- Introduce hash table improvement (by Vladimir Makarov)
- Binding#irb: Start a REPL session similar to binding.pry
- Unify Fixnum and Bignum into Integer
- String supports Unicode case mappings
- Performance improvements
- Thread#report_on_exception and Thread.report_on_exception changes
- Thread deadlock detection now shows threads with their backtrace and dependency
- Support OpenSSL 1.1.0 (drop support for 0.9.7 or prior)
- ext/tk is now removed from stdlib Feature #8539
- XMLRPC is now removed from stdlib Feature #12160
Signed-off-by: Luiz Angelo Daros de Luca <redacted>
ascob [Wed, 5 Oct 2016 16:24:19 +0000 (18:24 +0200)]
dynapoint: add new package (fixes #3304)
Dynapoint is a dynamic access point manager
Signed-off-by: Tobias Ilte <redacted> Acked-by: Thomas Huehn <redacted>
- minor whitespace fixes/cleanups
- squashed multiple commits into a single one before merge
Applying ./patches/001-sigval-ref-fix.patch using plaintext:
can't find file to patch at input line 3
Perhaps you used the wrong -p or --strip option?
The text leading up to this was:
--------------------------
|--- a/drivers/timers_unix/timers_unix.c
|+++ b/drivers/timers_unix/timers_unix.c
--------------------------
No file to patch. Skipping patch.
1 out of 1 hunk ignored
Patch failed! Please fix ./patches/001-sigval-ref-fix.patch!
...
-snap-
Reason is, that the tar ball created by hg checkout does not contain the
version string appended to the root source directory as expected by
default PKG_BUILD_DIR, so this patch adjusts the expected directory name.
Hauke Mehrtens [Sun, 25 Dec 2016 16:43:31 +0000 (17:43 +0100)]
libcoap: fix PKG_BUILD_DIR
libcoap is not building in LEDE any more because it is extracted to
/libcoap/ and the package makefiles try to access it at
/libcoap-2da31de732c0e51a9bc9e1d4aea21e25da89cf87/.
This patch defines a PKG_BUILD_DIR variable to make it use /libcoap/
without the hash.
Hauke Mehrtens [Sat, 17 Dec 2016 16:52:32 +0000 (17:52 +0100)]
boost: build always with -std=gnu++14 with recent compilers.
Some parts of boost need at least C++11, activate -std=gnu++14
when possible and otherwise use -std=gnu++11 instated of using the
default which is -std=gnu++98.
This fixes the build for me when I build everything expect
boost-coroutine2 with gcc 5.4.0 on LEDE.
ddns-scripts: no longer build public_suffix_list.dat.gz during build
- no longer build public_suffix_list.dat.gz during build #3678
- replace "\s" with "[[:space:]]" inside Makefile because "\s" ignored by some sed versions
- tools/public_suffix_list.sh still available to rebuild public_suffix_list.dat.gz outside OpenWRT/LEDE build system
ddns-scripts: get public_suffix_list.dat without using secure connection
Get public_suffix_list.dat without using secure connection. #3678
File generated during build, because it's the only option to have an
actual version packaged.
Long term Cloudflare_v1 package will be changed to no longer need
public_suffix_list.dat
Description: Bump to 1.4-5. Added new page for wifi configuration edit, similar how the current network configuration page works.
Also enabled collecting of wifi configuration in troubleshooting page.
danrl [Tue, 20 Dec 2016 10:06:56 +0000 (11:06 +0100)]
wireguard: add protocol dependency for endpoints
Endpoint dependency implemented. The actual endpoint is used exclusively. Using
this approach we are dual-stack safe (not errors on missing protocol) and create
only the dependency that are really necessary.
Dirk Brenken [Thu, 22 Dec 2016 18:33:59 +0000 (19:33 +0100)]
adblock: bugfix 2.0.1
* fixed a dnsmasq restart issue (udhcpc error)
* fixed a long standing corner case bug in "disabled" state (does not
remove active block lists!)
* simplified overall sort, removed needless 'for loop'
* cosmetics
David Thornley [Thu, 22 Dec 2016 07:11:38 +0000 (18:11 +1100)]
luarocks: Fix seeded uname (s/m) in siteconfig
LUAROCKS_UNAME_S/LUAROCKS_UNAME_M were being set automatically to host’s architecture and system/kernel name (e.g.. x86 and Darwin for the macOS)
The variables can now be supplied to configure via the supplied patch.
They are currently _S being hardcoded to “Linux” and _M set to the target architecture in the MakeFile.
Dirk Brenken [Mon, 19 Dec 2016 19:11:25 +0000 (20:11 +0100)]
adblock: release 2.0.0
* changed complete dnsmasq handling
- no longer redirect ad requests to internal web server via firewall
rules etc., in fact send back a simple NXDOMAIN for all
ad related (sub-)domains
- smaller memory footprint for block lists
- removed needless uhttpd instances
- removed needless firewall rules/redirects
* init/hotplug system migrated to procd
- removed hotplug.d script, now using procd interface trigger
* reduced code size/complexity
- removed needless internal pre-checks & function blocks,
no longer rely on a separate helper library
- removed flash writes to adblock config
* support different download tools like wget (default), aria2c,
uclient-fetch, curl (see online doc)
* adblock status/statistics via ubus call (see online doc)
* various bug fixes
* documentation update
* changed makefile copyright notice
Eric Luehrsen [Mon, 19 Dec 2016 03:38:03 +0000 (22:38 -0500)]
unbound: UCI updates to take advantage of 1.6.0
- UCI to take advantage of "qname-minimisation-strict:"
- UCI to block chaos reponses bind, server, and version
- UCI to limit or prefer recrusion over IP4 or IP6
- update_cloudflare_com_v4.sh: enclose urls in single quotes
- dynamic_dns_functions.sh: force to return only one ip, if using nslookup to get registered ip
Update to new snapshot version. We also make IPV6 optional, and
conditionally selecting the udptunnel6 module, using the same trick that
the strongswan package also uses for this kind of dependency expression.
Jo-Philipp Wich [Fri, 16 Dec 2016 01:22:41 +0000 (02:22 +0100)]
python: move includes after common package variables
Move the python-host.mk and python-package.mk includes after the definition of
common package variables. This is required to ensure that PKG_UNPACK is not set
to to the dummy "true" command which happens if PKG_SOURCE is not yet set.
Fixes the following error observed while attempting to build Python on a recent
LEDE version:
Applying ./patches/001-enable-zlib.patch using plaintext:
can't find file to patch at input line 14
Perhaps you used the wrong -p or --strip option?
[...]
Patch failed! Please fix ./patches/001-enable-zlib.patch!
Makefile:242: recipe for target '.../Python-2.7.12/.prepared_...' failed
Jo-Philipp Wich [Fri, 16 Dec 2016 00:14:54 +0000 (01:14 +0100)]
openswitch: improve build time for kmod-only builds
When only kmod-openvswitch is enabled, which commonly happens during LEDE
release builds, there is no need to build the entire userland of openvswitch
as this is done at a later stage on systems dedicated to build only userspace
packages.
This change conditionalizes the dependency on python, which means that python
prerequisites are only compiled if the openvswitch-python package is enabled,
which allows us to eliminate the entire python dependency chain.
Furthermore, this change sets MAKE_PATH to the kernel module sub directory
if only kmod-openvswitch is enabled which causes the openvswitch build system
to only process sources related to the .ko files.
These changes are a direct copy of [1]. The original commit
messages were:
> commit 0a6411b56b9edeba1809ffe45c03dbb7261bf45c
> Author: Matthias Schiffer <redacted>
> Date: Sun May 8 11:39:22 2016 +0200
>
> ecdsautils: update to current git, add libecdsautil
>
> The addition of libecdsautil required us to switch from statically
> linked libuecc to a shared library (as we also want shared
> libecdsautil).
Yousong Zhou [Tue, 6 Dec 2016 06:52:37 +0000 (14:52 +0800)]
qemu: packaging target {x86_64,arm}-softmmu and friends
At the moment, only build these softmmu emulators for x86_64 and sunxi
target. The decision was made for the following reasons
- It seems that interests of virtualization with qemu are mostly from
x86, and ARM recently.
- x86, sunxi boards/boxes capable of running qemu with accel=kvm are more
widely available
- Not all host, target combinations of qemu works, or even compiles
- Extra maintenance work and server resources
Test results are as the following
- Nested vmx works: lede-qemu-x86_64-kvm on lede-qemu-x86_64-kvm
- KVM on Cubieboard2 works
- tcg with malta works: lede-qemu-malta-tcg on lede-qemu-malta-tcg.
But it's too slow to be useful thus not included in this version
- mips64 host does not compile
git.99rst.org / openwrt-packages.git / log