If libcap happens to be present in the environment, coreutils will pick it up
and link some applets against it.
Since the idea of coreutils is to provide a full featured alternative to the
busybox applets, do not inhibit the optional dependency but explicitely
require libcap instead.
Fixes the following error spotted on the buildbots:
Package coreutils-dir is missing dependencies for the following libraries:
libcap.so.2
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
This adds the beep utility, which allows users to control a pc speaker device.
Maintainer: me / @riptidewave93
Compile tested: x86/64, LEDE Reboot SNAPSHOT r2709-b7677f0
Run tested: x86/64, LEDE Reboot SNAPSHOT r2709-b7677f0. Verify program works.
Signed-off-by: Chris Blake <chrisrblake93@gmail.com>
Bump the version of gnupg and use new PKG_HASH for
sha256sum and PKG_MD5SUM for md5sum (for backwards
compatibility).
Signed-off-by: Daniel Dickinson <lede@cshore.thecshore.com>
@Nico seems MIA; @hnyman asked if I'd be interested
in taking over. I'm willing so adding myself.
Signed-off-by: Daniel Dickinson <lede@cshore.thecshore.com>
gpgv is a stripped down gnupg useful for only verifying signatures.
Having this package can save space when all you need is signature
verification.
Signed-off-by: Daniel Dickinson <lede@cshore.thecshore.com>
With a LuCI app (of which I have one written) ucification makes
sense (and is in fact needed), so ucify the initscripts.
Also, rather than making selection of things to include an image
a matter of selecting compile-time config options, make optional
things into seperate packages that are built in default builds,
and leave selection of what to include or not up to the user
(e.g. using ImageBuilder, or adding packages via opkg).
Signed-off-by: Daniel Dickinson <lede@cshore.thecshore.com>
This fixes an alignment issue on some platforms, which renders
some tools, e.g. plcID, unusable.
While at, switch to xz compression for the tarball, drop an already
mainlined patch and remove a meanswhile absolete compiler switch.
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
As both LEDE and OpenWrt have STAGING_DIR_HOSTPKG now, we can start to rely
on it. See 73b7f55424 for more information on
STAGING_DIR_HOSTPKG.
STAGING_DIR_HOSTPKG won't actually be changed before the first LEDE release
(it is equivalent to $(STAGING_DIR)/host), so this simple search/replace
cleanup is safe to apply. Doing this cleanup now will be useful for the
Gluon project (an OpenWrt/LEDE based firmware framework) for experimenting
with modifying STAGING_DIR_HOSTPKG before doing this in the LEDE upstream.
Also fixes a typo in the dbus Makefile ("STAGIND_DIR").
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Add Sourceforge as primary download site and main site as secondary.
Drop fedoraproject.org repo
Signed-off-by: Daniel Engberg <daniel.engberg.lists@pyret.net>
Addition of sqlite storage
Tracking of SMTP traffic instead of P2P traffic
Based on NethServer bandwidthd implementation
Signed-off-by: Jean-Michel lacroix <lacroix@lepine-lacroix.info>
Should be PKG_BUILD_DEPENDS not BUILD_DEPENDS
Signed-off-by: Daniel Dickinson <lede@daniel.thecshore.com>
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
At the moment, only build these softmmu emulators for x86_64 and sunxi
target. The decision was made for the following reasons
- It seems that interests of virtualization with qemu are mostly from
x86, and ARM recently.
- x86, sunxi boards/boxes capable of running qemu with accel=kvm are more
widely available
- Not all host, target combinations of qemu works, or even compiles
- Extra maintenance work and server resources
Test results are as the following
- Nested vmx works: lede-qemu-x86_64-kvm on lede-qemu-x86_64-kvm
- KVM on Cubieboard2 works
- tcg with malta works: lede-qemu-malta-tcg on lede-qemu-malta-tcg.
But it's too slow to be useful thus not included in this version
- mips64 host does not compile
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
Package zip is currently broken with a runtime error.
When zip is run, it immediately exits with
"zip warning : Not supported (uzoff_t not same size as zoff_t)"
The issue boils down to the package's configure script which tries to
determine LARGE_FILE_SUPPORT on *host* side. The conftest.c is
compiled and ran on the building host to see whether LFS is given or not.
This will fail when cross-compiling. The patch here is created by Romain Naour,
taken from http://lists.busybox.net/pipermail/buildroot/2015-January/117909.html
Reworked and tested by me. Now LFS is detected and the built binaries work
on the target.
Signed-off-by: Paul Wassi <p.wassi@gmx.at>
Enable cpufreq plugin also for ipq806x,
as IPQ8065 supports CPU frequency scaling.
Run-tested with Netgear R7800.
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
- typo in path of uci-defaults scripts
- fix init script when there is no rngd config
- convert init script to procd
Signed-off-by: Stefan Hellermann <stefan@the2masters.de>
Enabling the default configuration will pipe /dev/urandom back into
/dev/random ala the current package behavior. Because this amounts to
disabling the in-kernel entropy estimation, default disabled.
While here, uci-paramaterize the high watermark.
While here, add a pre-command hook for real RNGs that need 'stty raw'
or other such hooks. (e.g. the TrueRNG devices)
As offered in github openwrt/packages#3422, take over maintainership.
See github openwrt/packages#3142.
Signed-off-by: Nathaniel Wesley Filardo <nwfilardo@gmail.com>
Add more dependencies needed by download template and replace all, not just
first occurence of : or _ in template name with space in webui.
Signed-off-by: Michal Hrušecký <Michal.Hrusecky@nic.cz>
Moved source folder in /sound folder, and changed all packages in the makefile
to be shown in Sound category (not in Sound subcategory of Utilities).
Removed a line calling a non-existent Package/alsa-utils/Default
and site URL to the package alsa-utils-tests.
Signed-off-by: Alberto Bursi <alberto.bursi@outlook.it>
Like on other common desktop Linux distributions, ship an /sbin/mount.ntfs
symlink in addition to the /sbin/mount.ntfs-3g one in order to let wrapper
programs successfully call external mount helpers by the filesystem name.
The assumption is that /sbin/mount.ntfs is only ever called by util-linux
mount and other mount-wrappers when kernel NTFS support is not available,
means shipping the additional symlink will not interfere with kernel mode
NTFS support.
This commit is mainly intended to prepare transparent fs-tools support for
mounting fuse filesystems, with focus on ntfs-3g in particular.
Please see http://git.lede-project.org/f027c68 for the corresponding
fs-tools support code.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
I needed also sgdisk (script-friendly) and while I was at it I added also
cgdisk (uses ncurses interface), all as separate packages.
re-arranged the package folder and set /usr/sbin for install folder
as most tools seem to go in there and not in /sbin.
Signed-off-by: Alberto Bursi <alberto.bursi@outlook.it>
- Merge mkfs.fat/fsck.fat/fatlabel into a single package.
- Remove compatibility symlinks.
- Take over maintainership.
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
This updates dosfstools to the latest release available and uses the new
automatically generated Makefiles instead of a hard coded compilation.
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
This change enables building of the built-in cpufreq module within Collectd, which is very useful on x86 and mvebu targets that support CPU frequency scaling. Note that luci-app-statistics currently does not have support for rendering this.
Signed-off-by: Chris Blake <chrisrblake93 at gmail.com>
Package maintainer reported off-list, that the package
also works with php7. So lets migrate the dependencies
to prepare the php5 removal.
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
In SConstruct the set of executables in the devenv variable should be adjusted to use the cross compile toolchain (as opposed to host's executables).
Achieved by setting target option to the toolchain prefix which corrects compilation on macOS.
Note, TARGET_CROSS has to have the hyphen stripped as SConstruct uses the target variable with an implied hyphen (env['target'] + '-' + toolname)
Signed-off-by: David Thornley <david.thornley@touchstargroup.com>
This commit adds a patch to remove a superfluous include.
Removing this include prevents the following error (detected by LEDE
project's buildbot while compiling for mips64 architecture):
(Note: directory names shorted for better readability)
In file included from mmc.h:20:0,
from mmc_cmds.c:37:
/srv/.../include/asm-generic/int-ll64.h:29:44: error: conflicting types for '__s64'
__extension__ typedef __signed__ long long __s64;
^
In file included from /srv/.../include/asm/types.h:22:0,
from /srv/.../include/linux/types.h:4,
from /srv/.../include/linux/fs.h:11,
from mmc_cmds.c:35:
/srv/.../include/asm-generic/int-l64.h:28:25: note: previous declaration of '__s64' was here
typedef __signed__ long __s64;
^
In file included from mmc.h:20:0,
from mmc_cmds.c:37:
/srv/.../include/asm-generic/int-ll64.h:30:42: error: conflicting types for '__u64'
__extension__ typedef unsigned long long __u64;
^
In file included from /srv/.../include/asm/types.h:22:0,
from /srv/.../include/linux/types.h:4,
from /srv/.../include/linux/fs.h:11,
from mmc_cmds.c:35:
/srv/.../include/asm-generic/int-l64.h:29:23: note: previous declaration of '__u64' was here
typedef unsigned long __u64;
^
Makefile:36: recipe for target 'mmc_cmds.o' failed
The patch was already sent upstream, no response yet.
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
* Add description to both Makefile and init script that
mentions the default of /dev/urandom as the entropy source.
* Remove the deprecated interval parameter from init script.
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
This package installs a couple of helper programs for the Flent testing
tool. These are useful when running tests and gathering remote data from
an OpenWrt/LEDE router.
Signed-off-by: Toke Høiland-Jørgensen <toke@toke.dk>
---
config.c: In function 'strndup':
config.c:87:10: error: nonnull argument 's' compared to NULL [-Werror=nonnull-compare]
if(!s)
^
cc1: all warnings being treated as errors
---
Signed-off-by: Hirokazu MORIKAWA <morikw2@gmail.com>
slide-switch translates slide switch position changes into normal button
presses and releases for OpenWrt / LEDE devices.
Signed-off-by: Jeffery To <jeffery.to@gmail.com>
Maintainer: @nwf
Compile tested: LEDE HEAD ar71xx
Run tested: LEDE HEAD on ar71xx, basic functionality testing
Description: A minimal kermit file transfer program
Signed-off-by: Nathaniel Wesley Filardo <nwfilardo@gmail.com>
This patch fixes the watchcat init script in case of mode 'always'.
This mode just needs the config parameters: mode, period and forcedelay.
But a wrong check against the variable pingperiod, needed in the mode 'ping'
prevented mode 'always' to work properly.
Signed-off-by: Tobias Ilte <tobias.ilte@campus.tu-berlin.de>
Enable tail_csv plugin that "follows" (tails) and parses CSV files.
Compile-tested with ar71xx.
This closes#3083
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
This has two effects
- Saves time and computing resources. Non-relevant tools like
qemu-img, qemu-io will not be built
- Fixes packaging failure on octeon target caused by failure of
building internal pixman library
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
Only mention busybox applets for the meta package and do not imply that
applets are enabled for any possible coreutils program.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
database
Signed-off-by: Jean-Michel Lacroix <lacroix@lepine-lacroix.info>
bandwidthd-php is a set of PHP files that allows to graph bandwidthd
data stored in a postgresql database. This bandwidthd data can be
generated using bandwidthd-pgsql in OpenWRT or LEDE or with
bandwidthd on different OS (Windows, Linux for example).
Addition of a config file and an init file.
Modification of the Makefile to take in account the changes above
Signed-off-by: Jean-Michel Lacroix <lacroix@lepine-lacroix.info>
Since 3048ebedec added a autoreconfig fixup to
procps-ng, the build fails on certain systems with errors like
configure.ac:72: error: possibly undefined macro: AM_GNU_GETTEXT_VERSION
or
configure:14998: error: possibly undefined macro: AM_ICONV_LINKFLAGS_BODY
Fix the issue by adding build time dependencies on the `gettext` and `libiconv`
stub libraries to ensure that `gettext.m4` and `iconv.m4` are installed into
the staging directory before `procps-ng` is getting reconfigured.
Fixes https://github.com/openwrt/packages/issues/2890#issuecomment-231115844
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
The license BSD-3-Clause.
Indicate license file.
Use @SF for PKG_SOURCE_URL instead of sourceforge.net URL.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
OpenWRT/LEDE only understands the PKG_MD5SUM variable, and detects if the
hash is e.g. SHA256 by looking at the length of the hash.
This affects libs/libmicrohttpd, mail/ssmtp and utils/mc.
Signed-off-by: Baptiste Jonglez <git@bitsofnetworks.org>
Standard assignment is immediate expansion without any extraneous spacing, RFC822 compliant email addresses without any quotation marks and long git hashes
Signed-off-by: Stephen Walker <stephendwalker+github@gmail.com>
Fix missing `PKG_SOURCE` variable leading to a
`Download/default is missing the FILE field` error in the buildroot.
Also remove now uneeded variable as the default one matches the git clone
structure. Finally use the common variable assignment style for
`PKG_SOURCE_PROTO` and `PKG_SOURCE_VERSION`.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
While normally busybox version of findutils is enough, sometimes you want the full deal.
Here it is.
Signed-off-by: Daniel Dickinson <openwrt@daniel.thecshore.com>
Package relayctl provides a command-line tool to control
SainSmart USB relay boards based on FTDI chips using bitbang
mode.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
- Use a temporary file path within the package build directory to avoid
cluttering the host systems `/tmp` directory.
- Switch to a gzipped, snapshotted copy of the `oui.txt` file since the
upstream IEEE server is extremely slow, also fetching an unversioned HTTP
resource during compilation breaks reproducable builds.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
[rework to use standard download facility]
Signed-off-by: Etienne CHAMPETIER <champetier.etienne@gmail.com>
The klish is a framework for implementing a CISCO-like CLI on a UNIX
systems. It is configurable by XML files. The KLISH stands for Kommand
Line Interface Shell.
klish is an active fork of the clish program created by Graeme
McKerrell.
Makefile from https://dev.openwrt.org/browser/packages/utils/klish/Makefile?rev=31310
add SOB.
add Maintainer.
change URL http://libcode.org/projects/klish/files
update to 2.1.1 (newest version).
merge klish and konf package.
[squash commits]
Signed-off-by: Etienne CHAMPETIER <champetier.etienne@gmail.com>
Signed-off-by: Takashi Umeno <umeno.takashi@gmail.com>
The current procps-ng Makefile calls the shipped autogen.sh script which
introduces incorrect implicit dependencies on host utilities, leading to
the following error observed on a minimal build system:
(cd .../procps-ng-3.3.11; echo "3.3.11" > ".../procps-ng-3.3.11/.tarball-version"; ./autogen.sh );
You must have autopoint installed to generate procps-ng build system.
The autopoint command is part of the GNU gettext package.
Makefile:96: recipe for target '.../procps-ng-3.3.11/.configured_yynyyyyy' failed
make[3]: *** [.../procps-ng-3.3.11/.configured_yynyyyyy] Error 1
Apply the following changes in order to fix compilation:
- Apply the generic autoreconf fixup to generate configure and Makefiles
- Use Build/Prepare to populate .tarball-version and revert Build/Configure
to its default implementation
- Disable to build of docs and tests as those require additional utilities
not guaranteed to be present
Fixes#2890.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
If we've enable POSIX ACL's and XATTR support as the default, then
make tar build with such support by default as well.
Signed-off-by: Daniel Dickinson <lede@daniel.thecshore.com>
There was harcoded list of available images. Instead of it, LuCI LXC app will
now try to download list of available container images.
Signed-off-by: Michal Hrusecky <michal.hrusecky@nic.cz>
As containers are basically virtual machines, it should not depend on the
build target but on the real hardware architecture. For example there is plenty
of ARM families (mvebu, sunxi, ...) but all armv7l arms should be able to run
armv7l containers.
Signed-off-by: Michal Hrusecky <michal.hrusecky@nic.cz>
I makes quite some sense to provide user the choice between various versions of
the distribution to install into LXC.
Signed-off-by: Michal Hrusecky <michal.hrusecky@nic.cz>
This contains format string fixes among others. procps-ng was not properly
displaying cpu usage in top and ps.
Signed-off-by: Rob Mosher nyt-lede@countercultured.net
Otherwise it will pull all compression libraries to install
(on newly configured targets) when maybe 1 or 2 are needed.
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Adjusting LuCI app for LXC to try to read path to your LXC containers from
configuration file and if not succeeding falling back to /srv/lxc to match
previous commit.
Signed-off-by: Michal Hrusecky <michal.hrusecky@nic.cz>
