CMake 3.0.12 uses SWIG_DIR to set SWIG_LIB, used as the library
install location, which defaults to
$(STAGING_HOSTPKG)/share/sig/(SWIGVERSION). If it is set, then the
original installed swig library directory is ignored, and compilation
fails:
[ 0%] Swig source
:1: Error: Unable to find 'swig.swg'
:3: Error: Unable to find 'python.swg'
Instead of setting it manually, let the default be used, which works
well.
Also, cmake picks up python3, while package wants 2.7, so we add
-DPython_ADDITIONAL_VERSIONS=2.7 to force use of 2.7, not 3.x.
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
CMake 3.0.12 uses SWIG_DIR to set SWIG_LIB, is used as the library
install location, which defaults to
$(STAGING_HOSTPKG)/share/sig/(SWIGVERSION). If it is set, then the
installed swig library directory is ignored, and compilation fails:
[ 32%] Swig source
:1: Error: Unable to find 'swig.swg'
:3: Error: Unable to find 'python.swg'
Instead of setting it manually, let the default be used, which works
well.
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
There seems to be no proper location for this, nor an up to date fork.
Arch uses this link so switch to that.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Applied a patch submitted upstream that fixes a compilation error with
cmake >= 3.12 Error: Unable to find 'swig.swg' [...] 'python.swg'
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
Package versions newer than 2.11 require meson/ninja which OpenWrt does not
support in its build environment. These files provide the minimum autotool
configuration necessary to build later versions.
Signed-off-by: Ted Hess <thess@kitschensync.net>
The eventlog package is no longer required by syslog-ng, and it conflicts
with the files installed by syslog-ng. No other package requires eventlog,
the external eventlog library project has not been updated in a long time,
and syslog-ng is no longer compatible with the external eventlog library.
Signed-off-by: W. Michael Petullo <mike@flyn.org>
Switched to tarball. Should be faster.
Codeload is used as the regular tarball is on the rtorrent github page
with rtorrent's versioning.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Switched to codeload instead of git for downloads. Should be faster and more reliable.
Re-organized the Makefile slightly
Signed-off-by: Rosen Penev <rosenp@gmail.com>
mariadb was sorted out by including nls.mk. Include it also in libzdb
and get rid of the previously introduced hack.
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
They are missed out from the FIXUP check probably because of a flaw in
the fixup-makefile.pl script
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
Error of the following kind can happen when parallel build with goals "all" and
"install" both specified on the command line. Seems like make can parallelly
build for both targets causing race conditions
make[3]: Entering directory `/tmp/tmp.HIR3tEIqBb/build_dir/target-mips_24kc_musl/libsoc-2016-12-22-5b788d4d'
Making all in lib
Making install in lib
make[4]: Entering directory `/tmp/tmp.HIR3tEIqBb/build_dir/target-mips_24kc_musl/libsoc-2016-12-22-5b788d4d/lib'
CC libsoc_la-gpio.lo
make[4]: Entering directory `/tmp/tmp.HIR3tEIqBb/build_dir/target-mips_24kc_musl/libsoc-2016-12-22-5b788d4d/lib'
CC libsoc_la-spi.lo
CC libsoc_la-gpio.lo
CC libsoc_la-file.lo
CC libsoc_la-i2c.lo
CC libsoc_la-spi.lo
CC libsoc_la-pwm.lo
CC libsoc_la-board.lo
CC libsoc_la-file.lo
CC libsoc_la-conffile.lo
CC libsoc_la-debug.lo
CC libsoc_la-pwm.lo
CC libsoc_la-conffile.lo
CCLD libsoc.la
OpenWrt-libtool: link: `libsoc_la-pwm.lo' is not a valid libtool object
make[4]: *** [libsoc.la] Error 1
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
libmariadb 10.2 needs to be linked in together with iconv. On musl and
glibc iconv is part of libc. That's not the case for uclibc, where
libiconv-full needs to be used. This commit adds -liconv to the linker
flags and aids libzdb in finding libiconv.
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
Some of them forgot to update MIRROR_HASH on version change, others
updated with wrong hash value. The new values were generated from
tarballs prepared by the newly introduced github-tarball download
methoded and confirmed consistent with those from sources.openwrt.org
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
This package is needed only to provide libcups.so and
libcups-image.so for hplip. No ipk is generated as
hplip-cups is currently disabled until full cups is provided.
cups was bumped to version 2.2.8
Signed-off-by: Luiz Angelo Daros de Luca <luizluca@gmail.com>
Update to ICU 62 (62.1)
ICU 62 includes number format skeleton support in MessageFormat, ICU4C DecimalFormat wrapping the new NumberFormatter implementation for improved performance.
Signed-off-by: Hirokazu MORIKAWA <morikw2@gmail.com>
In avahi 0.7 the runtime_dir was updated to be /run instead of /var/run
for better systemd compatibility. Patch out that change to fix
avahi-daemon.
Signed-off-by: Jeremiah McConnell <miah@miah.com>
* Version 2.0.5 (released 2018-05-18)
** Switched the default library behavior to IDNA2008 as amended
by TR#46 (non-transitional). That default behavior is enabled when
no flags are specified to function calls. Applications can utilize the
%IDN2_NO_TR46 flag to switch to the unamended IDNA2008. This is done in
the interest of interoperability based on the fact that this is what application
writers care about rather than strict compliance with a particular protocol.
** Fixed memleak in idn2_to_unicode_8zlz().
** Return error (IDN2_ICONV_FAIL) on charset conversion errors.
** Fixed issue with STD3 rules applying in non-transitional
TR46 mode.
** idn2: added option --usestd3asciirules.
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
When cross-compiling, the configure script doesn't use /dev/urandom as
entropy source, which would have been the case if we were building it
natively. Instead it tries to use an EGD, which openwrt doesn't supply.
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
Avoid calling deprecated and unnecessary library initialization
functions with openssl >= 1.1.0.
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
Use $(CP) instead of $(INSTALL) so that libtdb.so.1 is installed as
symlink, and not duplicated.
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
This patch is a backport of upstream changes.
I read the Makefile warning and made sure it only updates the openssl
calls and data types to the new API. It does not touch anything else!
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
Bump to v1.35
Refresh patches.
Update Makefile to use new '--disable-doc' configure flag.
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
Update to new upstream version. Needed to add a patch provided by upstream to compile with musl.
Signed-off-by: David Mora <iamperson347+public@gmail.com>
- Fixed package hyperlink
-> Now using the @SF macro to obtain the best mirror link
-> Added backup link in case Source Forge fails to provide the proper link
- Minor fix to package documentation
-> Help documentation was lacking the contract library info.
Signed-off-by: Carlos Miguel Ferreira <carlosmf.pt@gmail.com>
Switch to the @KERNEL alias
Use LTO optimization, reduces file size by ~20% (~10Kbyte)
With:
39K sbc_1.3-1_arm_cortex-a9_neon.ipk
Without:
51K sbc_1.3-1_arm_cortex-a9_neon.ipk
Signed-off-by: Daniel Engberg <daniel.engberg.lists@pyret.net>
This package update provides two new libraries
- Contract (compiled library) [1]: Contract programming for C++.
All contract programming features are supported: Subcontracting,
class invariants, postconditions (with old and return values),
preconditions, customizable actions on assertion failure (e.g.,
terminate or throw), optional compilation and checking of
assertions, etc, from Lorenzo Caminiti.
- HOF (header-only library) [2]:
Higher-order functions for C++, from Paul Fultz II.
More info can be found at the usual place [3].
[1]: https://www.boost.org/libs/contract
[2]: https://www.boost.org/libs/hof
[3]: https://www.boost.org/users/history/version_1_67_0.html
Signed-off-by: Carlos Miguel Ferreira <carlosmf.pt@gmail.com>
Add libidn2 & idn2 tool to repo
Heavily based on PR by Daniel Engberg <daniel.engberg.lists@pyret.net>
Added idn2 tool & heavily based on existing idn Makefile.
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
Required kernel interface was introduced in kernel v4.8,
so add the kernel version as dependency for this library.
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
Maintainer: me
Compile tested: ar71xx, LEDE trunk r5488-893a1ed
Run tested: NONE
Description:
ICU 61 upgrades to CLDR 33 locale data, has a new Java implementation
for number and currency parsing, and includes many small API additions,
improvements, and bug fixes.
(includes security fix CVE-2017-17484,CVE-2017-15422)
Signed-off-by: Hirokazu MORIKAWA <morikw2@gmail.com>
The FreeDesktop link does not work anymore. Switch to new upstream at GitHub.
Update the version while we're at it.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Upstream commit "gio: Port GThreadedResolver to use res_nquery() to fix
thread-safety" introduced a bug that causes compile failures on
platforms where res_nquery is unavailable. This patch fixes that.
Bug 794606 has been opened upstream: https://bugzilla.gnome.org/show_bug.cgi?id=794606
On the OpenWrt buildbots this bug is visible during the arc_archs
compile:
gthreadedresolver.c: In function 'do_lookup_records':
gthreadedresolver.c:875:16: error: 'res' undeclared (first use in this function); did you mean '_res'?
res_nclose (&res);
^~~
_res
gthreadedresolver.c:875:16: note: each undeclared identifier is reported only once for each function it appears in
Makefile:3334: recipe for target 'libgio_2_0_la-gthreadedresolver.lo' failed
make[8]: *** [libgio_2_0_la-gthreadedresolver.lo] Error 1
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
When building for ARC the compile fails with:
/var/lib/buildbot/slaves/dave-builder/arc_arc700/build/sdk/build_dir/target-arc_arc700_uClibc/postgresql-9.6.5/src/include/storage/s_lock.h:899:2: error: #error PostgreSQL does not have native spinlock support on this platform. To continue the compilation, rerun configure using --disable-spinlocks. However, performance will be poor. Please report this to pgsql-bugs@postgresql.org.
#error PostgreSQL does not have native spinlock support on this platform. To continue the compilation, rerun configure using --disable-spinlocks. However, performance will be poor. Please report this to pgsql-bugs@postgresql.org.
So disable spinlocks when compiling for this arch. This was done
likewise for the avr32 target, which is not supported anymore, so this
can be deleted.
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
- The extra include is removed; the postgresql build system adds this
include by itself.
- '-lpthread' is removed; the postgresql build system is able to detect
how to access pthread features by itself.
- '$(FPIC)' is removed; adding this to CFLAGS is not recommendable:
a) The postgresql build system adds its own PIC flag behind OpenWrt's
CFLAGS when compiling shared objects, overriding OpenWrt's setting.
b) Forcing applications into PIC mode will just slow them down.
c) This leads to build failures on aarch64:
access/transam/commit_ts.o: In function `CommitTsPagePrecedes':
commit_ts.c:(.text+0x14): relocation truncated to fit: R_AARCH64_LD64_GOTPAGE_LO15 against symbol `TransactionIdPrecedes' defined in .text section in access/transam/transam.o
commit_ts.c:(.text+0x14): warning: Too many GOT entries for -fpic, please recompile with -fPIC
/var/lib/buildbot/slaves/dave-builder/aarch64_generic/build/sdk/staging_dir/toolchain-aarch64_generic_gcc-7.3.0_musl/bin/../lib/gcc/aarch64-openwrt-linux-musl/7.3.0/../../../../aarch64-openwrt-linux-musl/bin/ld: final link failed: Symbol needs debug section which does not exist
collect2: error: ld returned 1 exit status
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
* add option to install additional executables - capsh, getcap, getpcaps, setcap
* capsh can be configured to execute a different shell rather than /bin/bash
Signed-off-by: Toni Uhlig <matzeton@googlemail.com>
This reverts 5b5659850d.
In hindsight I have to admit I did not correctly understand the
implications of the uclibc++.mk include.
The include allows a package to follow the user's choice regarding which
C++ library should be the standard. Linking against uClibc++ instead of
libstd++ is not a problem when running musl (which is what I had
incorrectly assumed), as both C++ libs are separate packages. And
uClibc++ is a lot smaller than libstd++, which is probably why it is
even the default C++ lib on OpenWrt currently.
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
Provide configuration items to select individual backends. This
also allows to fine-tune the package dependencies.
In order to address #5637, we make this dep unconditionally
until upstream allow to opt-in/opt-out libavahi support.
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
Recently added symbols CONFIG_LIBSSH2_MBEDTLS and CONFIG_LIBSSH2_OPENSSL
require a rerun of ./configure when their selection changes. So add them
to PKG_CONFIG_DEPENDS accordingly.
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
The change should also work fine with older kernels, as <linux/uaccess.h>
has existed for a long time, and it includes <asm/uaccess.h>.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Make sure that the c++ headers get installed to the staging directory
when libtiffxx was selected/built.
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
libxslt already makes sure that the compiler emits position-independent
code. Adding $(FPIC) makes no difference. In addition the libxslt build
system puts its own flag behind the CFLAGS, overriding whatever was set
before.
Also with $(FPIC) in CFLAGS even the libxslt utils get compiled with
PIC, which us undesirable.
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
