- Feb 27, 2016
-
-
Romain Naour authored
mraa build system is looking for a python interpreter >= 2.7 on the host due to a typo in MRAA_CONF_OPTS. We must use -DBUILDTESTS=OFF instead of -DTESTS=OFF. Fixes: http://autobuild.buildroot.net/results/5f0/5f047c4c28b34f6b1bf2306a2b00bcc6a7481c01 Signed-off-by:
Romain Naour <romain.naour@gmail.com> Signed-off-by:
Peter Korsgaard <peter@korsgaard.com>
-
Arnout Vandecappelle authored
Previous attempts to fix it failed [1], so disable it instead. [1] http://patchwork.ozlabs.org/patch/509731/ Signed-off-by:
Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be> Signed-off-by:
-
Romain Naour authored
pax-utils require a toolchain with kernel headers >= 3.5 to provide PR_SET_NO_NEW_PRIVS. Fixes: http://autobuild.buildroot.net/results/a9e/a9e0287e67c03e5e385bf54578aa6c5c85206ec2 Signed-off-by:
-
Yann E. MORIN authored
The way we use it, gzip will store the current time in the header, which leads to unreproducible archives. Fix that by telling gzip to not store the name and date of the file it compresses, with the -n option. Since it compresses its stdin, there was already no filename stored; now there's even no date stored. Note: gzip has had -n since at least 1.2.4, released in 1993, so virtually every gzip out there nowadays has it. Signed-off-by:
"Yann E. MORIN" <yann.morin.1998@free.fr> Signed-off-by:
-
Vicente Olivert Riera authored
We need to pass an argument to ld for setting the endianness when building it for MIPS architecture, otherwise the default one will always be used (which is big endian) and the compilation for little endian will always fail showing an error like this one: LD foo.o mips-linux-gnu-ld: foo.o: compiled for a little endian system and target is big endian Signed-off-by:
Vicente Olivert Riera <Vincent.Riera@imgtec.com> Reviewed-by:
-
Gustavo Zacarias authored
Fixes: CVE-2016-2521 - DLL hijacking vulnerability. CVE-2016-2522 - ASN.1 BER dissector crash. CVE-2016-2523 - DNP dissector infinite loop. CVE-2016-2524 - X.509AF dissector crash. CVE-2016-2525 - HTTP/2 dissector crash. CVE-2016-2526 - HiQnet dissector crash. CVE-2016-2527 - 3GPP TS 32.423 Trace file parser crash. CVE-2016-2528 - LBMC dissector crash. CVE-2016-2529 - iSeries file parser crash. CVE-2016-2530 - RSL dissector crash. CVE-2016-2531 - RSL dissector crash. CVE-2016-2532 - LLRP dissector crash. wnpa-sec-2016-12 - Ixia IxVeriWave file parser crash. wnpa-sec-2016-13 - IEEE 802.11 dissector crash. wnpa-sec-2016-14 - GSM A-bis OML dissector crash. wnpa-sec-2016-15 - ASN.1 BER dissector crash. wnpa-sec-2016-16 - SPICE dissector large loop. wnpa-sec-2016-17 - NFS dissector crash. wnpa-sec-2016-18 - ASN.1 BER dissector crash. Signed-off-by:
Gustavo Zacarias <gustavo@zacarias.com.ar> Signed-off-by:
Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
-
Ignacy Gawędzki authored
Signed-off-by:
Ignacy Gawędzki <ignacy.gawedzki@green-communications.fr> Acked-by:
-
Romain Naour authored
A variadic template template parameters [1] is used in QhullSet.h, this require a toolchain with (at least) gcc 4.4 [2][3]. Fixes (bfin toolchain with gcc 4.3): http://autobuild.buildroot.net/results/32f/32f324c7920ac46d393954de13b8a98a77a0c392 [1] http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2008/n2555.pdf [2] https://gcc.gnu.org/gcc-4.3/cxx0x_status.html [3] https://gcc.gnu.org/gcc-4.4/cxx0x_status.html Signed-off-by:
-
Romain Naour authored
The comment must be displayed with toolchain without C++ compiler support when BR2_STATIC_LIBS is not set. Signed-off-by:
-
Angelo Compagnucci authored
This patch fixes a leak in host compiler path for monodoc. This bug is solved upstream but the fix is not included in the latest stable release. Signed-off-by:
Angelo Compagnucci <angelo.compagnucci@gmail.com> Acked-by:
-
Peter Seiderer authored
With this implementation, the different licenses will be separated with <space><comma>. This is not nice, but to avoid it a pretty ugly construct is needed: QT_LICENSE = $(shell echo $(QT_LICENSE_BITS) | sed 's% , %, %g') Since the <space><comma> is just a minor nuisance, we're not going to that kind of complexity. Signed-off-by:
Peter Seiderer <ps.report@gmx.net> Acked-by:
-
- Feb 26, 2016
-
-
Luca Ceresoli authored
Several people have been asking what is the license of the patches provided by Buildroot. COPYING is the authoritative place to state it. Signed-off-by:
Luca Ceresoli <luca@lucaceresoli.net> Cc: "Yann E. MORIN" <yann.morin.1998@free.fr> Cc: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be> Cc: Peter Korsgaard <peter@korsgaard.com> Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Cc: Steve Calfee <stevecalfee@gmail.com> Acked-by:
-
Gustavo Zacarias authored
Signed-off-by:
-
Gustavo Zacarias authored
Signed-off-by:
-
- Feb 25, 2016
-
-
Romain Naour authored
As reported by Thomas Petazzoni [1], the libsmartcols library can't be used on no-MMU platform since fork() is used. [1] http://lists.busybox.net/pipermail/buildroot/2016-February/153414.html Signed-off-by:
-
Bernd Kuhls authored
When libcap was compiled before, avahi daemon will use it as optional dependency: $ output/host/usr/bin/x86_64-linux-readelf -a output/target/usr/sbin/avahi-daemon | grep NEEDED | sort 0x0000000000000001 (NEEDED) Shared library: [libavahi-common.so.3] 0x0000000000000001 (NEEDED) Shared library: [libavahi-core.so.7] 0x0000000000000001 (NEEDED) Shared library: [libcap.so.2] [...] The build system offers no option to en-/disable libcap support: http://git.0pointer.net/avahi.git/tree/configure.ac#n382 Signed-off-by:
Bernd Kuhls <bernd.kuhls@t-online.de> [Thomas: make the test unconditionally, even if libcap is only used by avahi-daemon for the moment.] Signed-off-by:
-
Maxime Hadjinlian authored
This patch fixes http://autobuild.buildroot.net/results/948/948f81d7ed2c080a675ee9fec754f6fda1fd905f/ The patches has been sent upstream: http://article.gmane.org/gmane.linux.nfs/76139 Signed-off-by:
Maxime Hadjinlian <maxime.hadjinlian@gmail.com> [Thomas: slightly tweak commit title.] Signed-off-by:
-
Gustavo Zacarias authored
Some egl/gles binary-only implementations need some extra help in the form of additional include directories, like rpi-userland. Fixes: http://autobuild.buildroot.net/results/083/083ce94b1ac61a60aa16906336612af59e32b18b/ Signed-off-by:
-
Rahul Bedarkar authored
wpa_supplicant is licensed under BSD-3c only and not dual licensed. Quoting from COPYING file This software was previously distributed under BSD/GPL v2 dual license terms that allowed either of those license alternatives to be selected. As of February 11, 2012, the project has chosen to use only the BSD license option for future distribution. As such, the GPL v2 license option is no longer used. Reviewed-by:
Abhimanyu Vishwakarma <Abhimanyu.Vishwakarma@imgtec.com> Signed-off-by:
Rahul Bedarkar <rahul.bedarkar@imgtec.com> Signed-off-by:
-
Rahul Bedarkar authored
hostapd is licensed under BSD-3c only and not dual licensed. Quoting from COPYING file This software was previously distributed under BSD/GPL v2 dual license terms that allowed either of those license alternatives to be selected. As of February 11, 2012, the project has chosen to use only the BSD license option for future distribution. As such, the GPL v2 license option is no longer used. Reviewed-by:
-
Rahul Bedarkar authored
dnsmasq is dual licensed so separate licenses with 'or' keyword. Reviewed-by:
Marcin Nowakowski <marcin.nowakowski@imgtec.com> Signed-off-by:
-
Angelo Compagnucci authored
This patch solves an host path leaking in /etc/mono/config that leads to searching libmonoposixhelper into the wrong path. This bug is fixed upstream, but not included in the current stable release. Fixes bug #8651. Signed-off-by:
-
Gustavo Zacarias authored
It uses __sync_fetch_and_add_4() and other 32-bit atomics, fixes: http://autobuild.buildroot.net/results/7ec/7ec4142d0f41234309a226a21a7da7acbece90e9/ Signed-off-by:
-
Gustavo Zacarias authored
It uses __sync_fetch_and_add_4() and other 32-bit atomics, fixes: http://autobuild.buildroot.net/results/610/610ef13ee1d62597f51b061cd1216e7a4e96abff/ Signed-off-by:
-
Luca Ceresoli authored
Adding a patch is a modification to Buildroot. Thus this sentence might be interpreted as "Patches are released under the Buildroot license". Being a legal matter, the last thing we want is to be ambiguous. Reword the sentence to avoid any misunderstandings. Signed-off-by:
-
- Feb 24, 2016
-
-
Gustavo Zacarias authored
It's BR2_aarch64, not aarch64 alone. Signed-off-by:
-
Gustavo Zacarias authored
They use __sync_add_and_fetch_4() and other 32-bit atomics, fixes: http://autobuild.buildroot.net/results/7ac/7acdbc54e26ad2ba4756512614192be0a086f563/ Also remove sparc from supported architectures since it won't grow atomics support magically. Signed-off-by:
-
Ignacy Gawędzki authored
Signed-off-by:
-
Gustavo Zacarias authored
Fixes: CVE-2016-0787 - diffie_hellman_sha256: convert bytes to bits. Signed-off-by:
-
Gustavo Zacarias authored
Fixes: CVE-2016-0739 - Bits/bytes confusion resulting in truncated Difffie-Hellman secret length. Signed-off-by:
-
Thomas Petazzoni authored
The patch added in commit 21ed7a92 to fix the musl build had the unfortunate consequence to break the build on uClibc. Thanks to Bernd's work with upstream, both the musl and uclibc issues have been fixed upstream, so this patch updates our numactl package to use the two upstream commits instead of our own private patch. Fixes: http://autobuild.buildroot.net/results/a8729c301a52473899a3e3110802a41cc8afe9db/ Signed-off-by:
-
Lada Trimasova authored
ARC gcc understands "-pie" option and attempts to generate PIE binaries as of today PIE is not really supported for user-space applications. So we provide option which checks if compiler understands relro and pie options and disable PIE detection if building for ARC. Also AUTORECONF option should be added because of modified configure.in and Makefile.am files. Signed-off-by:
Lada Trimasova <ltrimas@synopsys.com> Cc: Alexey Brodkin <abrodkin@synopsys.com> Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Cc: Peter Korsgaard <peter@korsgaard.com> Signed-off-by: Lada Trimasova <ltrimas@synopsys.com<mailto:ltrimas@synopsys.com>> Signed-off-by:
-
- Feb 23, 2016
-
-
Arnout Vandecappelle authored
Thomas P. has sent a few big feedback mails recently that explain how a patch should be formatted. Indeed, this was not explained much in the manual, so add a section that explains how patches should be formatted. This is based heavily on the feedback that Thomas P. gave. Also, specific examples for new packages and version bumps are added. This will allow us to refer to https://buildroot.org/manual.html#submitting-patches in the future instead of composing long mails. Signed-off-by:
Yegor Yefremov <yegorslists@googlemail.com> [Thomas: rewrap to our normal formatting practice.] Signed-off-by:
-
Arnout Vandecappelle authored
In subsequent patches, we will add more explanation about how to prepare patches, so it will be worthwhile to have a separate section for the series preparation. Signed-off-by:
-
Neha Bairathi authored
LUAPATH is used as the path for installing lua binding. As buildroot specifies the install prefix, STAGING_DIR gets appended to TARGET_DIR which is incorrect path for installation, so use relative path. Reviewed-by:
Neha Bairathi <Neha.Bairathi@imgtec.com> Acked-by:
-
Bernd Kuhls authored
When orc was compiled before, gnuradio will use it as optional dependency: $ output/host/usr/bin/x86_64-linux-readelf -a output/target/usr/bin/gnuradio-config-info | grep NEEDED [...] 0x0000000000000001 (NEEDED) Shared library: [liborc-0.4.so.0] [...] Apparently there is no option provided by the build system to en-/disable support for orc: https://github.com/gnuradio/volk/blob/41b4df055b2c8d4f2c23ad2bdc60c0b47b4f0dd8/lib/CMakeLists.txt#L449 Signed-off-by:
"Gwenhael Goavec-Merou" <gwenhael.goavec-merou@trabucayre.com> Tested-by:
-
Peter Korsgaard authored
Commit 24dfbe71 (arch/arm: do not distinguish revisions of ARM1136JF-S) removed the r0 variant of the arm1136jf-s, but didn't update the defconfig - So it ends up using arm926 instead. Fix it by selecting the correct symbol instead. Signed-off-by:
-
Peter Korsgaard authored
Similar to the issue with the Atmel A5 based boards, the VFP is optional on A9 - So we need to enable it to be able to use EABIhf. Signed-off-by: -
Alexandre Belloni authored
The recent changes in Config.in.arm regarding the VFP selection broke the atmel_sama5d* defconfigs. Ensure EABIhf is selected as all the sama5 have a vfp. Signed-off-by:
Alexandre Belloni <alexandre.belloni@free-electrons.com> Signed-off-by:
-
- Feb 22, 2016
-
-
Peter Korsgaard authored
Commit b3fa8e24 (configs: remove calao_snowball_defconfig) dropped the outdated snowball defconfig, but forgot to remove the board files. Signed-off-by:
-
