575e3d9)
- Drop autoreconf
- Update indentation in hash file (two spaces)

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

package/ghostscript: bump to version 9.53.2

- Drop first patch (already in version)
- Drop second patch (not needed since
  https://git.ghostscript.com/?p=ghostpdl.git;a=commit;h=f0d19bf181a156d011dba422ae4d165b36b0af7e)
- Drop autoreconf

https://www.ghostscript.com/doc/9.53.2/News.htm

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

package/bison: disable libtextstyle

Recent bison versions added support for fancy error output using gettext
provided libtextstyle. The Buildroot gettext version does not provided
all needed features. However host-bison might detect host installed
libtextstyle at configure time, but fail at run time because of missing
symbols in Buildroot provided libtextstyle.

We don't really need fancy output of host packages, so just disable
libtextstyle support.

Fixes (libpcap):
http://autobuild.buildroot.net/results/058e8ec90da3c06e31b31eb94541331cced44db3/
http://autobuild.buildroot.net/results/d99dda99722c53730b5964bd4ff1e1281c4ad759/
http://autobuild.buildroot.net/results/056fd1246554fe7def416429620175e86530c0a7/

Cc: Xogium <contact@xogium.me>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Tested-by: Xogium <contact@xogium.me>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/babeltrace2: fix tests with uclibc

It should be noted that upstream does not want to add a --disable-tests
option: https://review.lttng.org/c/babeltrace/+/4075

Fixes:
 - http://autobuild.buildroot.org/results/1e50e1ff90c1ac2af2223365dc0cd6e1914bcd75

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

package/frr: link with latomic if needed

Fixes:
 - http://autobuild.buildroot.org/results/c98/c989f58eabd8f012798ade75e5b0f2a93b91c82d

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

package/rsh-redone: disable rlogind on musl

Commit f89ca996b6deb660b53c33196b8b309fac659bf9 enabled linux-pam on
musl however rlogind fails to build on musl because it uses logwtmp:

/home/giuliobenetti/autobuild/run/instance-1/output-1/host/opt/ext-toolchain/bin/../lib/gcc/x86_64-buildroot-linux-musl/8.3.0/../../../../x86_64-buildroot-linux-musl/bin/ld: rlogind.c:(.text.startup+0x9ec): undefined reference to `logwtmp'

As this package is unmaintained, just disable back rlogind on musl

Fixes:
 - http://autobuild.buildroot.org/results/f2eb1c0bab7c6c32d6c561da0f85a530bd419206

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

package/log4cxx: fix build without wchar

Fix build with log4cxx in version 0.11.0

Fixes:
 - http://autobuild.buildroot.org/results/bab5329fdeb894471bfd5192ce04d3fbd2f9be5c

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

package/luarocks: refactor infrastructure with DEPS_DIR

DEPS_DIR is a new CLI option introduced by the version 3.4.0,
which avoids the use of `external_deps_dirs` in the config files.

now, the host config doesn't need to be altered.

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

package/keyutils: bump to version 1.6.3

- update home page URL
- change download to new location

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

package/gdb: also disable gprof

When gdb is built from sources fetched from Git, it contains both the
gdb and the binutils code base. In order to really build only gdb, we
disable a number of binutils components in the
GDB_DISABLE_BINUTILS_CONF_OPTS variable: --disable-binutils,
--disable-ld, --disable-gas, etc. However, gprof was still being
built, so disable it as well.

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

package/gdb: fix build of gdbserver-only on the ARC gdb version

The GDB version used on ARC is based on a recent gdb code base, post
gdb 9.2. This recent gdb code base, which pre-figures what will be in
gdb 10, has a significant change: gdbserver is not longer in
gdb/gdbserver, but at the top-level, and the mechanism to build
gdbserver only has changed. Due to this change, a build of ARC GDB for
gdbserver only fails with:

/bin/bash: line 0: cd: /opt/output/build/gdb-arc-2020.03-release-gdb/gdb/gdbserver: No such file or directory

This commit adjusts gdb.mk to support four cases:

 - "old" gdb, gdbserver only
 - "old" gdb, with the gdb debugger (and optionally gdbserver as well)
 - "new" gdb, gdbserver only
 - "new" gdb, with the gdb debugger (and optionally gdbserver as well)

A boolean GDB_GDBSERVER_TOPLEVEL is introduced to differentiate
between the old and new gdb, it is set to "y" for gdb versions that
have the gdbserver code at the top-level. For now, only the ARC
version sets it, but in the future, upstream gdb version 10 will also
have to set it.

Here is the behavior, for each case:

 (1) "old" gdb, gdbserver only

     We set GDB_SUBDIR to gdb/gdbserver, so only the configure script
     in this folder gets called.

     --enable-gdbserver --disable-gdb are passed in CONF_OPTS.

 (2) "old" gdb, with the gdb debugger (and optionally gdbserver as well)

     We set GDB_SUBDIR to build/, an empty directory which allows to
     do an out of tree build, which is mandatory for a full gdb build
     since gdb 9.x.

     --enable-gdb is passed in CONF_OPTS as well as --enable-gdbserver
     or --disable-gdbserver depending on whether gdbserver is enabled
     as well.

 (3) "new" gdb, gdbserver only

     We set GDB_SUBDIR to build/, an empty directory which allows to
     do an out of tree build, which is mandatory for a full gdb build
     since gdb 9.x.

     --enable-gdbserver --disable-gdb are passed in CONF_OPTS.

 (4) "new" gdb, with the gdb debugger (and optionally gdbserver as well)

     We set GDB_SUBDIR to build/, an empty directory which allows to
     do an out of tree build, which is mandatory for a full gdb build
     since gdb 9.x.

     --enable-gdb is passed in CONF_OPTS as well as --enable-gdbserver
     or --disable-gdbserver depending on whether gdbserver is enabled
     as well.

In addition to these changes, some related changes are done as well:

 - We re-enable building both gdb and gdbserver on ARC, as it works
   again.

 - We only pass --with-curses when curses is really provided, i.e when
   the full debugger is being built.

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

package/sentry-cli: fix dependency on host-rustc

Replace SENTRY_CLI_DEPENDENCIES by HOST_SENTRY_CLI_DEPENDENCIES as
sentry-cli is a host-generic-package

Fixes:
 - http://autobuild.buildroot.org/results/ce566ec91a8f7d69f2525892a7a034c829db9dad

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

package/qt5base: XCB support needs xcb-util-renderutil

The qt5base feature 'xcb_syslibs' is a compile check
using a xcb_render_util call (see qt5base-5.15.1/src/gui/configure.json).

Fixes:
  - http://autobuild.buildroot.net/results/3c05e597c073b823f8b091e17b676ed3ecf6b5b3

  ERROR: Feature 'xcb' was enabled, but the pre-condition 'features.thread && libs.xcb && tests.xcb_syslibs && features.xkbcommon-x11' failed.

Signed-off-by: Peter Seiderer <ps.report@gmx.net>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/linuxptp: bump version to 3.1

From release note:

Version 3.1 - September 2020

This release offers substantial new functionality. Users are encouraged
to upgrade. The most notable changes are as follows.

* Fixes non-functional NMEA mode of the ts2phc program (since 3.0)
* Improves system time synchronization via the PTP_SYS_OFFSET_PRECISE
  and PTP_SYS_OFFSET_EXTENDED ioctls.
* Supports implementing a PTP GM clock by using a GPS radio or other
  PPS time source.
* Provides per-port statistics via the management interface.
* Implements IEEE 1588 v2.1 slave event monitoring.
* Adds Automotive Profile support.y

 utils/test-pkg -p linuxptp
                             br-arm-full [1/6]: OK
                  br-arm-cortex-a9-glibc [2/6]: OK
                   br-arm-cortex-m4-full [3/6]: OK
                          br-x86-64-musl [4/6]: OK
                      br-arm-full-static [5/6]: OK
                            sourcery-arm [6/6]: OK
6 builds, 0 skipped, 0 build failed, 0 legal-info failed

Signed-off-by: Heiko Thiery <heiko.thiery@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/liberation: bump to version 2.1.1

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

package/qt5/qt5base: allow to use imx-gpu-viv as GBM provider

Needed in order to use eglfs_kms platform backend.

Signed-off-by: Gary Bisson <gary.bisson@boundarydevices.com>
Reviewed-by: Peter Seiderer <ps.report@gmx.net>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/webkitgtk: bump to version 2.30.1

This is a minor release which fixes a regression found in 2.30.0.
Release notes:

  https://webkitgtk.org/2020/09/21/webkitgtk2.30.1-released.html

Signed-off-by: Adrian Perez de Castro <aperez@igalia.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/open2300: switch to github

Upstream repository has been dead for quite a while now, and in fact
we are using s.b.o as the "official" location now.

There however exists a Github repository, which is still alive. That
repository has not been touched in a looooong while, but it's still
up, and there is a revision which actually gives the exact same
content (although the hash changes, because the directory name has
changed, from open2300-12/ to open2300-f2198.../).

Also fix spaces in hash file.

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

arch/Config.in.s390x: drop redundant depends on BR2_s390x

The whole arch/Config.in.s390x file is only included by arch/Config.in
if BR2_s390x=y, so having "depends on" conditions on each options of
the choice doesn't make much sense.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
[yann.morin.1998@free.fr: drop first, empty line]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/pango: bump to version 1.46.2

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

package/python-werkzeug: bump version to 1.0.1

Signed-off-by: Michael Nosthoff <buildroot@heine.tech>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/elfutils: bump to version 0.181

Refreshed patch #1 and disable debuginfod.

The debuginfod tools needs to hook into serverinfrastructure,
this currently only exists for Fedora.

Signed-off-by: Norbert Lange <nolange79@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

support/config-fragments: add s390x z13 to autobuild configs

Add a minimal s390x s13 autobuild configuration for the
internal toolchain with glibc.

Signed-off-by: Romain Naour <romain.naour@gmail.com>
Cc: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/python-txtorcon.mk: drop _py3 file for python 2.x to fix pycompile issue

Fixes:
http://autobuild.buildroot.net/results/76b580000e6311e88584874f942517badd6fadf6/

python-txtorcon DOES support python 2.x, but it contains some optional
python 3 / async code in controller_py3.py which is conditionally used from
controller.py:

try:
    from .controller_py3 import _AsyncOnionAuthContext
    HAVE_ASYNC = True
except Exception:
    HAVE_ASYNC = False

pycompile unfortunately errors out on the async code:

../scripts/pycompile.py ..
error:   File "/usr/lib/python2.7/site-packages/txtorcon/controller_py3.py", line 13
    async def __aenter__(self):
            ^
SyntaxError: invalid syntax

As a workaround, simply drop the unusable _py3 file from TARGET_DIR if
building for python 2.x.

Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>