The ARC gcc compiler now defines _REENTRANT when -pthread is
passed. This issue was fixed upstream in gcc commit
de4c7f60f2891193bf3f5da823b17fa0d7fd4830, which is part of gcc
7.x. Therefore, both the gcc 7.x and 8.x versions, which can be
selected for the ARC architecture, are fixed, making the libmicrohttpd
work around useless.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
This reverts commit 2e57e835bf, which is
no longer needed now that the RISC-V gcc has been patched to define
_REENTRANT when -pthread is passed.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Commit 19448f40a0 (gnutls: use included unistring unless libunistring is
selected) made libunistring an optional dependency. So now gnutls no longer
depend on wchar.
Drop wchar dependency of libmicrohttpd and libsoup, which are gnutls
reverse dependencies that do not depend themselves on wchar.
Update the comments in libsoup and taskd; the wchar dependency is now
not due to gnutls.
Cc: Matt Weber <matthew.weber@rockwellcollins.com>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Drop libgcrypt dependency for SSL support; with newer version of gnutls
libgcrypt is not needed.
Drop the LIBS fix for static linking with gnutls that has been
introduced in commit 2d3d6d258b (libmicrohttpd: fix static linking with
gnutls). gnutls does not support static linking anymore.
Fix a comment typo.
Add license hash.
Cc: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The gnutils code uses __attribute__((constructor)) and
__attribute__((destructor)) to call constructor/desctructor when a
shared library is loaded.
Constructor/desctructor are not used when a static library is used
(except when if -Wl,--whole-archive -lgnutls -Wno-whole-archive is
used, not tested).
Even if gnutls initialization (_gnutls_global_init()) may be
called manually, the gnutls maintainer said it's not supported [1].
"Note that static linking applications with gnutls is not something
supported. gnutls relies on library constructors and destructors
which are not loaded when linking statically."
Now the gnutls script warns about static linking [2].
So disable gnutls statically by adding "depends on !BR2_STATIC_LIBS"
at Kconfig level and --disable-static in GNUTLS_CONF_OPTS.
Fixes:
[taskd] http://autobuild.buildroot.net/results/c2d/c2dd5c1c9dc87d2943c15e58ee56e67d7375368c
[ffmpeg] http://autobuild.buildroot.net/results/892/8926d319d6d1cd1ee72239ad7d9ca869d2355628
[sngrep] http://autobuild.buildroot.net/results/f7f/f7fb42d3742f6f01000a0d181e0c785640284405
[1] https://gitlab.com/gnutls/gnutls/issues/203
[2] 6b74888679
Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
[Romain: merge our two patches together
add some option comment
disable static libgnutls.a
add sngrep autobuilder reference]
Signed-off-by: Romain Naour <romain.naour@gmail.com>
Tested-by: Bernd Kuhls <bernd.kuhls@t-online.de>
[Thomas: do not disable libgnutls.a]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
We want to use SPDX identifier for license string as much as possible.
SPDX short identifier for LGPLv2.1/LGPLv2.1+ is LGPL-2.1/LGPL-2.1+.
This change is done using following command.
find . -name "*.mk" | xargs sed -ri '/LICENSE( )?[\+:]?=/s/LGPLv2.1(\+)?/LGPL-2.1\1/g'
Signed-off-by: Rahul Bedarkar <rahulbedarkar89@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
This reverts commit 4807388410.
The commit broke the build with the pre-built Blackfin toolchain.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Bfin workaround is not required if the gcc bfin patch is applied.
(gcc: enable _REENTRANT for bfin when -lpthread is used)
Signed-off-by: Waldemar Brodkorb <wbx@openadk.org>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Due to compiler deficiences, gcc on ARC and Blackfin doesn't define
_REENTRANT, which the latest version of the ax_pthread.m4
autoconf-archive macros looks at to determine whether thread support is
available or not.
Fixing the compilers is definitely the right thing to do, but it's not
doable for pre-built external toolchains like the Blackfin one, so for
now, we simply define _REENTRANT manually on ARC and Blackfin. If more
packages are affected by this problem, a more general solution will have
to be found.
Fixes
http://autobuild.buildroot.net/results/593/593b5050473a83a9ddcada9de39f8f14ab38d554/http://autobuild.buildroot.net/results/ff0/ff058a378b91740b5b399f32a6b375fbc3c8df06/
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
This patch is based on a patch sent by Vicente Olivert Riera and commented by
Arnout Vandecappelle [1].
- Bump version to 1.23
- Add a hook to fix cross-compilation
- Fix license and license files
- Remove patch applied upstream
- Add a BR2_PACKAGE_LIBGPG_ERROR_ARCH_SUPPORTS variable
- Propagate the dependencies using that variable:
* package/cppcms
* package/crda
* package/gnupg2
- package/gcr
- package/midori
* package/kodi
* package/libaacs
* package/libassuan
* package/libgcrypt
* package/libgpgme
* package/libksba
* package/libmicrohttpd
- package/janus-gateway
- package/kodi
- package/ola
- package/systemd
* package/libssh
* package/libssh2
- package/php-ssh2
* package/netatalk
* package/network-manager
* package/ntfs-3g
* package/opkg
* package/php-gnupg
* package/rng-tools
* package/strongswan
* package/vpnc
[1] http://patchwork.ozlabs.org/patch/416427/
Cc: Arnout Vandecappelle <arnout@mind.be>
Cc: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Signed-off-by: Jörg Krause <joerg.krause@embedded.rocks>
[Thomas:
- rebase on master
- changing systemd no longer needed, as it no longer selects
libgcrypt.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
[Maxime:
- rebase on master
- bump to new version
- propagate dependencies to missing packages]
Signed-off-by: Maxime Hadjinlian <maxime.hadjinlian@gmail.com>
Reviewed-by: Romain Naour <romain.naour@gmail.com>
[Thomas:
- fix hash file.
- change the way to handle the various arch so that it works properly
for uClibc.
- add nios2 arch support.
- Maxime Hadjinlian learned some basic Emacs-fu to do the final fixups
of this commit.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
The _gp link issue has been fixed in CS nios2 2015.11.
Signed-off-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
See the conclusion about external toolchains during the Buildroot
meeting [1]:
"In the future, we stick to a single external toolchain version. The
Kconfig symbol should not encode the version (avoid legacy handling)"
[1] http://elinux.org/index.php?title=Buildroot:DeveloperDaysELCE2015#Report
Signed-off-by: Romain Naour <romain.naour@openwide.fr>
Reviewed-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Tested-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Experimental SPDY support was dropped so drop the configure option.
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fixes
http://autobuild.buildroot.net/results/76f/76f838b2b33311897f3c2ce82a65f3b73af2c046/
Propagate reverse dependency to janus-gateway. I did not propagate the
reverse dependency to kodi, ola and systemd because they are not
available for nios.
Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
It is often difficult to know exactly when make will expand the
variable, and usually it can only be expanded after the dependencies
have been built (e.g. pkg-config or the .pc file). Using a backtick
instead makes it very clear that it will be expanded only while
executing the command.
This change is useful for two cases:
1. The per-package staging (and host) directory will be created as part
of the configure step, so any $(shell ...) variable that is used in
the configure step will fail because the directory doesn't exist
yet.
2. 'make printvars' evaluates the variables it prints. It will therefore
trigger a lot of errors from missing .pc files and others. The
backticks, on the other hand, are not expanded, so with this change
the output of 'make printvars' becomes clean again.
This commit contains only the easy changes: replace $(shell ...) with
`...`, and also replace ' with " where needed. Follow-up commits will
tackle the more complicated cases that need additional explanation.
After this change, the following instances of $(shell ...) will remain:
- All assignments that use :=
- All variables that are used in make conditionals (which don't expand
the backticks).
- All variables that only refer to system executables and make
variables that don't change.
- The calls to check-host-* in dependencies.mk, because it is eval'ed.
[Original patch by Fabio Porcedda, but extended quite a bit by Arnout.]
Signed-off-by: Fabio Porcedda <fabio.porcedda@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Bump to new version, and disable building examples.
Signed-off-by: Simon Marchi <simon.marchi@polymtl.ca>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
To be consistent with the recent change of FOO_MAKE_OPT into FOO_MAKE_OPTS,
make the same change for FOO_CONF_OPT.
Sed command used:
find * -type f | xargs sed -i 's#_CONF_OPT\>#&S#g'
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
In the Config.in file of package foo, it often happens that there are other
symbols besides BR2_PACKAGE_FOO. Typically, these symbols only make sense
when foo itself is enabled. There are two ways to express this: with
depends on BR2_PACKAGE_FOO
in each extra symbol, or with
if BR2_PACKAGE_FOO
...
endif
around the entire set of extra symbols.
The if/endif approach avoids the repetition of 'depends on' statements on
multiple symbols, so this is clearly preferred. But even when there is only
one extra symbol, if/endif is a more logical choice:
- it is future-proof for when extra symbols are added
- it allows to have just one strategy instead of two (less confusion)
This patch modifies the Config.in files accordingly.
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>