While running `make menuconfig`, it was discovered then there is a
recursive dependency like this:
tmp/.config-package.in:59138:error: recursive dependency detected!
tmp/.config-package.in:59138: symbol PACKAGE_libwebsockets-openssl is selected by PACKAGE_libwebsockets-mbedtls
tmp/.config-package.in:59122: symbol PACKAGE_libwebsockets-mbedtls depends on PACKAGE_libwebsockets-openssl
It is not possible with the recently added conflicts that two packages
(OpenSSL and full variant, which uses OpenSSL as well), which are almost the same
provides the same named package libwebsockets as their conflict - Mbed
TLS.
Fixes: 676c5c72b5 ("libwebsockets: OpenSSL
and mbedTLS variants should conflict")
Signed-off-by: Josef Schlehofer <pepe.schlehofer@gmail.com>
(cherry picked from commit a4e8cbb89a)
They provide the same files, but they don't conflict to each other, this
means that users can install them side by side.
Signed-off-by: Josef Schlehofer <pepe.schlehofer@gmail.com>
(cherry picked from commit 676c5c72b5)
For some time, it is not possible to install ttyd and mosquitto-ssl at the
same time, so let's solve it that libwebsockets-full provides
libwebsockets-openssl. This allows to install ttyd and mosquitto at
the same time.
Also, we need to add conflict, because we should not have installed
libwebsockets-openssl and libwebsockets-full at the same time as they
provides the same files.
Signed-off-by: Josef Schlehofer <pepe.schlehofer@gmail.com>
(cherry picked from commit 77e682a11c)
Fixes security issues:
* CVE-2021-25220 -- The rules for acceptance of records into the cache
have been tightened to prevent the possibility of
poisoning if forwarders send records outside
the configured bailiwick.
* CVE-2021-25219 -- The "lame-ttl" option is now forcibly set to 0. This
effectively disables the lame server cache, as it could
previously be abused by an attacker to significantly
degrade resolver performance.
Signed-off-by: Noah Meyerhans <frodo@morgul.net>
The FreeBSD project stopped publishing HTTP date headers and seeks to
limit further resource taxing by distributed htpdate clients using the
www.freebsd.org host as default time source.
Fixes: #17924
Reported-by: Allan Jude <allanjude@freebsd.org>
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit e871318002)
The following security issues are addressed with this change:
CVE-2020-8619
CVE-2020-8622
CVE-2020-8623
CVE-2020-8624
CVE-2020-8625
CVE-2021-25214
CVE-2021-25215
CVE-2021-25216
CVE-2021-25219
A complete description of the changes with this BIND release is
available in the release notes at
https://ftp.isc.org/isc/bind9/9.11.36/RELEASE-NOTES-bind-9.11.36.html
Signed-off-by: Noah Meyerhans <frodo@morgul.net>
When installing a host perl module, the host perl binary in the staging
dir is replaced by using 'cp'. However, if the binary is running in a
parallel job, cp will fail with a text file busy error. Use
$(INSTALL_BIN), which unliks the file first to avoid the error.
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
(cherry picked from commit 19c7496648)
(cherry picked from commit 67f403b5e6)
By default, there was used sks-keyservers.net pool, which has invalid
SSL certificate and they also announced that their service is deprecate
and no longer maintained.
Use the same GPG server as LXC is using by default in the newer
releases.
Signed-off-by: Josef Schlehofer <pepe.schlehofer@gmail.com>
Fixes two CVEs:
CVE-2020-25613: Potential HTTP Request Smuggling Vulnerability in WEBrick
CVE-2021-28965: XML round-trip vulnerability in REXML
After this release, Ruby 2.5 reaches EOL.
Signed-off-by: Luiz Angelo Daros de Luca <luizluca@gmail.com>
Includes fixes for:
* CVE-2021-3177 - ctypes: Buffer overflow in PyCArg_repr
* CVE-2021-23336 - urllib parse_qsl(): Web cache poisoning - semicolon
as a query args separator
Signed-off-by: Jeffery To <jeffery.to@gmail.com>
It was provided by both the freeradius3 and freeradius3-mod-realm packages.
Now provided by the freeradius3 package only.
Signed-off-by: Robby K <robbyke@gmail.com>
Fixes: #13149
(cherry picked from commit c81d176e80)
command-count.h generated by makefile was wrong
when using default shell in mac,
set shell to bash to fix it.
Signed-off-by: Liangbin Lian <jjm2473@gmail.com>
(cherry picked from commit 9bb0962d6e)
Config files
/etc/freeradius3/policy.d/accounting
/etc/freeradius3/policy.d/filter
/etc/freeradius3/proxy.conf
/etc/freeradius3/sites-available/default
and link
/etc/freeradius3/sites-enabled/default
are in the freeradius3 package and are mentioned in the main config file
/etc/freeradius3/radiusd.conf
Thus, they must be explicitly specified in the Makefile.
File
/etc/freeradius3/sites/default
is not included in the package, is not created during installation,
is not mentioned in the main config file and should therefore be excluded
from the Makefile.
(backported from commit f6974b8)
Signed-off-by: Alexey Dobrovolsky <dobrovolskiy.alexey@gmail.com>
* samba4 complains that it cant find the libasn1 host bins
Signed-off-by: Andy Walsh <andy.walsh44+github@gmail.com>
(cherry picked from commit 42e9057d41)
This backports fixes for security issues, including:
* CVE-2020-14422: Hash collisions in IPv4Interface and IPv6Interface
* CVE-2019-20907: Infinite loop in the tarfile module
Signed-off-by: Jeffery To <jeffery.to@gmail.com>
This is intended as a match with standard urngd. They serve same purpose
and urngd starts as first with 00. Starting haveged later can create
issues if you replace urngd with it. The example problem is if
uci-defaults script decides to generate certificate. Haveged can supply
entropy but it is started later and to mitigate this urngd would still
have to be installed. This means that haveget can't serve as replacement
without moving it to match start order of urngd.
Signed-off-by: Karel Kočí <karel.koci@nic.cz>
(cherry picked from commit 1ef38f45d9)
The ssl module assumes OpenSSL can load the default trust anchors (root
CA certificates).
From https://github.com/openwrt/packages/issues/12209
Signed-off-by: Jeffery To <jeffery.to@gmail.com>
Add -rpath linker option to host build, pointing to staging/hostpkh/lib.
It's needed to find the correct host libs during runtime, without it the
hosts libs may be used instaead, causing failures.
Signed-off-by: Jan Kardell <jan.kardell@telliq.com>
(cherry picked from commit 0311e58bb6)
The linker option -rpath is required to find libs in staging_dir. Now it
is included when building host modules. Without it the import test of
the _ctypes and _uuid modules would fail. The _ctypes module uses
libffi.so.6 from staging, but OpenSUSE LEAP 15 has libffi.so.7.
It will also fail on LEAP 42.x, Fedora28 and 29 and future or old
versions of Ubuntu.
Fix needed in master and 18.06 branches.
Signed-off-by: Jan Kardell <jan.kardell@telliq.com>
(cherry picked from commit 6ade5a1e3a)
Fixes:
CVE-2020-2752
CVE-2020-2812
CVE-2020-2814
This commit also moves mysql_upgrade to the client package and installs
the configuration files readable for all, so that the clients can read
them.
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
Sometimes the return value of `ubus -S call network.interface.wan status`
cause `json_load` to return `Failed to parse message data` error.
To avoid this, the JSON data always should be quoted with double quotes.
Signed-off-by: Evren Yurtesen <eyurtese@abo.fi>
Removed quoatation marks from commit heading
Signed-off-by: Florian Eckert <fe@dev.tdt.de>
(cherry picked from commit 94e0c78826)
Switch to normal tarballs. Remove autoreconf as a result.
Several Makefile cleanups for consistency.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry picked from commit c22dd1bf74)
(cherry picked from commit 2085b990ff)
When during the build the openssl extension is also selected, then
the mysqlnd extension depends on it, too.
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
(cherry picked from commit 50b1cd3757)
This is to address the need for specifying dependency on a wget
implementation with ssl support.
Now we have a game of names for opkg
1. uclient-fetch: minimal version by openwrt project
2. wget-nossl: gnu wget w/o ssl support
3. wget-ssl: for the moment since this commit, gnu wget w/ ssl support
4. wget: uclient-fetch, wget-nossl, or wget-ssl
5. gnu-wget: wget-nossl or wget-ssl
By the time we provide some dummy package like uclient-fetch-ssl and
make it also provide wget-ssl, I guess by then we will also need
gnu-wget-ssl...
Ref: https://github.com/openwrt/packages/issues/11534
Ref: https://github.com/openwrt/packages/pull/9941
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
(cherry picked from commit 585e4a38fe &
resolved conflict)
The package wget should not say that it provides itself.
This also make gnu-wget provide general so it is not written in Makefile
twice.
Signed-off-by: Karel Kočí <karel.koci@nic.cz>
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
(cherry picked from commit 75f2be7d50 &
resolved conflict)
e921ca0a Add further commonly used protocols to the protocol file.
Signed-off-by: DENG Qingfang <dengqf6@mail2.sysu.edu.cn>
(cherry picked from commit 5e140462e4)
Not all interfaces may have been allocated address at nlbwmon startup so
it may not collect statistics as expected/configured.
Add interface triggers to catch dhcp events and restart as required.
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
(cherry picked from commit 662d3f16d2)
Even on a powerful platform a nlbwmon process'
activities are sometimes affecting throoughput and
latency. This is a backgroud process, that should not
be running with default priority.
Even if it is a little deplayed, that is not a worry in
this case. The routing should be the main priority,
bandwidth stats collection can wait a bit.
Tested on Netgear R7800
Signed-off-by: Marc Benoit <marcb62185@gmail.com>
(cherry picked from commit 9b3ecbd64a)
- It brings oauthbearer support
- It switches from OpenSSL to GnuTLS, which is fully compatible and
brings more parameters
Makefile changes:
- Fix SPDX License Identifier
- Add PKG_CPE_ID
- The Project does not use sourceforce anymore.
- Uses GnuTLS instead of OpenSSL, which was discouraged
Signed-off-by: Josef Schlehofer <pepe.schlehofer@gmail.com>
Update nano editor to version 4.9.1
2020.03.31 - GNU nano 4.9.1 "Sapperdeflap"
Two bugs introduced in version 4.9 are fixed: the cursor
getting misplaced when undoing line cuts, and filtering
of the whole buffer to a new buffer not working.
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
(cherry picked from commit f5c47a2606)
FIX CVE-2020-10531
An integer overflow, leading to a heap-based buffer overflow, exists in the UnicodeString::doAppend() function in common/unistr.cpp.
Signed-off-by: Hirokazu MORIKAWA <morikw2@gmail.com>
- Use HTTPS protocol for updating urls for afraid.org to protect the password.
- Bump/align package version number.
Signed-off-by: Kwonjin Jeong <gram25gwh@gmail.com>
(cherry-picked from e25b558999)
This fixes how GOARM is selected for arm platforms, based on support for
VFP/VFPv3 rather than CPU version.
Fixes#10967.
Signed-off-by: Jeffery To <jeffery.to@gmail.com>
It seems semantics changed slightly.
Cleaned up Makefile for consistency between packages.
Added PKG_BUILD_PARALLEL for faster compilation.
Fixed license information.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry-picked from 16fb706641)
Updates vapigen-generated vapi files to vala-0.34.
Files are now copied to $(STAGING_DIR_HOSTPKG)/share/vala/vapi, without
the vala version being hardcoded.
Removed copyright line, and redundant PKG_BUILD_DEPENDS and
TARGET_LDFLAGS.
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
(cherry-picked from 1b794e93e3)
Musl only specifies in/outb for x86/x86. Use the fallback path in case
musl is used.
This should fail compilation during the linking stage but for some reason
does not. Will do if -Werror=implicit-function-declaration is specified.
Fixed up license information.
Other small cleanups.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry-picked from fa40b85f20)
Add patch to fix compile issue:
stats.c: In function 'setcounter':
stats.c:300:29: error: 'val' undeclared (first use in this function); did you mean 'value'?
stats->counters[counter] = val;
^~~
value
Signed-off-by: Josef Schlehofer <pepe.schlehofer@gmail.com>
Commit c1aa1f784c which backported
a number of fixes from master, missed one fix for nut-cgi. If
nut-cgi is installed standalone or nut-cgi starts before nut-server
(the default in 18.04.5) then the configuration file directory does
not get create and therefore there is no CGI configuration file.
This commit fixes that.
Closes: #10687
Signed-off-by: Daniel F. Dickinson <cshored@thecshore.com>
Fixes to init-script were not cherry-picked properly, sync it with
version found in master branch.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
(cherry-picked from a79d6df674)
Removed unused flags, and added --enable-compile-warnings=minimum to
avoid a build failure with fortify-headers:
In file included from [staging_dir/target]/usr/include/libxml2/libxml/tree.h:15:0,
from [staging_dir/target]/usr/include/libxml2/libxml/parser.h:16,
from grl-jamendo.c:32:
[staging_dir/toolchain]/include/fortify/stdio.h: In function 'snprintf':
[staging_dir/toolchain]/include/fortify/stdio.h:99:2: error: format not
a string literal, argument types not checked [-Werror=format-nonliteral]
return __orig_snprintf(__s, __n, __f, __builtin_va_arg_pack());
^~~~~~
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
gnome-common.m4 and gnome-compiler-flags.m4 were taken from version
3.18.0, and match current master branch.
Disable building help and test directories.
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
(cherry picked from commit a87108fe40)
Some firewalls mandate a minimum size of 4k for SYN packets, which
transmission does not do by default. Upstream issue here:
https://github.com/transmission/transmission/issues/964
Cleanup:
Fixed license info.
Removed two unnecessary patches.
Ran shell script through shellcheck.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry-picked from 730a1697fe)
PHP7 fails to load xmlreader.so (php7-mod-xmlreader) module without
dom.so (php7-mod-dom) module loaded:
-snip-
PHP Warning: PHP Startup: Unable to load dynamic library 'xmlreader.so'
(tried: /usr/lib/php/xmlreader.so (Error relocating /usr/lib/php/xmlreader.so:
dom_node_class_entry: symbol not found), /usr/lib/php/xmlreader.so.so (Error
loading shared library /usr/lib/php/xmlreader.so.so: No such file or
directory)) in Unknown on line 0
^C
-snap-
However, this dependency only exists when during build also php7-mod-dom
is selected.
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
(cherry picked from commit b8c22fc1ec)
While at, update the SPDX license id to most recent format.
Compile and run tested on mxs platform.
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
(cherry picked from commit 5805da860e)
This fixes CVE-2019-11042 and CVE-2019-11041.
Compile and run tested on mxs platform
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
(cherry picked from commit 8e419c6d4c)
All symbols on MacOS are prefixed with an underscore which
interfered with the filtering mechanism (added in perl 5.28)
for extension libraries to be linked into static perl.
Signed-off-by: Jakub Piotr Cłapa <jpc@loee.pl>
HAVE_ICONV and HAVE_ICONV_H are two different headers that both need to
evaluate to false. Added the extra CONFIGURE_VARS.
This can be verified by passing -Werror=implicit-function-declaration
Added PKG_LICENSE_FILES
Updated homepage URL.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry-picked from bf2f1a0263)
So that packages like acme requiring features from it can depend on it
explicitly, not the more basic "wget" which is also provided by
"uclient-fetch"
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
[port to 18.06 to fix acme package backport]
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
There are quite a few bugfixes in the version of the ACME package in
master, and the old version in 18.06 have some issues as seen in #10328.
This commit ports over all changes from the master branch in one go.
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
2.5.7 fixes:
* CVE-2019-16255: A code injection vulnerability of Shell#[] and Shell#test
* CVE-2019-16254: HTTP response splitting in WEBrick (Additional fix)
* CVE-2019-15845: A NUL injection vulnerability of File.fnmatch and File.fnmatch?
* CVE-2019-16201: Regular Expression Denial of Service vulnerability of WEBrick’s Digest access authentication
2.5.6 fixes:
* Multiple jQuery vulnerabilities in RDoc
* About 40 bugs
Changelog: https://github.com/ruby/ruby/compare/v2_5_5...v2_5_7
Signed-off-by: Luiz Angelo Daros de Luca <luizluca@gmail.com>
Patches already merged and so removed:
* 019-bpo-36216-Add-check-for-characters-in-netloc-that-normalize-to-separators-GH-12216.patch
* 020-bpo-36216-Only-print-test-messages-when-verbose-GH-12291.patch
* 021-2.7-bpo-35121-prefix-dot-in-domain-for-proper-subdom.patch
* 027-bpo-38243-Escape-the-server-title-of-DocXMLRPCServer.patch
* 028-bpo-34155-Dont-parse-domains-containing-GH-13079.patch
Patches no longer necessary and so removed:
* 017_lib2to3_fix_pyc_search.patch
Signed-off-by: Jeffery To <jeffery.to@gmail.com>
(cherry picked from 83b300aa83)
- Use HTTPS for downloading tarball and for their website
- Add PKG_CPE_ID
- Remove inactive maintainer
- Refreshed patches
Fixes: CVE-2019-14287
Signed-off-by: Josef Schlehofer <pepe.schlehofer@gmail.com>
This fixes the symlinks for libGammu.so and libgsmsd.so. Previously, the
symlinks were overwritten by $(INSTALL_BIN) with copies of their
sources.
Signed-off-by: Jeffery To <jeffery.to@gmail.com>
This seems to have slipped for some time. No idea if it ever worked.
It could be that this worked at some point.
In any case, the shebang is properly updated now.
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
(cherry-picked from commit 1b96dc0171)
Signed-off-by: Josef Schlehofer <pepe.schlehofer@gmail.com>
(bump PKG_RELEASE for python3-pip)
Update nano editor to version 4.5.
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
(cherry picked from commit bfd66f2d23)
(fix also license tag and add CVE)
These patches address issues:
CVE-2019-16056: email.utils.parseaddr mistakenly parse an email
CVE-2019-16935: A reflected XSS in python/Lib/DocXMLRPCServer.py
Links to Python issues:
https://bugs.python.org/issue34155https://bugs.python.org/issue38243
Signed-off-by: Jeffery To <jeffery.to@gmail.com>
Simplified the Makefile and fixes compilation with uClibc-ng. Also added
IPv6 support.
Took the time to clean up the Makefile with other useful options.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry-picked from commit 012e4c1)
Additionally to the fix issued for #6923, we need to disable the docbook
usage also for the host build. This prevents the following error:
checking for docbook2man... docbook2man
configure: error: Your local docbook2man was found to work with SGML rather
than XML. Please install docbook2X and use variable DOCBOOK_TO_MAN to point
configure to command docbook2x-man of docbook2X.
Or use DOCBOOK_TO_MAN="xmlto man --skip-validation" if you have xmlto around.
You can also configure using --without-docbook if you can do without a man
page for xmlwf.
Signed-off-by: Michael Heimpold <michael.heimpold@i2se.com>
CVE-2013-7459 and CVE-2018-6594. Both patches taken from Fedora.
Also took the liberty to update the PKG_SOURCE_URL to a standard one.
Updated the home URL as well.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry-picked from 32b23e28ad)
- Use HTTPS in their website
- Remove unnecessary space between PKG_SOURCE_URL
Signed-off-by: Josef Schlehofer <pepe.schlehofer@gmail.com>
Signed-off-by: Jan Pavlinec <jan.pavlinec@nic.cz>
Fixes compilation with -Werror=implicit-function-declaration .
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Signed-off-by: W. Michael Petullo <mike@flyn.org>
This patch updates OpenLDAP to 2.4.47, introduces new build
parameters and places openldap-server, openldap-utils and
libopenldap under a separate menu item in Network.
OpenLDAP is difficult to find in menuconfig at present. Making
a separate menu item for OpenLDAP for selection of packages and
enabling or disabling build parameters makes better sense.
To have access to the loglevel directive, OpenLDAP must be built
with debugging information. Having access to the loglevel directive
is essential during the initial configuration of OpenLDAP server.
International users may want to enable ICU support to have access
to international characters.
Signed-off-by: Val Kulkov <val.kulkov@gmail.com>
Signed-off-by: W. Michael Petullo <mike@flyn.org>
GNU patch through 2.7.6 is vulnerable to OS shell command injection that
can be exploited by opening a crafted patch file that contains an ed style
diff payload with shell metacharacters. The ed editor does not need to be
present on the vulnerable system. This is different from CVE-2018-1000156.
https://nvd.nist.gov/vuln/detail/CVE-2019-13638
Signed-off-by: Russell Senior <russell@personaltelco.net>
In GNU patch through 2.7.6, the following of symlinks is mishandled in
certain cases other than input files. This affects inp.c and util.c.
https://nvd.nist.gov/vuln/detail/CVE-2019-13636
Signed-off-by: Russell Senior <russell@personaltelco.net>
Make sure we force shutdown of UPS only when we should, and when
we should that shutdown happens.
Signed-off-by: Daniel F. Dickinson <cshored@thecshore.com>
The server and driver were not starting/restarting reliably. In
addition on interface changes NUT got very confused. So we fix
handling of restarts and add a reload trigger for interface
changes.
Signed-off-by: Daniel F. Dickinson <cshored@thecshore.com>
Extra parameters for the UPS driver were not being handled correctly.
Fix that (was wrong variable name).
Signed-off-by: Daniel F. Dickinson <cshored@thecshore.com>
The statepath was getting the wrong permission and/or not created
at the right time. This commit includes fixes for handling the
statepath (typically /var/run/nut).
Signed-off-by: Daniel F. Dickinson <cshored@thecshore.com>
Running as non-root was failing due to misplace local keyword
causing runas to be unset from calling value.
Signed-off-by: Daniel F. Dickinson <cshored@thecshore.com>
1) For upsmon start and stop were at wrong position in rc.d
2) Stop needs more than just killing the procd instead but rather
needs a stop command to be issued.
3) Interface up/down was causing not to enter a crashloop (we fix this
with procd trigger on interface changes).
Signed-off-by: Daniel F. Dickinson <cshored@thecshore.com>
Running as non-root was failing due to misplace local keyword
causing runas to be unset from calling value.
Signed-off-by: Daniel F. Dickinson <cshored@thecshore.com>
CONFIG_ARGS has --without-wrap so libwrap as a dependency is
extraneous as it is not actually used.
Signed-off-by: Daniel F. Dickinson <cshored@thecshore.com>
Needed to avoid error in case xxd is not installed:
[ 16%] Generating html.h from index.html
/bin/sh: 1: CMAKE_XXD-NOTFOUND: not found
CMakeFiles/ttyd.dir/build.make:61: recipe for target 'html.h' failed
make[6]: *** [html.h] Error 127
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry picked from commit 773c19afad)
The speedtest-netperf.sh script measures the network throughput while
monitoring latency under load and capturing key CPU usage and frequency
statistics. The script can emulate a web-based speed test by downloading
and then uploading from an internet server, or perform simultaneous
download and upload to mimic the stress of the FLENT test program.
It simplifies tasks such as validating ISP provisioned speeds or setting
up and fine-tuning SQM, directly on the router. The CPU usage details
can also help determine if the demands of SQM, routing and other tasks
such as the test itself are exhausting the device's CPUs.
This script leverages earlier scripts from the CeroWrt project used for
bufferbloat mitigation, betterspeedtest.sh and netperfrunner.sh. They are
used with the permission of the author, Rich Brown.
Signed-off-by: Tony Ambardar <itugrok@yahoo.com>
(cherry-picked from 463590e2bc)
This project seems abandoned. Updated to latest version.
Also cleaned up the Makefile quite a bit.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry-picked from dff6d2639e)
Fixes CVEs:
CVE-2018-5738
CVE-2018-5740
CVE-2018-5743
CVE-2018-5744
CVE-2018-5745
CVE-2019-6465
CVE-2019-6471
Signed-off-by: DENG Qingfang <dengqf6@mail2.sysu.edu.cn>
Signed-off-by: Josef Schlehofer <pepe.schlehofer@gmail.com>
[mention fixed CVEs;patches: refreshed and removed those which are in
upstream now]
Fixes CVE-2019-9923
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Signed-off-by: Josef Schlehofer <pepe.schlehofer@gmail.com>
[mention CVE in commit message]
Do not inlcude <netinet/in.h> when using glibc to avoid various
redefinitions.
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
(cherry-picked from 0c995cabe7)
To return the number of connected clients.
At present this can be partially inferred by using a count() over one of
the existing metrics, however this doesn't handle the case when there
are no connected clients. When that happens, the count() will return no
data instead of 0.
Signed-off-by: Alex Tomlins <alex@tomlins.org.uk>
(cherry picked from commit 1237e196b4)
These output a count of the number of packets transmitted/received, so
should be tracked as a counter. As it stands, promtool is warning that
these shouldn't be named ending _total if they're a gauge.
Signed-off-by: Alex Tomlins <alex@tomlins.org.uk>
(cherry picked from commit a45c702baa)
If the /proc/sys/net/netfilter/nc_conntrack_* files are not present,
this exporter was outputting a blank value, which is invalid. These
files will not be present when using an image that doesn't include the
iptables and firewall packages (eg a minimal access-point type image).
This updates the collector to only output the metrics if the
corresponding /proc files are present.
Signed-off-by: Alex Tomlins <alex@tomlins.org.uk>
(cherry picked from commit 0100a2cb26)
we should pass -DUNSUPPORTED_POSIX_THREADS_HACK to CFLAGS to openssh
to prevent function redefine, I don't know why pam module use
micro UNSUPPORTED_POSIX_THREADS_HACK to detect whether define
pthread functions, but not detect whether define
UNSUPPORTED_POSIX_THREADS_HACK.
Signed-off-by: Guo Li <uxgood.org@gmail.com>
ac_cv_header_libunwind_h needs to be set to 0, as ac_cv_have_libunwind_h
overwritten based on the former's value.
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
(cherry-picked from ec317934b2)
DEPENDS:=@!arc is causing a recursive dependency because of the optional
selection of libseccomp by util/lxc. The workaround hides the package
in Package/libseccomp/config instead.
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
(cherry-picked from e29483d7e2)
libseccomp does not compile for arc, so we can't show that option for
arc.
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
(cherry-picked from 57b4d94d98)
Otherwise one gets a failure if the libraries (e.g. Python
header file) exist in the build system. Worse in some cases
is host headers being found if one doesn't specifically
disable a library search in autotools. It is especially
important that Python is disabled by default.
Signed-off-by: Daniel F. Dickinson <cshored@thecshore.com>
(cherry-picked from cc039d801e)
Now that libaio compiles on arc targets we need to add the dependency to
libaio on these targets as well.
resolves#9298
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
(cherry-picked from c571c37e5c)
Remove patches 001-fix-unused.patch and 002-replace-nitems.patch,
similar fixes to make libudev-fbsd compile were applied upstream.
Refresh remaining patches.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
(cherry-picked from 58f1de60f8)
Copy vapi files to unversioned vala dir.
Added vala/host to PKG_BUILD_DEPENDS.
Removed TARGET_LDFLAGS
Removed copyright lines
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
(cherry-picked from 2cf76d5278)
Backport the Makefile from branches 'master' and 'openwrt-19.07'
- It means that now it will be created as Python package
with one difference that this one uses Python2
Signed-off-by: Josef Schlehofer <pepe.schlehofer@gmail.com>
Currently, all files in usr/bin (presumably all Python scripts) are run
through sed to replace the shebang; sed will overwrite the file whether
or not a match is found. This causes symlinks to be overridden and made
into copies of their targets. python[3]-base and python[3]-dev are
affected by this.
This adds the --follow-symlinks flag to sed, in addition to using
$(SED), so that symlinks are not overridden.
Signed-off-by: Jeffery To <jeffery.to@gmail.com>
This extends the Python[3] shebang fixup to all packages.
Only Python scripts in `/usr/bin` will be handled at the moment. Later it
may make sense to also cover executables in `/bin`, though typically Python
executables shouldn't be placed there.
Previously the shebang handling was only done for python[3]-pip &
python[3]-setuptools.
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Most interesting changes:
- Fixes CVEs
1.10.6
- CVE-2018-11782
- CVE-2019-0203
1.10.4
- CVE-2018-11803
Other misc changes:
- Fixes indentation in Makefile
- Reorder some things in Makefile to be sync with other packages
- Use HTTPS in URL
Signed-off-by: Josef Schlehofer <pepe.schlehofer@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>
(cherry-picked from 19bde8da68)
The code from 1996 is using pre-c99 inline semantics that cause compilation
failures on i386. Added std-gnu89 to fix this.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry-picked from 16f623b387)
Fixes issue where OpenWrt CFLAGS like relro were not being passed.
Updated license information.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry-picked from 5e34f4a32b)
It throws new warnings which triggers -Werror. Remove it.
Added support for full NLS.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
)cherry picked from 623d25a86b)
res_nsend and res_send are both not available in uClibc-ng as configured
in OpenWrt. Having this function return an error is the only sensible way
to fix.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry-picked from b5b776fc9a)
New string warnings were introduced. Adding -Wno-error to TARGET_CFLAGS
does not work so patch it out.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry-picked from 9eea417e92)
Previous GCC7 fix was incorrect and GCC8 correctly complains about it.
Fix it properly.
Also cleaned up the Makefile for consistency between packages.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry-picked from fac6f0e0a5)
Mainly endian definitions are needed.
Removed sys/io.h patch. It's a GNU header unsupported by uClibc-ng.
Cleaned up Makefile slightly.
Added HTTPS everywhere.
Added PKG_BUILD_PARALLEL for faster compilation.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry-picked from 12378cd4d3)
Upstream has a patch for this dated 2016.
Also ran the init script through shellcheck.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry-picked from f4f665c7db)
It applies the single patch here as well as adding some fixes.
The upstream project is dead.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry-picked from d2af82a6f5)
softethervpn overrides nls.mk by defining its iconv functions. This only
works if the libc has iconv. In addition, it does not allow external
libiconv usage. TARGET_LDFLAGS is also the wrong place to add -liconv.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry-picked from fb480e297c)
Remove hack to avoid readline host dependency, now that readline is
being host/built.
Pass on HOST_CFLAGS, HOST_CPPFLAGS, & HOST_LDFLAGS, to fix buildbots
host-compile errors about not finding openssl headers.
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
(cherry-picked from e3c6fcb796)
This change updates Python to version 2.7.16, which is a bugfix release
in the Python 2.7 series.
This also removes patches back-ported from upstream.
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
3.6.9 is the latest security fix release of Python 3.6.
- Fixes: CVE-2018-20852, CVE-2019-9948, CVE-2019-9740, CVE-2019-5010
- Refreshed patches
- Removed a few backports patches
Signed-off-by: Josef Schlehofer <pepe.schlehofer@gmail.com>
Busybox brctl applet conflicts with the version from bridge-utils.
Fix this by using ALTERNATIVE support for brctl in bridge-utils.
Fixesopenwrt/packages#8812
Signed-off-by: Konstantin Demin <rockdrilla@gmail.com>
[PKG_RELEASE bump]
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
(cherry-picked from commit fabd9efdea)
CFLAGS, CPPFLAGS & LDFLAGS need to be passed with CC because they are being ingored.
This is already contained in the master branch but was missing in 18.06. Without these flags the compilation fails.
Compile tested: Atheros AR7xxx/AR9xxx, TP-LINK Archer C7 v2, 18.06.4
Signed-off-by: Philipp Schuster <philippschuster@gmx.com>
Patch taken from upstream fixes an Invalid argument error while trying
to get the IP address of an interface.
Makefile was updated to current style.
(cherry-picked from 5ab9f3e357)
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
Small Makefile polishing
- removed PKG_BUILD_DIR
- PKG_MAINTAINER is above PKG_LICENSE
- URL should be under TITLE
Signed-off-by: Josef Schlehofer <pepe.schlehofer@gmail.com>
This reverts commit a176ffa0a1.
I just noticed that I accidentally used wrong openwrt branch
to compile this, with 18.09 branch the compilation fails.
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
This reverts commit 8e9429f3e9.
I just noticed that I accidentally used wrong openwrt branch
to compile this, with 18.09 branch the compilation fails.
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
This reverts commit 8abaf8a308.
I just noticed that I accidentally used wrong openwrt branch
to compile this, with 18.09 branch the compilation fails.
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
Also drop the CVE patches which are already covered by this new release.
Compile tested for and run tested on mxs platform.
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
(cherry picked from commit d5f0331c91)
Embarrasingly, I missed this one last time.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
(cherry picked from commit b7870ea711)
Update nano to 4.3
Release notes at https://nano-editor.org/news.php
2019.06.18 - GNU nano 4.3
* The ability to read from and write to a FIFO has been regained.
* Startup time is reduced by fully parsing a syntax only when needed.
* Asking for help (^G) when using --operatingdir does not crash.
* The reading of a huge or slow file can be stopped with ^C.
* Cut, zap, and copy operations are undone separately when intermixed.
* M-D reports the correct number of lines (zero for an empty buffer).
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
(cherry picked from commit 2bdd3856fd)
- Update haproxy download URL and hash
- Add new patches (see https://www.haproxy.org/bugs/bugs-1.8.20.html)
- Make halog use our CFLAGS
- Update statically linked LUA to v5.3.5
Signed-off-by: Christian Lachner <gladiac@gmail.com>
- fix package makefile resulted in resolved symlink and copied
double file contents of libunbound.so during install
- treat RFC6762 'local.' as nxdomain because avahi and other services
will disable if SOA or NS records appear in central DNS
- NLNetLabs has moved download and documentation sites
cherry-pick and squash from openwrt-19.07
add3e6c699936a71739c4618c74150
Signed-off-by: Eric Luehrsen <ericluehrsen@gmail.com>
All of these are either not needed or not valid.
Added a patch to remove the OPENSSL_WITH_DEPRECATED dependency.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry-picked from 430906bd78)
(refreshed patches)
INSTALL_DATA turns all of the symlinks to files, increasing size.
Signed-off-by: Deng Qingfang <dengqf6@mail2.sysu.edu.cn>
(cherry-picked from 3417ee0122)
This commit fixes the bug described in issue #8146 [1], where the
package fails to build if the boost package is selected without
selecting any of the internal non-header-only libraries.
[1]: https://github.com/openwrt/packages/issues/8146
Signed-off-by: Carlos Miguel Ferreira <carlosmf.pt@gmail.com>
(cherry-picked from 92c93ca0f4)
Forum user portuquesa raised a topic (see [1]) about being unable to use
Asterisk on his armeb xscale device. We narrowed it down to sqlite3.
Asterisk was unable to insert a simple table into its db.
In short, sqlite3 assumes little endian for every ARM device. This worked OK
for 4 Byte (unaligned) access. But once upstream (back in 2015) added a
function which accesses 2 Bytes (see [2]) this failed for some (if not all) ARM
big endian devices. ARM CPUs are bi-endian for 4 Byte reads but not for 2 Byte
reads.
This patch fixes the problem by setting the endianness adequately for
ARM targets, for both 32 bit and 64 bit varieties. The patch was applied
upstream (see [3]).
[1] https://forum.openwrt.org/t/solved-asterisk13-or-15-sqlite3-database-problem/36856
[2] 329428e208
[3] https://www.sqlite.org/src/info/b7aad929619f7043
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
- Fixes CVE-2019-9740, CVE-2019-11324
- RFC 3986 compliant
- Fix TITLE, which was too long for make menuconfig
- Add PKG_LICENSE_FILES
- Remove current maintainer
Signed-off-by: Josef Schlehofer <pepe.schlehofer@gmail.com>
libbsd may compile before nfs-kernel-server, it will make
nfs-kernel-server depends libbsd.so.0, that is not we want to see. so
gave option to 'configure' to disable libbsd detect and tell it we have
no libbsd
Signed-off-by: Guo Li <uxgood.org@gmail.com>
(cherry picked from commit 3fc7f7b8cc)
* include /etc/netconfig
* cleanup old patches
* update via patch to 1.0.4-rc2
Signed-off-by: Andy Walsh <andy.walsh44+github@gmail.com>
(cherry picked from commit 4896e9c36e)
Some of these hacks were needed when uClibc++ was used. Now that it is not
we can remove them.
Remove libpthread dependency. Not only is it unneeded, package-defaults
already specifies it.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
[jonas.gorski: cherry-pick from master]
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
(cherry picked from commit 9cb7cc71e7da9d06194b813d6263c0b8427eb1c3)
* update nano to 4.2
Release notes at https://nano-editor.org/news.php
2019.04.24 - GNU nano 4.2
* The integrated spell checker does not crash when 'spell' is missing.
* Option --breaklonglines works also when --ignorercfiles is used.
* Automatic hard-wrapping is more persistent in pushing words to the
same overflow line.
Tested with ipq806x/R7800 and mvebu/WRT3200ACM
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
(cherry picked from commit aee8628696)
The LIBDIR and INCDIR assignments are duplicate of the original
Makefile, changing LIB_PATH and INC_PATH to LDFLAGS and CPPFLAGS.
Setting LIB_PATH and INC_PATH to empty strings will do the same
without duplicating the flags.
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
Linking with ld is not portable and was causing problems for some
targets, e.g. i386_pentium4:
i486-openwrt-linux-musl-ld: x509.o: in function `push_asn1_objname':
x509.c:(.text+0x61): undefined reference to `__stack_chk_fail_local'
...
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
(cherry-picked from e3f1532297)
Currently i386 and the PPC targets have issues linking issues.
https://github.com/openwrt/packages/issues/3319
says that replacing -fPIC with -fpic works.
Patch added to avoid package overriding settings set by toolchain and make
compilation less noisy
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry-picked from cf23dd2eb0)
This is causing linking errors on i3486 and maybe other platforms. Linking
with LD does not seem to be very portable.
Also cleaned up the Makefile by getting rid of whitespace, HTTPS,
duplicated entries, etc...
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Added PKG_BUILD_PARALLEL for faster compilation.
Removed PKG_FORMAT_SECURITY as it's not needed to be disabled anymore.
Removed patch as musl has been updated.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
If ld is used directly to link libmilter.so, linking is not done
correctly with glibc, and produces a library that can't be used.
Linking with the bad library produces a strange error:
ld: conftest: hidden symbol `stat' in [...]libc_nonshared.a(stat.oS)
is referenced by DSO
Using ld with musl works fine, but using gcc is a more portable way of
linking the shared library anyway.
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
[ Upstream commit 2bc28eb40c ]
Update to 10.03.0000.
--with-unixodbc should point to the odbc_config binary, not to the top
of the install directory $(STAGING_DIR)/usr.
Acked-by: Rosen Penev <rosenp@gmail.com>
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
Probably a result of GCC7 that it fails now.
Added PKG_CPE_ID for proper CVE tracking.
Added PKG_BUILD_PARALLEL for faster compilation.
Reorganized Makefile slightly for consistency with other packages.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
darkstat includes its own strlcat and strlcpy, making the dependency
somewhat pointless.
Fixes compilation ever since glibc dependency on libbsd was removed.
Also removed std=gnu99 as it's not needed with GCC7.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Since the switch to GCC7, this has not compiled as it assumes gnu89 behavior.
-fgnu89-inlining is not enough so use std=gnu89.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
The previous source location was an SVN mirror from long ago. This one is
from the author himself.
Reduced overall patch count. A lot of them are no longer necessary.
Switched back to libstdcpp as the necessary uclibc++ changes are not being
merged.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Now that the library has been updated, we can also update this.
Switched to codeload as we don't need the submodule anymore.
Various other Makefile consistency updates.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
This version contains NVME fixes.
Edited the patches to modify the package Makefile directly so as to reduce
hacks in the OpenWrt Makefile.
Added LTO support to lower filesize by ~4KB.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
These patches address issue:
CVE-2019-9636: urlsplit does not handle NFKC normalization
Link to Python issue:
https://bugs.python.org/issue36216
Signed-off-by: Jeffery To <jeffery.to@gmail.com>
Save a copy of unixodbc_conf.h in STAGING_DIR to be used by host build.
Use STAGING_DIR/tmp/unixodbc instead of include.
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
For host compilation, the configure-generated config.h from the target
compilation is used in place of the host-generated file. When the
target package is compiled with clean-build, that file is gone. This
saves the file under $(STAGING_DIR), and fetch it from there.
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
When copying config.h from PKG_BUILD_DIR to HOST_BUILD_DIR, LIB_PREFIX
is set to /usr/lib. Then when odbc_config is run, it reports /usr/lib
as the --lib-dir, and in --libs as well, and dependent packages may
fail. Set it to $(STAGING_DIR)/usr/lib to make it right.
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
Very minor bugfix.
Also adjusted standard to gnu89 to fix compilation issues (lot of missing
prototypes).
Signed-off-by: Rosen Penev <rosenp@gmail.com>
This should fix compilation as wolfSSL currently does not define
wolfTLSv1_client_method. And as the comment suggests, this is only TLS 1,
not 1.0 and above.
SSLv23 is TLS 1.1 and above as currently configured in the wolfssl package
Signed-off-by: Rosen Penev <rosenp@gmail.com>
* update nano to 4.1
* implement Makefile style changes proposed in #8483
Release notes at https://nano-editor.org/news.php
2019.04.15 - GNU nano 4.1
* By default, a newline character is again automatically added at the
end of a buffer, to produce valid POSIX text files by default, but
also to get back the easy adding of text at the bottom.
* The now unneeded option --finalnewline (-f) has been removed.
* Syntax files are read in alphabetical order when globbing, so that
the precedence of syntaxes becomes predictable.
* In the C syntax, preprocessor directives are highlighted differently.
* M-S now toggles soft wrapping, and M-N toggles line numbers.
* The jumpy-scrolling toggle has been removed.
* The legacy keystrokes ^W^Y and ^W^V are recognized again.
* Executing an external command is disallowed when in view mode.
* Problems with resizing during external or speller commands were fixed.
Tested with ipq806x R7800
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
(cherry picked from commit 5c212b1a13)
Correct option is "password_file" not "passwd_file"
Originally reported as: https://github.com/openwrt/packages/pull/8642
Added the package bump.
Signed-off-by: Karl Palsson <karlp@etactica.com>
Also refresh patch which does not apply cleanly anymore.
Run tested on Duckbill for mxs platform.
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
(cherry picked from commit 80cdd53134)
While at, add --with-pic to configure arguments. This prevents the following
build errors spotted by the build bots for i386 targets:
-snip-
...
ext/openssl/.libs/openssl.o: direct GOT relocation R_386_GOT32X against
`X509_REQ_free' without base register can not be used when making a shared object
...
-snap-
This parameter seems to make no difference on other targets, nor
improve or make worse the package size.
Run tested for i386 in VirtualBox VM and on Duckbill for mxs platform.
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
(cherry picked from commit 963c841463)
This orders loading of openssl extension before extensions
which require openssl functions.
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
(cherry picked from commit 5afeb3f9c8)
Update nano editor to version 4.0.
Release notes at
http://git.savannah.gnu.org/cgit/nano.git/plain/NEWS?h=v4.0
2019.03.24 - GNU nano 4.0 "Thy Rope of Sands"
* An overlong line is no longer automatically hard-wrapped.
* Smooth scrolling (one line at a time) has become the default.
* A newline character is no longer automatically added at end of buffer.
* The line below the title bar is by default part of the editing space.
* Option --breaklonglines (-b) turns automatic hard-wrapping back on.
* Option --jumpyscrolling (-j) gives the chunky, half-screen scrolling.
* Option --finalnewline (-f) brings back the automatic newline at EOF.
* Option --emptyline (-e) leaves the line below the title bar unused.
* <Alt+Up> and <Alt+Down> now do a linewise scroll instead of a findnext.
* Any number of justifications can be undone (like all other operations).
* When marked text is justified, it becomes a single, separate paragraph.
* Option --guidestripe=<number> draws a vertical bar at the given column.
* Option --fill=<number> no longer turns on automatic hard-wrapping.
* When a line continues offscreen, it now ends with a highlighted ">".
* The halfs of a split two-column character are shown as "[" and "]".
* A line now scrolls horizontally one column earlier.
* The bindable functions 'cutwordleft' and 'cutwordright' were renamed
to 'chopwordleft' and 'chopwordright' as they don't use the cutbuffer.
* The paragraph-jumping functions were moved from Search to Go-to-Line.
* Option --rebinddelete is able to compensate for more misbindings.
* Options --morespace and --smooth are obsolete and thus ignored.
* The --disable-wrapping-as-root configure option was removed.
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
(cherry picked from commit f1d51dbf76)
- CVE-2019-3855
Possible integer overflow in transport read allows out-of-bounds write
- CVE-2019-3856
Possible integer overflow in keyboard interactive handling allows
out-of-bounds write
- CVE-2019-3857
Possible integer overflow leading to zero-byte allocation and out-of-bounds
write
- CVE-2019-3858
Possible zero-byte allocation leading to an out-of-bounds read
- CVE-2019-3859
Out-of-bounds reads with specially crafted payloads due to unchecked use of
`_libssh2_packet_require` and `_libssh2_packet_requirev`
- CVE-2019-3860
Out-of-bounds reads with specially crafted SFTP packets
- CVE-2019-3861
Out-of-bounds reads with specially crafted SSH packets
- CVE-2019-3862
Out-of-bounds memory comparison
- CVE-2019-3863
Integer overflow in user authenicate keyboard interactive allows
out-of-bounds writes
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
2.5.5: Bug fix for a deadlock in multi-thread/multi-process (using Process.fork) applications, like for example Puma
2.5.4: Fixes multiple vulnerabilities:
CVE-2019-8320: Delete directory using symlink when decompressing tar
CVE-2019-8321: Escape sequence injection vulnerability in verbose
CVE-2019-8322: Escape sequence injection vulnerability in gem owner
CVE-2019-8323: Escape sequence injection vulnerability in API response handling
CVE-2019-8324: Installing a malicious gem may lead to arbitrary code execution
CVE-2019-8325: Escape sequence injection vulnerability in errors
Signed-off-by: Luiz Angelo Daros de Luca <luizluca@gmail.com>
When the server hostname resolved to both IPv4 and IPv6 addresses,
connecting would fail with nothing in syslog. This corrects that oversight.
Signed-off-by: Daniel Gimpelevich <daniel@gimpelevich.san-francisco.ca.us>
(cherry picked from ca56324 and PKG_MIRROR_HASH removal from 494ce71)
Revert the addition of build dependency in commit 2d1694ff7
to a non-existent host build of zlib.
The host build of zlib was removed already in April 2018 by
8dcd941d8b (diff-1ed408c61d79f9c6c5d197333e94ce8d)
which made zlib a build tool defined in /tools
The newly introduced build dependency causes always a warning like:
WARNING: Makefile 'package/feeds/packages/postgresql/Makefile'
has a build dependency on 'zlib/host', which does not exist
Not sure what was the error that 2d1694ff7 tried to fix,
but reference to a non-existent host build is not the solution.
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
(cherry picked from commit d8e61d49da)
spotted on buildbot trying postgresql/host build:
configure: error: zlib library not found
Fix this by adding zlib/host to HOST_BUILD_DEPENDS.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
(cherry-picked from commit 2d1694ff7c)
cherry-pick and squash commits from master for GNUnet
04eb431cb libgabe: add package
7831fb63b libgabe: update to shared library version
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
cherry-pick commit 4c5d25458 libpbc: add new package
from master as GNUnet started to depend on libgabe which depends on
libpbc.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
revert 7b2bf511c gnunet: Specify libmicrohttpd-ssl dependency
which was accidentally merged from master while the rename of the
libmicrohttpd* packages has happened only on master.
Revert it for openwrt-18.06.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Backport and squash the following commits from master:
43ec390bd postgresql: security bump to 9.6.10
845aab78a postgresql: Update to 9.6.11
fe6597dd7 postgresql: update to version 9.6.12
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Backport and squash the following commits from master:
853e9d1c3 libextractor: Update to 1.7
1a23de5db libextractor: update to version 1.8
a50f26941 libextractor: fix PKG_HASH
6709d9b82 libextractor: update to version 1.9
Backport and squash the following commits from master:
af06f6fd5 gnurl: update to version 7.61.1
7cdbb7569 gnurl: build without libpsl
d34eda733 gnurl: update to version 7.63.0
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
This is a minor bugfix release. Full changelog available at:
https://mosquitto.org/blog/2019/02/version-1-5-7-released/
Most relevant to OpenWrt are probably:
* fixing persistent store bloat
* fix sorting of included config files
* fix errors related to per_listener_settings
Signed-off-by: Karl Palsson <karlp@etactica.com>
Link to Python bug:
https://bugs.python.org/issue34656
Upstream commit:
71a9c65e74
OpenWrt 18.06 contains version Python 3.6.5, which doesn't contain this
fix.
Python 2.7 is not affected.
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
This is a bugfix and security release.
CVE-2018-12551: If Mosquitto is configured to use a password file for
authentication, any malformed data in the password file will be
treated as valid. This typically means that the malformed data becomes
a username and no password. If this occurs, clients can circumvent
authentication and get access to the broker by using the malformed
username. In particular, a blank line will be treated as a valid empty
username. Other security measures are unaffected.
=> Users who have only used the mosquitto_passwd utility to create and
modify their password files are unaffected by this vulnerability.
CVE-2018-12550: If an ACL file is empty, or has only blank lines or
comments, then mosquitto treats the ACL file as not being defined,
which means that no topic access is denied. Although denying access to
all topics is not a useful configuration, this behaviour is unexpected
and could lead to access being incorrectly granted in some
circumstances.
CVE-2018-12546. If a client publishes a retained message to a topic
that they have access to, and then their access to that topic is
revoked, the retained message will still be delivered to future
subscribers. This behaviour may be undesirable in some applications,
so a configuration option `check_retain_source` has been introduced to
enforce checking of the retained message source on publish.
Plus the following bugfixes:
* wills not sent to websocket clients
* spaces now allowed in bridge usernames
* durable clients not receiving offline messages with
per_listener_settings==true
* compilation with openssl without deprecated apis
* TLS working over SOCKS
* better comment handling in config files
Full changelog available at: https://github.com/eclipse/mosquitto/blob/fixes/ChangeLog.txt#L1
Signed-off-by: Karl Palsson <karlp@etactica.com>
Upstream Release Notes:
- MDEV-17475: Maximum value of table_definition_cache is now 2097152
- MDEV-13671: InnoDB should use case-insensitive column name comparisons
like the rest of the server
- ALTER TABLE fixes: MDEV-17230, MDEV-16499, MDEV-17904, MDEV-17833,
MDEV-17470, MDEV-18237, MDEV-18016
- Improvements to InnoDB page checksum, recovery, and Mariabackup:
MDEV-17957, MDEV-12112, MDEV-18025, MDEV-18279, MDEV-18183
- Galera
- MDEV-15740: Galera durability fix
- New configuration variable wsrep_certification_rules, used for
controlling whether to use new/optimized
(--wsrep_certification_rules=optimized) certification rules or the
old/classic ones (--wsrep_certification_rules=strict). Setting the
variable to strict can cause more certification failures.
- Fixes for the following security vulnerabilities:
- CVE-2019-2537
- CVE-2019-2529
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
This upstream release adds support for trust_anchors_backoff_time
configuration parameter. UCI support has been added for this.
This commit also includes a number of clean-ups:
o change START=50 to START=30 in init file
Starting earlier in the boot means less chance of missing interface
trigger events. See: https://github.com/openwrt/packages/pull/4675
o remove unused variables from init file
o separate local declarations and assignments in init file
o add defensive quoting in init file
o use default values for procd respawn in init file
o make use of {} in variables consistent in init file
o remove unused variable from init file
Signed-off-by: Jonathan G. Underwood <jonathan.underwood@gmail.com>
I am no longer able to support maintaining the stubby daemon for openwrt. I suggest Jonathan Underwood <jonathan.underwood@gmail.com> as a replacement.
This commit brings UCI support to the stubby package.
o All options are documented in the README.md file.
o The README.md file has been re-written to include a short usage
manual.
o The default configuration now includes more Cloudflare addresses.
o The stubby service is (re)started using procd triggers from a
specified interface with a configurable time delay.
o Round robin use of upstream resolvers is now activated by
default.
o Client privacy is now activated by default.
o Options are added for specifying the log level of the daemon and
command line options passed to the stubby command.
Signed-off-by: Jonathan G. Underwood <jonathan.underwood@gmail.com>
Remove the limit setting core="unlimited", since this shouldn't be needed
in production use (i.e. non-debug) and on an embedded platform, which is
why it's rarely used by any existing packages.
Signed-off-by: Tony Ambardar <itugrok@yahoo.com>
Add an SPKI pin for Cloudflare to help prevent MITM and downgrade attacks,
as described in RFC7858 (DNS over TLS). The setup of SPKI and the specific
SHA256 certificate hash are taken from Cloudflare's DoT configuration guide
published at https://developers.cloudflare.com/1.1.1.1/dns-over-tls/.
Note that the certificate is valid to March 25th 2020, 13:00 CET, which
provides ample time for issuance of a backup pin to support future key
rollover.
Signed-off-by: Tony Ambardar <itugrok@yahoo.com>
Retain the upstream value since privacy is usually the key user motivation
for using DNS-over-TLS, and simply note that those encountering sub-optimal
routing may consider disabling the setting.
Signed-off-by: Tony Ambardar <itugrok@yahoo.com>
The config file /etc/stubby/stubby.yml is not registered properly and any
local changes are being overwritten on upgrade or reinstall.
Signed-off-by: Tony Ambardar <itugrok@yahoo.com>
Move loading credential function before cert renewal call as credentials might be needed for some renewal operations ( ex: DNS )
Signed-off-by: Adrien DAURIAT <16813527+dauriata@users.noreply.github.com>
The first is needed for 4.14 (maybe the relevant parts got packported and
the second is for when OpenWrt migrates to 4.19.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry picked from commit 4f28139a3c)
Signed-off-by: Daniel Albers <Daniel.Albers@public-files.de>
Since the new hotplug script in master was not backport (new feature),
for 18.06 branch revert the old behavior of running NUT daemons and
drivers as root by default to avoid permisions problems, but backport
fix the support for running as another user for those who can set the
appropriate permissions on the USB (or other) device.
Closes: #7742
Signed-off-by: Daniel F. Dickinson <cshored@thecshore.com>
Defaulting to -fPIC is a bad idea, especially for executables (here:
sqlite3-cli). In short, there are certain security implications as well
as overhead/performance penalties. Details see:
https://wiki.gentoo.org/wiki/Project:Hardened/Position_Independent_Code_internals
The configure script is able to detect the need for PIC and adds the
flag when needed anyway (when compiling the library).
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
A remote code execution vuln has been found in sqlite. Infos available
here:
https://blade.tencent.com/magellan/index_en.html
sqlite 3.26.0 contains the fix.
This commit also changes source URL to https.
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
Backport and squash the following commits from master:
5790053eb nut: Add missing conffiles
ceff68837 nut: Reorganize nut-server to clarify nut-driver
f6a2a97d2 nut: Use 'real' procd init for nut-monitor
918a62f91 nut: Make FSD really work
a2f64b3ba nut: Reduce user error with POWERDOWNFLAG
461393810 nut: Use quotes around filenames
1b6dbe7a7 nut: Remove duplicate/extraneous lines
0a49d0ffb nut: Fix checking for path before it exists
3b5a8eee8 nut: Various startup fixes for monitor and server
44e57d4bd nut: Fix variables for NUT drivers
36fd59dc7 nut: Fix extraneous config_get
192b0f164 nut: Fix a typo in setting a driver parameter
f48b060fa nut: Fix upsd runs as root
And bump PKG_RELEASE
Signed-off-by: Daniel F. Dickinson <cshored@thecshore.com>
This bumps libtiff's minor version from 9 to 10. In addition to the CVE
fixes that we already included this fixes:
CVE-2017-17095
CVE-2018-17101
CVE-2018-18557
The update is 100% backwards compatible, no symbol changes.
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
Adds support for acl_plugin, and acl_opt_* options.
acl_opt_* requires some care as it relies on the internal behaviour of
cfg_load setting environment variables in a certain form. However,
given that _all_ of the cfg_load infrastructure relies on that, we can
be pretty sure that it won't change in a way that will hurt us.
Originally reported as: https://github.com/openwrt/packages/pull/7434
Signed-off-by: Karl Palsson <karlp@etactica.com>
This is the same change as the one on master
This is to change the init script to a procd init script
This also enable some additional parameters in the binary that
were present but not enabled:
The export file (option export_file)
The import file (option import_file)
The daylog (option daylog_file)
These are disabled by default. Also, the option to run as a daemon
is removed, as not compatible with procd.
There is no change in the binary.
Signed-off-by: Jean-Michel Lacroix <lacroix@lepine-lacroix.info>
Checking:
- Pull request does not contain unwanted merges
- signed-off-by tag exists and matches author
- Subject line has package name
- Author name has 'firstname lastname' (no nicknames)
Signed-off-by: Ted Hess <thess@kitschensync.net>
[Use git instead of CircleCI variables]
Signed-off-by: Etienne Champetier <champetier.etienne@gmail.com>
libmariadb 10.2 needs to be linked in together with iconv.
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
(cherry picked from commit b695c1150a)
Even on a powerful platform a collectd process'
activities are sometimes affecting throoughput and
latency. This is a backgroud process, that should not
be running with default priority.
Even if it is a little deplayed, that is not a worry in
this case. The routing should be the main priority,
stats collection can wait a bit.
Tested on Netgear R7800
Signed-off-by: Marc Benoit <marcb62185@gmail.com>
Make niceness more moderate, bump version.
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
(cherry picked from commit b33ec70c95)
Hard float includes: mp3lame
Soft float includes: shine (mp3 encoder)
libx264 is included when selected iff BUILD_PATENTED is true.
fdk-aac will not be available in libffmpeg-full due to incompatible license with libx264.
Custom builds can override licensing restrictions but results may not be re-distributable.
Signed-off-by: Ted Hess <thess@kitschensync.net>
Notable Changes (copied from release notes):
Various fixes from MySQL 5.6.42: MDEV-17533, MDEV-17532, MDEV-17531
MDEV-16465: fixed a bug with DDL and FOREIGN KEY
Fulltext index fixes:
MDEV-12547: extended the range of innodb_ft_result_cache_limit on 64-bit systems
MDEV-16865: InnoDB fts_query() ignores KILL
Fixes for the following security vulnerabilities:
CVE-2018-3282
CVE-2016-9843
CVE-2018-3174
CVE-2018-3143
CVE-2018-3156
CVE-2018-3251
OpenWrt changes:
- dropped obsolete ucontext patch (issue fixed upstream)
- refreshed 130-c11_atomics.patch
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
This is basically same commit that took place in master 3 weeks ago.
gcc-7 with -Os makes inline functions disappeard. It is caused by
the new C11 inline semantics. pass option -fgnu89-inline to gcc let
it use gnu inline semantics.
see https://wiki.debian.org/GCC7#Porting_help
Compile tested on 18.06. Run tested on OpenWrt 18.06.1 r7258-5eb055306f
QEMU Virtual CPU version (cpu64-rhel6)
Signed-off-by: Jean-Michel Lacroix <lacroix@lepine-lacroix.info>
I used podman/buildah to build this image, and the local sha256 is not the same than
the docker hub sha256. The layers are the same, so maybe just docker hub changing the manifest
Signed-off-by: Etienne Champetier <champetier.etienne@gmail.com>
Fix only release, including:
* CVE-2018-16396: Tainted flags are not propagated in Array#pack
and String#unpack with some directives
* CVE-2018-16395: OpenSSL::X509::Name equality check does not work
correctly
Signed-off-by: Luiz Angelo Daros de Luca <luizluca@gmail.com>
(cherry picked from commit 74216a55e1)
The only known user of this library is currently unable to get their
application to work with with the fixed 0.7.6 release of this library.
To prevent accidental use by unknown parties of a flawed library, mark
it as BROKEN.
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
(cherry picked from commit 14ad4cb976)
The last commit added PKG_CPE_ID and now uscan detects a CVE that I missed
Reordered patches by date
Signed-off-by: Rosen Penev <rosenp@gmail.com>
[tweaked commit message]
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
(cherry picked from commit 394ff73e53)
Patches taken from official git repository.
Added PKG_CPE_ID for proper CVE tracking.
Added PKG_BUILD_PARALLEL for faster compilation.
Also adjusted Makefile to be more similar to other projects.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry picked from commit 6f0ff25503)
Full changelog at https://github.com/eclipse/mosquitto/blob/v1.5.3/ChangeLog.txt
Primary change:
CVE fix for CVE-2018-12543 - prevent crash on topics that begin with $
but are not $SYS
Selected other fixes relevant to OpenWrt since 1.5.1:
- Fix retained messages not sent by bridges on outgoing topics at the first
connection. Closes#701.
- Fix duplicate clients being added to by_id hash before the old client was
removed. Closes#645.
- Fix excessive CPU usage when the number of sockets exceeds the system limit.
Closes#948.
- Fix for bridge connections when using WITH_ADNS=yes.
- Fix round_robin false behaviour. Closes#481.
- Fix segfault on HUP when bridges and security options are configured.
Closes#965.
Signed-off-by: Karl Palsson <karlp@etactica.com>
musl libc (http://musl-libc.org lack the non-standard <fpu_control.h>
header, which is used in src/os/linux/{i386,x86_64}/init.c files to
setup the floating point precision. This patch makes it use the
standard C <fenv.h> header instead.
Original patch at Felix Janda at
https://sourceforge.net/p/jamvm/patches/6/
Signed-off-by: Guo Li <uxgood.org@gmail.com>
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
Makefile and package changes to support builds with both Python 2.x and Python 3.x versions.
Python versioning is automatically configured from lang/python repository xxx-version.mk files.
Signed-off-by: Ted Hess <thess@kitschensync.net>
Commit 6cd8fcabe added ipsec hotplug script support by calling "exec
/sbin/hotplug-call ipsec".
Using the exec call breaks the insertion of iptables rules by the _updown.in
script as hotplug-call just replaces the current shell meaning the commands
following exec do not run since the shell is replaced and as a result lead to
connectivity issues.
Fix this by removing the exec command in front of /sbin/hotplug-call.
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
(cherry picked from commit d0ac611bf0)
Ipsec user script (/etc/ipsec.user) now get called indirectly by openwrt
"/sbin/hotplug-call". So other packages could also install their scripts
in "/etc/hotplug.d/ipsec".
Signed-off-by: Florian Eckert <fe@dev.tdt.de>
(cherry picked from commit 6cd8fcabe6)
ibmariadb 10.2 needs to be linked in together with iconv.
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
(cherry picked from commit 7a0aebbff2)
Commit ae5ee6ba6c added support for inbound
firewall rule support but some corner cases were not covered.
In case net-snmp is started and the network interface is already up
the procd firewall rule is created but not applied by fw3 as
service_started calling procd_set_config_changed firewall was missing.
When stopping net-snmp clean up the net-snmp inbound firewall rules in
iptables by calling procd_set_config_changed firewall in stop_service
which will trigger fw3 to remove the inbound firewall rules.
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
(cherry picked from commit 0bd19db049)
libbsd gets picked up since it's no longer limited to glibc.
Patch identical to libtalloc one. Same codebase.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry picked from commit 33dc529e00)
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>
(cherry picked from commit 75d9ab331d)
Automatic detection of the arm architecture does not work well.
http://downloads.lede-project.org/snapshots/faillogs/arm_arm1176jzf-s_vfp/packages/node/compile.txt
```
../deps/v8/src/arm/assembler-arm.cc:176:2: error: #error "CAN_USE_ARMV7_INSTRUCTIONS should match CAN_USE_VFP3_INSTRUCTIONS"
#error "CAN_USE_ARMV7_INSTRUCTIONS should match CAN_USE_VFP3_INSTRUCTIONS"
^~~~~
```
https://github.com/openwrt/packages/issues/5728
Explicitly set cpu arch optimization flag to the compiler option so that "configure" script correctly identifies "arm version".
(Raspberry Pi Zero W)
Raspbian:
```
raspberrypi:~ $ echo | gcc -dM -E - | grep ARM_ARCH
```
OpenWrt (cross-env):
```
ubuntu:~ $ echo | ./arm-openwrt-linux-muslgnueabi-gcc -dM -E - | grep ARM_ARCH
```
```
ubuntu:~ $ echo | ./arm-openwrt-linux-muslgnueabi-gcc -mcpu=arm1176jzf-s -dM -E - | grep ARM_ARCH
```
Also specifying an option lines compactly.
Signed-off-by: Hirokazu MORIKAWA <morikw2@gmail.com>
(cherry picked from commit 3482320c2a)
modify patch.
https://github.com/nodejs/node/pull/19196
made not to use libressl headers
fix to include path not to use "host/include"
Signed-off-by: Hirokazu MORIKAWA <morikw2@gmail.com>
(cherry picked from commit 818770d273)
* Update nano editor to 3.1
* Apply a post-release upstream patch to fix compilation
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
(cherry picked from commit af86b17064)
Fix the init script to allow access from IPv6 subnets of the interface
specified in allow section in /etc/config/chrony.
Fixes issue #7039.
Signed-off-by: Miroslav Lichvar <mlichvar0@gmail.com>
add extra command "export_storage" to export data for use with Radicale 2.x.x
remove myself as PKG_MAINTAINER
Signed-off-by: Christian Schoenebeck <christian.schoenebeck@gmail.com>
The following error shows that mysqlnd depends on functions
provided by hash:
root@OpenWrt:/etc/php7# php-cli -m
PHP Warning: PHP Startup: Unable to load dynamic library
'mysqlnd.so' (tried: /usr/lib/php/mysqlnd.so (Error
relocating /usr/lib/php/mysqlnd.so: PHP_SHA256Final: symbol
not found), /usr/lib/php/mysqlnd.so.so (Error loading shared
library /usr/lib/php/mysqlnd.so.so: No such file or
directory)) in Unknown on line 0
So let's model this dep in package metadata.
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
Bugfix release. Full changelog at:
https://mosquitto.org/blog/2018/08/version-151-released/
Of most interest to OpenWrt:
* Remove use of AI_ADDRCONFIG, which means the broker can be used on systems where only the loopback interface is defined.
* Fix IPv6 addresses not being able to be used as bridge addresses.
* Fix problem opening listeners on Pi caused by unsigned char being default.
* Fix segfault on startup if bridge CA certificates could not be read.
* Fix possible endian issue when reading the memory_limit option.
* library and client bugfixes including: https://github.com/openwrt/packages/issues/6765
Signed-off-by: Karl Palsson <karlp@etactica.com>
Backport Rosen's commit in master to 18.06 to address open CVEs. This
fixes:
CVE-2017-11613
CVE-2018-5784
CVE-2018-7456
CVE-2018-8905
CVE-2018-10963
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
This update includes fixes for the following CVEs:
- CVE-2018-1115
- CVE-2018-10925
- CVE-2018-10915
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
Currently external modules and non-base packages are numbered
from their own internal number space, and even though the Perl
ABI number is embedded into them this isn't externally visible.
For example, perl-html-parser-3.72.1 could be built for ABI
5.26 or for 5.28, we can't easily tell. This changes all of
that by embedding the ABI number into the filename.
Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com>
(cherry picked from commit 0d9584724f)
Bump minor version. Bugfix release. 100% backward compatible.
Includes fixes for:
CVE-2018-3064
CVE-2018-3063
CVE-2018-3058
CVE-2018-3066
Also includes CPPFLAGS fix from master (to get fortify-source headers
etc.).
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
* fix restart behaviour after successful connection
* fix labeling of faulty stations
* optimize re-connect behaviour at locations where multiple uplinks with
the same SSID are in range
* use procd pidfile handling
* refine logging
* small fixes
Signed-off-by: Dirk Brenken <dev@brenken.org>
(cherry picked from commit bc78ad8200)
Since https://git.openwrt.org/d0e0b7049f88774e67c3d5ad6b573f7070e5f900,
OpenWrt SDKs ship the appropriate sources for building usbip userspace
packages, so special nonshared handling is not required anymore.
Sucessfully tested by compiling usbip utilities for various architectures
using self built SDKs after applying the change linked above.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit eded6ed7cf)
Refresh patches 900 and 910.
Add fix (920) for improperly gated variable.
Add workaround (020) for Storable's run-time check for stacksize.
Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com>
(cherry picked from commit b94744496f)
Since Travis runs as a batch job and logs are typically looked at
after-the-fact (if at all), there's not any point to seeing progress
when cloning git repos.
Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com>
(cherry picked from commit fc3beffeff)
The internal nameservers and the DHCP default domain should be
squirted into /tmp/resolv.conf.
Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com>
(cherry picked from commit db6adb616d)
Looking at Configure and Porting/config.sh, it seems that Perl
requires both of these options to build correctly.
Should fix FS #1464.
Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com>
nut-monitor failed to create required dir /var/etc/nut, as
well as failing to set appropriate user on the directory and
conf files. Fixing this closes
https://github.com/openwrt/packages/issues/6644
Signed-off-by: Daniel F. Dickinson <cshored@thecshore.com>
(cherry picked from commit 98fdf16ecb)
We need to force this since a *.m4 file is patched.
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
(cherry picked from commit 6543859dab)
OpenWrt changed the way the uci shell parsing functions deal with list
configuration items.
This change broke the generation of the privoxy runtime configuration
because no callbacks were emitted anymore.
Fix the problem by defining a list_cb() that simply calls the existing
option_cb() to deal with list item values.
Ref: c9c0fc28a9 ("base-files: fix UCI config parsing and callback handling")
Ref: https://forum.lede-project.org/t/openwrt-snapshot-privoxy-error/15919
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit 3113d62393)
The sources for usbip are within the kernel. A patch that was included
with the package, which changed the old signal name SIGCLD to the new
one, SIGCHLD, was merged upstream. However, different targets use
different kernel versions. Current version 4.14 and 4.9 are fine, but
older versions do not have the patch applied. So, I used
-DSIGCLD=SIGCHLD to please both worlds.
libudev-fbsd currently used by openwrt does not implement the
udev_device_get_devpath function. eudev's implementation of libudev
sets it as (src/libudev/libudev-device.c):
udev_device->devpath = udev_device->syspath + strlen("/sys");
I used a command-line define to use the same logic, as it works with
new and old versions of the kernel--the use of ..devpath is quite
recent.
I also linked with libbsd, when using glibc.
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
(cherry picked from commit 633fe0dbdf)
This backports upstream fix from the master branch.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit 13d57a5e9f)
(cherry picked from commit 6e1104cc6d)
Having "select PACKAGE_zoneinfo-core" wasn't enough for builds without
php7-cli=y or php7-cgi=y. It didn't result in installing zoneinfo-core
when using "opkg install" (during runtime or when building images with
CONFIG_TARGET_PER_DEVICE_ROOTFS).
Missing zoneinfo results in PHP fatal errors, e.g.:
Fatal error: DateTime::createFromFormat(): Timezone database is corrupt - this should *never* happen!
For years users were told to manually install zoneinfo-core package.
This problem was hidden for some time (including 17.01 release) due to
disabled support for CONFIG_PHP7_SYSTEMTZDATA. It's now back as support
for --with-system-tzdata was enabled again.
The proper solution is to simply make php7 package depend on
zoneinfo-core when PHP7_SYSTEMTZDATA is used.
Fixes: 84e5012e88 ("php7: re-enable system timezone data usage")
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
This is in anticipation of the MariaDB upgrade from mariadb 10.1.x to
10.2.x. With the latter the PHP MySQL plugins fail to compile, e.g.:
In file included from /home/equeiroz/src/openwrt-asus/build_dir/target-mipsel_74kc_musl/php-7.2.6/ext/mysqli/php_mysqli_structs.h:63:0,
from /home/equeiroz/src/openwrt-asus/build_dir/target-mipsel_74kc_musl/php-7.2.6/ext/mysqli/mysqli.c:34:
/home/equeiroz/src/openwrt-asus/staging_dir/target-mipsel_74kc_musl/usr/include/mysql/my_global.h:3:2: warning: #warning This file should not be included by clients, include only <mysql.h> [-Wcpp]
#warning This file should not be included by clients, include only <mysql.h>
^~~~~~~
In file included from /home/equeiroz/src/openwrt-asus/build_dir/target-mipsel_74kc_musl/php-7.2.6/ext/mysqli/php_mysqli_structs.h:79:0,
from /home/equeiroz/src/openwrt-asus/build_dir/target-mipsel_74kc_musl/php-7.2.6/ext/mysqli/mysqli.c:34:
/home/equeiroz/src/openwrt-asus/staging_dir/target-mipsel_74kc_musl/usr/include/mysql/my_sys.h:3:2: warning: #warning This file should not be included by clients, include only <mysql.h> [-Wcpp]
#warning This file should not be included by clients, include only <mysql.h>
^~~~~~~
In file included from /home/equeiroz/src/openwrt-asus/build_dir/target-mipsel_74kc_musl/php-7.2.6/ext/mysqli/mysqli.c:34:0:
/home/equeiroz/src/openwrt-asus/build_dir/target-mipsel_74kc_musl/php-7.2.6/ext/mysqli/php_mysqli_structs.h:82:10: fatal error: my_list.h: No such file or directory
#include <my_list.h>
Upstream seems unlikely to fix this, see
https://bugs.php.net/bug.php?id=75612. In the bug report it is suggested
to not use libmysqlclient/libmariadbclient and instead utilize mysqlnd
(MySQL Native Driver) provided by PHP. This is the default anyway.
So add the mysqlnd module and remove the libmariadbclient depends.
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
AARCH64 compilation fails due to upstream bug in 1.2.0
that has been later fixed. Backport the fix commit.
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
(cherry picked from commit 0cb6e15f01)
This also changes PKG_SOURCE to use .tar.xz, and changes the copyright
line. (I believe this is more accurate, as I haven't done a copyright
assignment.)
Signed-off-by: Jeffery To <jeffery.to@gmail.com>
Defining _GLIBCXX_INCLUDE_NEXT_C_HEADERS instructs gcc 7.x libstdc++ to
include the system's stdlib.h and math.h, and not their own.
Signed-off-by: Eneas U de Queiroz <cote2004-github@yahoo.com>
(cherry picked from commit 054d5273cb)
Unbound UCI in OpenWrt 18.06 has limited forward configuration lacking
DNS over TLS connection setup tools. User override files
'unbound_srv.conf' and 'unbound_ext.conf' can implement this.
Signed-off-by: Eric Luehrsen <ericluehrsen@gmail.com>
Maybe we should use something other than -Os for this code anyway; it's
generally quite CPU-intensive.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Prometheus introduced some new conventions on how to name metrics.
Read here https://prometheus.io/docs/practices/naming/
This PR breaks compatibility with past versions, just like the officials
node exporter! 💥
Signed-off-by: Paul Spooren <mail@aparcar.org>
This updates GoPackage/Build/Compile in golang-package.mk to accept
additional arguments that are passed to the go command line.
Signed-off-by: Jeffery To <jeffery.to@gmail.com>
This also changes the bootstrap source URL to the official GitHub
mirror, to make building easier for places where Google sites aren't
accessible. Fixes#6326.
Signed-off-by: Jeffery To <jeffery.to@gmail.com>
The files in ozwcp/ shouldn't be compressed as there's no gzip handling
for those.
Also enable Python support — since it can dynamically link with
libpython optionally, it's harmless to enable it. Those who want Python
plugins can use it. I still want lua-based hardware plugins though.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Upstream has merged a simplified version of the FHS patch, with a few
changes...
Scripts are actually configuration. There are examples, but the point is
that you write your own.
So they should live in the data directory (e.g. /var/lib/domoticz) not
in /usr/share/domoticz. The only exception is the dzVents runtime.
So.... the upstream patch handles the dzVents runtime bit. Drop the part
of our patch which added -scripts, because it can just be based in the
userdata directory and we don't need to change that.
Ship the default scripts/ directory in /etc/domoticz/scripts, and on
startup make a *symlink* to it from /var/lib/domoticz/scripts.
Symlink from /etc/domoticz/scripts/dzVents{data,generated_scripts} to
temporary directories under /var/lib/domoticz/dzVents so that those
directories (which are written to by Domoticz) don't land on the root
file system. Anyone with a writeable file system who *wants* the data/
directory to be persistent, can change that. Just as they can change
the userdata config option to point to a real file system somewhere.
Also drop the renaming of the OpenZWave Config/ directory. It's purely
cosmetric so there's no need for us to carry that change. It can go
upstream first, if it really offends anyone.
Drop the patches which are now merged upstream, and turn off the newly
added USE_OPENSSL_STATIC. Add -noupdates to the command line.
Finally, gzip the static www files to save space. In the common case,
clients will use "Accept-Encodiong: gzip" and Domoticz will serve them
as-is. It can also decompress on the fly if it really has to, but now we
aren't asking it to *compress* on the fly, which is probably a losing
proposition on an OpenWRT box.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
This makes quite a significant difference to the executable size:
text data bss dec hex filename
7921421 87804 31692 8040917 7ab1d5 domoticz
5862321 86180 31212 5979713 5b3e41 domoticz-lto
As an added bonus, it still seems to work.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Make OLA more useful for untrained users which depend on the built-in
webserver. We may split the ola package into smaller parts to allow
not having web-stuff in case this breaks the space-constraints for some
users.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Recent protobuf requires C++11 while OLA was forcing C++98 in order
to keep using auto_ptr without getting warnings... Use gnu++11 to make
everyone happy and live with the warnings about auto_ptr being
deprecated.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Some releases may have non letters in it's name currently resulting in
an empty ("") output which is then discarded, resulting in *missing*
labels in the metric.
Now it uses `.-` to catch as little as possible, but anything.
Signed-off-by: Paul Spooren <mail@aparcar.org>
gptfdisk is a gpt-aware disk partitioning tool. It can be used to
convert mbr partitioned disk to gpt and vice versa.
It supports reading GPT, MBR, and BSD disklabels.
Signed-off-by: Alif M. Ahmad <alive4ever@live.com>
(cherry picked from commit eac2e12a6b)
From golang.org:
The Go programming language is an open source project to make
programmers more productive.
This commit consists of two "parts":
* golang/host: Main Go compiler for host (installed to
STAGING_DIR_HOST/lib/go-cross), used to cross-compile Go programs to
be packaged.
* golang (and golang-src/golang-doc): Main Go compiler for on-target
development. These packages are quite large, but I would expect only
developers to install these.
Signed-off-by: Jeffery To <jeffery.to@gmail.com>
Signed-off-by: Paul Spooren <mail@aparcar.org>
Update commit 3d59ce6f50 ("lighttpd: update to 1.4.48") resulted in
plain auth regression: it simply stopped working with:
(mod_auth.c.525) password doesn't match for (...)
appearing on every authentication try.
This regression was fixed in 1.4.49 release. Backport the fix instead of
updating to the 1.4.49 to avoid risking more/other regressions.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Commit dcd68100c23f980a7bbd1d7d9567a315ee584bdf fixed the zlib pkgconfig
file. But libxml2/host is stuck in the compile phase on the build bots.
Bumping the revision will force a clean build.
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
If an interface is not tracked by mwan3 or enabled and this interface is
setup by netifd, then the connected ipset is not update by mwan3.
To fix this also call connected ipset update code even if the interface
is not tracked or enabled by mwan3.
Signed-off-by: Florian Eckert <fe@dev.tdt.de>
(cherry picked from commit 7ea690ea3c)
Some resource options bundled many Unbound.conf options and
made customizing on top of UCI difficult. Make it easier to
use Unbound built defaults (blank conf sections).
Signed-off-by: Eric Luehrsen <ericluehrsen@hotmail.com>
Update to flashrom version 1.0 and cherry-pick an upstream fix for deprecated
libusb api usage.
Fixes the following error spotted by the buildbots:
ch341a_spi.c: In function 'ch341a_spi_init':
ch341a_spi.c:447:2: error: 'libusb_set_debug' is deprecated: [...]
libusb_set_debug(NULL, 3); // Enable information, warning and [...]
^~~~~~~~~~~~~~~~
In file included from ch341a_spi.c:25:0:
.../libusb-1.0/libusb.h:1300:18: note: declared here
void LIBUSB_CALL libusb_set_debug(libusb_context *ctx, int level);
^~~~~~~~~~~~~~~~
cc1: all warnings being treated as errors
Makefile:1030: recipe for target 'ch341a_spi.o' failed
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit cb6e98ba44)
The second source is just a redirect to SourceForge. Removed.
Also change URL to HTTPS as the website now supports it.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
(cherry picked from commit 54a35e465f)
nano 2.9.7 was released on 2018 May 15.
Release notes:
GNU nano 2.9.7 "Hvide Sande" adds the option '--afterends'
for making Ctrl+Right (the nextword function) stop at word
ends instead of beginnings, accepts multibyte letters for
the Yes/No/All answers, does emergency saves of changed
buffers in the unlikely event that nano crashes, adds the
until-now missing bindable function 'linenumbers', and
renames the toggles 'constupdate' to 'constantshow' and
'cuttoend' to 'cutfromcursor', for consistency with the
corresponding options -- adjust your nanorc files soon.
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
(cherry picked from commit d9e9b59159)
Only execute an ifup event with the command "mwan3 ifup <iface>" if the
l3_device is found in the ubus "network.interface.<iface>".
Signed-off-by: Florian Eckert <fe@dev.tdt.de>
(cherry picked from commit ab4e62b885)
Please make sure that the issue subject starts with `<package-name>: `
Also make sure that the package is maintained in this repository and not in base which should be submitted at https://bugs.openwrt.org or in the LuCI repository which should be submitted at https://github.com/openwrt/luci/issues.
This repo here is only for packages maintained in this repo. For base packages residing in the same repo as the build system and maintained by core devs, please consider opening tickets there for more timely responses
Issues related to releases below 18.06 and forks are not supported or maintained and will be closed.
- OpenWrt base system: https://bugs.openwrt.org
- Most LuCI packages: https://github.com/openwrt/luci/issues
# Issue template (remove lines from top till here)
(See <https://openwrt.org/docs/guide-developer/packages> for overall format and construction)
Ref: <https://openwrt.org/docs/guide-developer/packages> for overall format and construction
## Basic guidelines
### Basic guidelines
All packages you commit or submit by pull-request should follow these simple
guidelines:
All packages you commit or submit by pull-request should follow these simple guidelines:
* Package a version which is still maintained by the upstream author.
* Will be updated regularly to maintained and supported versions.
* Have no dependencies outside the OpenWrt core packages or this repository feed.
* Have been tested to compile with the correct includes and dependencies. Please also test with "Compile with full language support" found under "General Build Settings" set if language support is relevant to your package.
* Do NOT use a rolling source file (e.g. foo-latest.tar.gz) or the head of a branch as source for the package since that would create unpredictable builds which change over time.
* Best of all -- it works as expected!
- Package a version which is still maintained by the upstream author and will
be updated regularly with supported versions.
- Have no dependencies outside the OpenWrt core packages or this repository
feed.
- Have been tested to compile with the correct includes and dependencies.
Please also test with "Compile with full language support" found under
"General Build Settings" set if language support is relevant to your package.
- Best of all -- it works as expected!
#### Makefile contents should contain:
## Package Sources (archives and repositories)
* An up-to-date copyright notice. Use OpenWrt if no other present or supply your own.
* A (PKG_)MAINTAINER definition listing either yourself or another person in the field.
(E.g.: PKG_MAINTAINER:= Joe D. Hacker `<jdh@jdhs-email-provider.org`>)
* A PKG_LICENSE tag declaring the main license of the package.
(E.g.: PKG_LICENSE:=GPL-2.0+) Please use SPDX identifiers if possible (see list at the bottom).
* An optional PKG_LICENSE_FILES tag including the filenames of the license-files in the source-package.
(E.g.: PKG_LICENSE_FILES:=COPYING)
* PKG_RELEASE should be initially set to 1 or reset to 1 if the software version is changed. You should increment it if the package itself has changed. For example, modifying a support script, changing configure options like --disable* or --enable* switches, or if you changed something in the package which causes the resulting binaries to be different. Changes like correcting md5sums, changing mirror URLs, adding a maintainer field or updating a comment or copyright year in a Makefile do not require a change to PKG_RELEASE.
* Avoid reuse of PKG_NAME in call, define and eval lines to improve readability.
- PKG_SOURCE should reference the smallest available archive. In order of
preference: xz (most compressed), bzip2, gz and zip. As a last resort,
downloads from source repositories can be used.
- PKG_SOURCE_URL should link to an official release archive. Use of HTTPS:
is preferred. If a source archive is not available, a locally generated
archive fetched using git, svn, cvs or in rare circumstances, hg or bzr.
- Convenience macros for popular mirrors are defined. Using these macros will
make your package downloads more robust by mapping to a list of possible
source mirrors for archive availability.
- @SF - Sourceforge (downloads.sourceforge.net) with 5 retries due to
re-directs
- @GITHUB - Github (raw.githubusercontent.com) with 5 retries due to
re-directs
- @GNU - 8 regional servers
- @GNOME - 8 regional servers
- @SAVANNAH - 8 regional servers
- @APACHE - 8 regional servers
- @KERNEL - Linux kernel archives & mirrors
- Please _DO NOT_ use an archive which changes over time. A version labeled
"latest" is not constant each download. Also, using the head of a branch will
create unpredictable results which can be different each build.
#### Commits in your pull-requests should:
### Makefile contents should contain
* Have a useful description prefixed with the package name
(E.g.: "foopkg: Add libzot dependency")
* Include Signed-off-by in the comment
(See <https://openwrt.org/submitting-patches#sign_your_work>)
- Provide an up-to-date Copyright notice or **none**. Copyright should not be
assigned to OpenWrt unless you are explicitly requested by or working under
contract to OpenWrt. Assigning a Copyright to yourself or organization you
represent is acceptable.
- A (PKG\_)MAINTAINER definition listing either yourself and/or another person
responsible for this package (E.g.: PKG_MAINTAINER:= Joe D. Hacker
`<jdh@jdhs-email-provider.org`>). Listing multiple maintainers is encouraged in
order to keep the package active and up-to-date. Leaving this blank will also
be accepted, however the review process may not be as quick as one with a
maintainer.
- A PKG_LICENSE tag declaring the main license of the package. (E.g.:
PKG_LICENSE:=GPL-2.0-or-later) Please use SPDX identifiers if possible (see
list at the bottom).
- An optional PKG_LICENSE_FILES tag including the filenames of the
license-files in the source-package. (E.g.: PKG_LICENSE_FILES:=COPYING)
- PKG_RELEASE should be initially set to 1 or reset to 1 if the software
version is changed. You should increment it if the package itself has
changed. For example, modifying a support script, changing configure options
like --disable_ or --enable\_ switches, or if you changed something in the
package which causes the resulting binaries to be different. Changes like
correcting md5sums, changing mirror URLs, adding a maintainer field or updating
a comment or copyright year in a Makefile do not require a change to
PKG_RELEASE.
- Avoid reuse of PKG_NAME in call, define and eval lines to improve
readability.
### Advice on pull requests:
### Commits in your pull-requests should
Pull requests are the easiest way to contribute changes to git repos at Github. They are the preferred contribution method, as they offer a nice way for commenting and amending the proposed changes.
- Have a useful description prefixed with the package name (E.g.: "foopkg: Add
libzot dependency")
- Include Signed-off-by tag in the commit comments. See: [Sign your
Pull requests are the easiest way to contribute changes to git repos at Github.
They are the preferred contribution method, as they offer a nice way for
commenting and amending the proposed changes.
- You need a local "fork" of the Github repo.
- Use a "feature branch" for your changes. That separates the changes in the
pull request from your other changes and makes it easy to edit/amend commits
in the pull request. Workflow using "feature_x" as the example:
* You need a local "fork" of the Github repo.
* Use a "feature branch" for your changes. That separates the changes in the pull request from your other changes and makes it easy to edit/amend commits in the pull request. Workflow using "feature_x" as the example:
- Update your local git fork to the tip (of the master, usually)
- Create the feature branch with `git checkout -b feature_x`
- Edit changes and commit them locally
- Push them to your Github fork by `git push -u origin feature_x`. That
creates the "feature_x" branch at your Github fork and sets it as the
remote of this branch
- When you now visit Github, you should see a proposal to create a pull
request
- Push them to your Github fork by `git push -u origin feature_x`. That creates the "feature_x" branch at your Github fork and sets it as the remote of this branch
- When you now visit Github, you should see a proposal to create a pull request
- If you later need to add new commits to the pull request, you can simply
commit the changes to the local branch and then use `git push` to
automatically update the pull request.
* If you later need to add new commits to the pull request, you can simply commit the changes to the local branch and then use `git push` to automatically update the pull request.
- If you need to change something in the existing pull request (e.g. to add a
missing signed-off-by line to the commit message), you can use `git push -f`
to overwrite the original commits. That is easy and safe when using a feature
branch. Example workflow:
* If you need to change something in the existing pull request (e.g. to add a missing signed-off-by line to the commit message), you can use `git push -f` to overwrite the original commits. That is easy and safe when using a feature branch. Example workflow:
- Checkout the feature branch by `git checkout feature_x`
- Edit changes and commit them locally. If you are just updating the commit
message in the last commit, you can use `git commit --amend` to do that
- If you added several new commits or made other changes that require
cleaning up, you can use `git rebase -i HEAD~X` (X = number of commits to
edit) to possibly squash some commits
- Push the changed commits to Github with `git push -f` to overwrite the
original commits in the "feature_x" branch with the new ones. The pull
request gets automatically updated
- Edit changes and commit them locally. If you are just updating the commit message in the last commit, you can use `git commit --amend` to do that
- If you added several new commits or made other changes that require cleaning up, you can use `git rebase -i HEAD~X` (X = number of commits to edit) to possibly squash some commits
- Push the changed commits to Github with `git push -f` to overwrite the original commits in the "feature_x" branch with the new ones. The pull request gets automatically updated
## If you have commit access
### If you have commit access:
- Do NOT use git push --force.
- Do NOT commit to other maintainer's packages without their consent.
- Use Pull Requests if you are unsure and to suggest changes to other
maintainers.
* Do NOT use git push --force.
* Do NOT commit to other maintainer's packages without their consent.
* Use Pull Requests if you are unsure and to suggest changes to other maintainers.
### Gaining commit access
#### Gaining commit access:
- We will gladly grant commit access to responsible contributors who have made
* We will gladly grant commit access to responsible contributors who have made
useful pull requests and / or feedback or patches to this repository or
OpenWrt in general. Please include your request for commit access in your next
pull request or ticket.
OpenWrt in general. Please include your request for commit access in your
next pull request or ticket.
## Release Branches
### Release Branches:
- Old stable branches were named after the following pattern "for-XX.YY" (e.g.
for-14.07) before the LEDE split. During the LEDE split there was only one
release branch with the name "lede-17.01". After merging the LEDE fork with
OpenWrt the release branches are named according to the following pattern
"openwrt-XX.YY" (e.g. openwrt-18.06).
- These branches are built with the respective OpenWrt release and are created
* Branches named "for-XX.YY" (e.g. for-14.07) are release branches.
* These branches are built with the respective OpenWrt release and are created
during the release stabilisation phase.
- Please ONLY cherry-pick or commit security and bug-fixes to these branches.
- Do NOT add new packages and do NOT do major upgrades of packages here.
- If you are unsure if your change is suitable, please use a pull request.
* Please ONLY cherry-pick or commit security and bug-fixes to these branches.
* Do NOT add new packages and do NOT do major upgrades of packages here.
* If you are unsure if your change is suitable, please use a pull request.
## Common LICENSE tags (short list)
### Common LICENSE tags (short list)
(Complete list can be found at: <http://spdx.org/licenses>)
(Complete list can be found at: <https://spdx.org/licenses>)
COMMAND if grep -q "Welcome to ${CMAKE_PROJECT_NAME} ${GIT_RELEASE_VERSION}" "${PROJECT_SOURCE_DIR}/NEWS" \; then echo "NEWS current version ${GIT_RELEASE_VERSION}" \; else echo "NEWS current version does not match release version ${GIT_RELEASE_VERSION}" \; exit 1 \; fi