John Crispin [Tue, 16 Sep 2014 18:35:15 +0000 (18:35 +0000)]
ubox: update to latest git HEAD
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 42582
John Crispin [Tue, 16 Sep 2014 18:35:09 +0000 (18:35 +0000)]
procd: update to latest git HEAD
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 42581
John Crispin [Tue, 16 Sep 2014 18:35:05 +0000 (18:35 +0000)]
Lantiq: fix build warning from dsl_fs script
fix inclusion of functions.sh in dsl_fs init script
without this, the following error is seen during build:
./etc/init.d/dsl_fs: line 4: /lib/functions.sh: No such file or directory
Signed-off-by: Thomas Langer <thomas.langer@lantiq.com>
Backport of r42575
SVN-Revision: 42580
John Crispin [Tue, 16 Sep 2014 18:35:03 +0000 (18:35 +0000)]
fs-tools: dont use mips16
this causes problems on 47xx/74k
Signed-off-by: John Crispin <blogic@openwrt.org>
Backport of r42570
SVN-Revision: 42579
John Crispin [Tue, 16 Sep 2014 18:35:00 +0000 (18:35 +0000)]
lantiq: fix switch init on vr9 SoC (3.10)
Signed-off-by: John Crispin <blogic@openwrt.org>
Backport of r42544
Backport of r42551
Backport of r42574
SVN-Revision: 42578
John Crispin [Tue, 16 Sep 2014 18:34:44 +0000 (18:34 +0000)]
comgt: add directip proto
add a proto handler for sierra wireless directip modems
Signed-off-by: John Crispin <blogic@openwrt.org>
Backport of r42465
SVN-Revision: 42577
Felix Fietkau [Mon, 15 Sep 2014 16:09:36 +0000 (16:09 +0000)]
hostapd: remove bogus default setting for wps_pin (#17873)
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r42553
SVN-Revision: 42554
Felix Fietkau [Thu, 11 Sep 2014 15:14:05 +0000 (15:14 +0000)]
ar71xx: ack completed tx descriptors only after the full frame has been completed
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r42457
SVN-Revision: 42475
Felix Fietkau [Wed, 10 Sep 2014 13:50:53 +0000 (13:50 +0000)]
atheros: enable the low_mem flag to better deal with 16M RAM devices
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r42460
SVN-Revision: 42461
Felix Fietkau [Wed, 10 Sep 2014 12:57:07 +0000 (12:57 +0000)]
procd: allow adding empty strings to arrays
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r42456
SVN-Revision: 42458
Felix Fietkau [Tue, 9 Sep 2014 08:03:34 +0000 (08:03 +0000)]
ath9k: fix pll clock initialization on newer soc devices (fixes #14916)
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r42453
SVN-Revision: 42454
Felix Fietkau [Mon, 8 Sep 2014 16:45:58 +0000 (16:45 +0000)]
ath9k: add some ani related stability / performance fixes
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r42443
SVN-Revision: 42444
John Crispin [Mon, 8 Sep 2014 15:14:03 +0000 (15:14 +0000)]
mac80211: add mt7620a/n to the rt2x00 depends list
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 42441
Steven Barth [Mon, 8 Sep 2014 14:47:26 +0000 (14:47 +0000)]
netifd: work-around kernel IPv6 on-link route issue
SVN-Revision: 42440
John Crispin [Mon, 8 Sep 2014 14:30:00 +0000 (14:30 +0000)]
lantiq: add ath9k led pin support
Signed-off-by: John Crispin <blogic@openwrt.org>
Backport of r42437
SVN-Revision: 42438
Felix Fietkau [Sun, 7 Sep 2014 09:46:41 +0000 (09:46 +0000)]
ath79: dev-eth: Don't advertise 1gbit in link code word on ar9331
While the AR9331 has a gigabit MAC towards the internal switch, the
integrated PHYs however are only 100-base-tx capable. The existing code
however advertieses gigabit capability in the link status word. If you
attach such a PHY to a gigabit capable switch on the remote end, with
some probability it attempts to negotiate gigabit and fails, falling
baco to the AR9331 assuming a 10mbit half-duplex link. This has been
observed quite frequently with the Carambola2 and gigabit capable
switches.
In ath79_register_eth(), "pdata->has_gbit = 1;" is set unconditionally
for both AR9331 ethernet ports. This is most likely wrong. Despite the
two MAC IP cores being gigabit MACs, the MAC for eth1 is connected to a
100base-T PHY via MII. The has_gbit attribute is used in the ethernet
driver to determine the supported link modes.
So either pdata->has_gbit is not set to 1 anymore, or the ethernet
driver needs to be modified to determine the advertised link code word
on another criteria than pdata->has_gbit. This patch implements the
former solution.
Signed-off-by: Harald Welte <laforge@gnumonks.org>
Backport of r42432
SVN-Revision: 42434
Felix Fietkau [Sun, 7 Sep 2014 09:46:39 +0000 (09:46 +0000)]
ar71xx: merge ethernet tx descriptor splitting from trunk (as of r42429) to fix ar716x performance issues
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 42433
Felix Fietkau [Thu, 4 Sep 2014 17:46:39 +0000 (17:46 +0000)]
uclient: update to the latest version, fixes some crash issues
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r42419
SVN-Revision: 42420
Felix Fietkau [Tue, 2 Sep 2014 21:06:59 +0000 (21:06 +0000)]
mtd: disable trx_check() for ar71xx, fixes a regression introduced in r42403
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r42405
SVN-Revision: 42406
Felix Fietkau [Tue, 2 Sep 2014 17:14:24 +0000 (17:14 +0000)]
mtd: Fix trx check after partition rename ("linux" to "firmware")
On brcm47xx (Huawei E970), I noticed that sysupgrade now tries to flash trx images which don't fit into the flash, resulting in non-booting system.
I found that mtd does trx size checking only when flashing to a partition called "linux". This patch changes this to be "firmware".
This matches the following patch which changed partition name in bcm47xxpart.c driver to "firmware":
commit
86b4d5ef68ca77ff6724ebb9bddd6b93239c87fc
Author: hauke <hauke@
3c298f89-4303-0410-b956-
a3cf2f4a3e73>
Date: Sun Dec 1 16:32:32 2013 +0000
brcm47xx: use "firmware" partition name
...
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Backport of r38973
Signed-off-by: Mathias Adam <m.adam--openwrt@adamis.de>
Backport of r42403
SVN-Revision: 42404
Felix Fietkau [Tue, 2 Sep 2014 17:12:34 +0000 (17:12 +0000)]
cns3xxx: support isolated PCI interrupts on newer Laguna PCBs
The cns3xxx uses irq61 for pcie0_intr which in the case of a PCIe-to-PCI
bridge ends up combining INTA/B/C/D on a single ARM CPU interrupt. This is
not optimal when you have multiple cores. To overcome this limitation an
enhancement was made on newer Laguna PCB's that support miniPCI cards
to route the INTA/B/C/D signals to unique external ARM CPU interrupts which
can help balance CPU core utilization and in some cases increase overall
system performance or responsiveness.
For more details see:
http://trac.gateworks.com/wiki/multicoreprocessing#PCIInterruptsteering
Signed-off-by: Tim Harvey <tharvey@gateworks.com>
Backport of r42400
SVN-Revision: 42402
John Crispin [Tue, 2 Sep 2014 15:51:05 +0000 (15:51 +0000)]
BB: point old.packages at the real tree
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 42399
Felix Fietkau [Mon, 1 Sep 2014 22:34:49 +0000 (22:34 +0000)]
sdk: keep $(STAGING_DIR)/usr/bin, at least curl needs it
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r42397
SVN-Revision: 42398
John Crispin [Mon, 1 Sep 2014 13:22:16 +0000 (13:22 +0000)]
mac80211: rt2x00 correctly set ht20/ht40 filter
Simple patch to correct ht20/ht40 switching. Tested with rt3290/rt3352/rt5350
Signed-off-by: Serge Vasilugin <vasilugin@yandex.ru>
Backport of r42348
SVN-Revision: 42396
John Crispin [Mon, 1 Sep 2014 13:22:14 +0000 (13:22 +0000)]
ramips: fix switch config for Edimax 3G-6200n
Switch in Edimax 3G-6200n also require full switch reset, not only vlan definitions. Tested on Edimax 3G-6200n.
Signed-off-by: Cezary Jackiewicz <cezary.jackiewicz@gmail.com>
Backport of r42194
SVN-Revision: 42395
John Crispin [Mon, 1 Sep 2014 13:22:13 +0000 (13:22 +0000)]
ramips: Hame: change LED name mpr-x -> hame
Signed-off-by: Cezary Jackiewicz <cezary.jackiewicz@gmail.com>
Backport of r42190
SVN-Revision: 42394
John Crispin [Mon, 1 Sep 2014 13:22:12 +0000 (13:22 +0000)]
add support for the NC-LINK SMART-300
Signed-off-by: Imre Kaloz <kaloz@openwrt.org>
Backport of r42130
SVN-Revision: 42393
John Crispin [Mon, 1 Sep 2014 13:22:10 +0000 (13:22 +0000)]
Add 'subversion' build dependency to README
Fixes #17725.
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
Backport of r42343
SVN-Revision: 42392
John Crispin [Mon, 1 Sep 2014 13:22:09 +0000 (13:22 +0000)]
ralink: DIR-645 picks up incorrect MAC address in DTS file
Signed-off-by: John Crispin <blogic@openwrt.org>
Backport of r42321
SVN-Revision: 42391
John Crispin [Mon, 1 Sep 2014 13:22:08 +0000 (13:22 +0000)]
dnsmasq: Make the --dhcp-host logic easier to understand
Use an if/else statement to cover the two different syntaxes. Add
comments explaining what the end results should look like.
This patch should not change the script's output.
Signed-off-by: Kevin Cernekee <cernekee@gmail.com>
Backport of r42320
SVN-Revision: 42390
John Crispin [Mon, 1 Sep 2014 13:22:06 +0000 (13:22 +0000)]
dnsmasq: Fix hosts file format when MAC address is not specified
An entry like this in /etc/config/dhcp:
config 'host'
option 'name' 'pc2'
option 'ip' '192.168.100.56'
option 'dns' '1'
results in a /tmp/hosts/dhcp entry that looks like this:
192.168.100.56 .lan
Obviously it should say "pc2.lan".
This happens because $name is set to "" in order to support the MAC-less
syntax: "--dhcp-host=lap,192.168.0.199". Fix this by reordering the
operations. Also, refuse to add a DNS entry if the hostname or IP is
missing.
Fixes #17683
Reported-by: Kostas Papadopoulos <kpapad75@travelguide.gr>
Signed-off-by: Kevin Cernekee <cernekee@gmail.com>
Backport of r42319
SVN-Revision: 42389
John Crispin [Mon, 1 Sep 2014 13:22:05 +0000 (13:22 +0000)]
dnsmasq: Create rDNS records for LuCI "Hostnames"
LuCI creates "domain" UCI config sections, which the dnsmasq init file
then, currently, translates into "address" config lines. This is not
the correct usage of "address" (see r36943), and also causes rDNS
records to not be created. This patches dnsmasq.init to utilize the
additional hosts file introduced in r40799 for such domain names,
resolving both issues.
Signed-off-by: Tyler Fenby <tylerf@securecominc.com>
Backport of r42318
SVN-Revision: 42388
John Crispin [Mon, 1 Sep 2014 13:22:04 +0000 (13:22 +0000)]
ar71xx: add Support for RB2011UiAS-2HnD
Backport of r42317
SVN-Revision: 42387
John Crispin [Mon, 1 Sep 2014 13:22:02 +0000 (13:22 +0000)]
lantiq: BTHOMEHUBV2B use bigger mtd partition for kernel
The bb-rc3 image for the BTHOMEHUBV2B is too big for its
mtd partition. This patch corrects the partition sizes in
the device tree. This patch should really go in before
bb-final, otherwise the BTHOMEHUBV2B images won't be useable.
I do apologise for not spotting this straight away.
Many thanks,
Ben
Signed-off-by: Ben Mulvihill <ben.mulvihill@gmail.com>
Backport of r42316
SVN-Revision: 42386
John Crispin [Mon, 1 Sep 2014 13:22:01 +0000 (13:22 +0000)]
ralink: the mt7620 evb has a usb port
Signed-off-by: John Crispin <blogic@openwrt.org>
Backport of r42293
SVN-Revision: 42385
John Crispin [Mon, 1 Sep 2014 13:22:00 +0000 (13:22 +0000)]
ralink: fix mt7620 ohci 3.10
the ohci phy was not reset properly
Signed-off-by: John Crispin <blogic@openwrt.org>
Backport of r42290
SVN-Revision: 42384
John Crispin [Mon, 1 Sep 2014 13:21:59 +0000 (13:21 +0000)]
ralink: mt7620 has obchi/ehci and not dwc2
fix the default package selection
Signed-off-by: John Crispin <blogic@openwrt.org>
Backport of r42289
SVN-Revision: 42383
John Crispin [Mon, 1 Sep 2014 13:21:57 +0000 (13:21 +0000)]
mac80211/rt2x00: allow disabling bands through dts
Add dts property ralink,2ghz and ralink,5ghz
e.g. disable 2ghz band:
wmac@
10180000 {
status = "okay";
ralink,2ghz = <0>;
}
Signed-off-by: Alexander Couzens <lynxis@fe80.eu>
Backport of r42281
SVN-Revision: 42382
John Crispin [Mon, 1 Sep 2014 13:21:56 +0000 (13:21 +0000)]
ar71xx: use ETH_ALEN for consistency
Signed-off-by: Forest Crossman <cyrozap@gmail.com>
Backport of r42280
SVN-Revision: 42381
John Crispin [Mon, 1 Sep 2014 13:21:55 +0000 (13:21 +0000)]
ar71xx: enable sysupgrade for the EnGenius ESR900
Signed-off-by: Forest Crossman <cyrozap@gmail.com>
Backport of r42279
SVN-Revision: 42380
John Crispin [Mon, 1 Sep 2014 13:21:54 +0000 (13:21 +0000)]
package/*: remove useless explicit set of function returncode
somebody started to set a function returncode in the validation
stuff and everybody copies it, e.g.
myfunction()
{
fire_command
return $?
}
a function automatically returns with the last returncode,
so we can safely remove the command 'return $?'. reference:
http://tldp.org/LDP/abs/html/exit-status.html
"The last command executed in the function or script determines the exit status."
Signed-off-by: Bastian Bittorf <bittorf@bluebottle.com>
Backport of r42278
SVN-Revision: 42379
John Crispin [Mon, 1 Sep 2014 13:21:52 +0000 (13:21 +0000)]
ramips: enable second SPI for VoCore
This patch enables the second SPI as spidev and also fixes some typos.
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
Backport of r42277
SVN-Revision: 42378
John Crispin [Mon, 1 Sep 2014 13:21:51 +0000 (13:21 +0000)]
ramips: Add support for SPI_CS1 pinmux
This patch adds support for setting SPI_CS1 as Chip Select, Watchdog reset output and GPIO#27.
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
Backport of r41938
SVN-Revision: 42377
John Crispin [Mon, 1 Sep 2014 13:21:49 +0000 (13:21 +0000)]
ramips: Add support for VoCore
This is based and tested on VoCore Alpha, but other stuff like status/eth LEDs are present on the final board revision + VoDock.
All GPIOs are exported, except spi/i2c.
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
Backport of r41939
SVN-Revision: 42376
John Crispin [Mon, 1 Sep 2014 13:21:48 +0000 (13:21 +0000)]
ramips: second spi device on rt5350
This is based on Jon Smirl's patch with the following changes:
- Set CS polarity as low by default.
- Add support for changing CS polarity.
- Add support for changing LSB/MSB.
- Add support for changing SPI mode.
- Fix indentations.
I tested it on a VoCore. Works fine connected to a second flash, but fails to detect MMC/SD cards due to SPI clock speed.
Signed-off-by: Jon Smirl <jonsmirl@gmail.com>
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
Backport of r42276
SVN-Revision: 42375
John Crispin [Mon, 1 Sep 2014 13:21:46 +0000 (13:21 +0000)]
firewall: the firewall did not start properly on boot
https://dev.openwrt.org/ticket/17593
Signed-off-by: John Crispin <blogic@openwrt.org>
Backport of r42233
SVN-Revision: 42374
John Crispin [Mon, 1 Sep 2014 13:21:45 +0000 (13:21 +0000)]
libubox: update to latest git revision
Signed-off-by: John Crispin <blogic@openwrt.org>
Backport of r42230
SVN-Revision: 42373
John Crispin [Mon, 1 Sep 2014 13:21:44 +0000 (13:21 +0000)]
valgrind: fix build failure on uml
Signed-off-by: Nicolas Thill <nico@openwrt.org>
Backport of r42229
SVN-Revision: 42372
John Crispin [Mon, 1 Sep 2014 13:21:42 +0000 (13:21 +0000)]
brcm47xx: image: build alternative TRX using less optimized LZMA
There is a group of devices that lzma-loader doesn't work with. They
simply hang at "Starting program at 0x80001000" which is really hard to
debug and we didn't find any solution for this for years.
Broadcom doesn't use lzma-loader on these devices anyway. They decided
to drop lzma-loader and use less optimal LZMA compression that can be
handled by CFE itself (it doesn't use dictionary).
So support these devices we will need kernel compressed with different
parameters and trx without a loader.
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Backport of r42205
SVN-Revision: 42371
John Crispin [Mon, 1 Sep 2014 13:21:40 +0000 (13:21 +0000)]
ramips: Add support for Omnima MiniPlug
This patch adds support for the Omnima MiniPlug RT5350 based board, 8MB SPI flash
Signed-off-by: Ivan Ignjatic<ivan@omnima.co.uk>
Backport of r42200
SVN-Revision: 42370
John Crispin [Mon, 1 Sep 2014 13:21:38 +0000 (13:21 +0000)]
gmp: add host compile
Currently, any package that uses host compile and depends on
libgmp.so will fail. This is because gmp is not compile for
host machine. So, staging_dir/host has only static lib for gmp
$ ls staging_dir/host/lib/libgmp* -1
staging_dir/host/lib/libgmp.a
staging_dir/host/lib/libgmp.la
staging_dir/host/lib/libgmpxx.a
staging_dir/host/lib/libgmpxx.la
Addind host compile in gmp, the dependent package can use:
PKG_BUILD_DEPENDS:=gmp/host
That will compile gmp to staging_dir/host with *.so files
Signed-off-by: Luiz Angelo Daros de Luca <luizluca@gmail.com>
Backport of r42196
SVN-Revision: 42369
John Crispin [Mon, 1 Sep 2014 13:21:36 +0000 (13:21 +0000)]
busybox: Fix pam_start link errors when PAM support is enabled
Patch to BusyBox Makefile to fix linker errors when PAM support is enabled by
CONFIG_BUSYBOX_CONFIG_PAM=y. Typical errors include:
login.c:(.text.login_main+0x122): undefined reference to `pam_start'
Signed-off-by: Stephen Parry <sgparry@mainscreen.com>
Backport of r42195
SVN-Revision: 42368
John Crispin [Mon, 1 Sep 2014 13:21:34 +0000 (13:21 +0000)]
ar71xx: TL-WR1043N/ND v2: Fix MAC address of WMAC
This sets the MAC address of the WLAN interface to the "official" primary MAC
address (the one on the label under the devices, and the one used with the stock
firmware). The MAC address used so far (primary-1) isn't even used at all with
the stock firmware, which sets (primary) on LAN and WLAN and (primary+1) on the
WAN interface (like OpenWrt does with this patch).
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Backport of r42193
SVN-Revision: 42367
John Crispin [Mon, 1 Sep 2014 13:21:32 +0000 (13:21 +0000)]
ralink: clean up HAME profiles, remove unnecessary packages.
Signed-off-by: Cezary Jackiewicz <cezary.jackiewicz@gmail.com>
Backport of r42192
SVN-Revision: 42366
John Crispin [Mon, 1 Sep 2014 13:21:31 +0000 (13:21 +0000)]
ramips: add mpr-a1 to default profile
Add hame mpr-a1 to default ramips profile.
Signed-off-by: Cezary Jackiewicz <cezary.jackiewicz@gmail.com>
Backport of r42191
SVN-Revision: 42365
John Crispin [Mon, 1 Sep 2014 13:21:29 +0000 (13:21 +0000)]
ar71xx: build factory image for EnGenius ESR900 and ESR1750
Signed-off-by: Forest Crossman <cyrozap@gmail.com>
Backport of r42189
SVN-Revision: 42364
John Crispin [Mon, 1 Sep 2014 13:21:27 +0000 (13:21 +0000)]
ar71xx: Fix 2.4G WiFi LED of the OpenMesh MR600 board
The OpenMesh MR600(v1) can only enable the 2.4G WiFi PHY LED through the
mini-PCIe device. Not configuring the LED pin inside the platform data
makes it impossible to configure it through any standard OpenWrt tool.
Signed-off-by: Sven Eckelmann <sven@open-mesh.com>
Backport of r42184
SVN-Revision: 42363
John Crispin [Mon, 1 Sep 2014 13:21:25 +0000 (13:21 +0000)]
lantiq: generate unique ARV7519RW22 firmware
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
Backport of r42183
SVN-Revision: 42362
John Crispin [Mon, 1 Sep 2014 13:21:23 +0000 (13:21 +0000)]
lantiq: add support for xrx200 firmware depending on soc type
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
Backport of r42182
SVN-Revision: 42361
John Crispin [Mon, 1 Sep 2014 13:21:21 +0000 (13:21 +0000)]
lantiq: export soc type
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
Backport of r42181
SVN-Revision: 42360
John Crispin [Mon, 1 Sep 2014 13:21:19 +0000 (13:21 +0000)]
lantiq: improve ARV7519RW22 support
- use full board name
- rename uboot-env partition
- add dsl_fw partition
- remove unneeded pinmux groups
- move gigabit ethernet to LAN
- load mac address from mtd
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
Backport of r42180
SVN-Revision: 42359
John Crispin [Mon, 1 Sep 2014 13:21:17 +0000 (13:21 +0000)]
ramips: add a workaround to fix failsafe on RT5350
Due to TCP connections not working when VLAN is disabled, this is
needed to get failsafe functional.
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Backport of r42179
SVN-Revision: 42358
John Crispin [Mon, 1 Sep 2014 13:21:15 +0000 (13:21 +0000)]
ramips: enable additional GPIOs on HAME MPR-A1
According to the pcb tracing results[1] by anton.rad[2] MPR-A1s expose
6 unused GPIOs, only one of them working as configured in the current
DTS. This patch enables GPIO22-26.
Tested on hardware.
[1] http://i.imgur.com/kHVW2Ox.jpg
[2] https://forum.openwrt.org/viewtopic.php?pid=222698#p222698
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Backport of r42178
SVN-Revision: 42357
John Crispin [Mon, 1 Sep 2014 13:21:13 +0000 (13:21 +0000)]
ar71xx: changed the EnGenius ESR900 mtdparts to match that of the ESR1750 since they have the same flash layout
Signed-off-by: Forest Crossman <cyrozap@gmail.com>
Backport of r42177
SVN-Revision: 42356
John Crispin [Mon, 1 Sep 2014 13:21:11 +0000 (13:21 +0000)]
ppp: enable IPv6CP by default
Signed-off-by: Steven Barth <steven@midlink.org>
Backport of r42158
SVN-Revision: 42355
John Crispin [Mon, 1 Sep 2014 13:21:09 +0000 (13:21 +0000)]
firewall: fix potential NULL pointer access
Properly skip struct ifaddr entries with NULL ifa_addr, thanks Kostas Papadopoulos for reporting.
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Backport of r42138
SVN-Revision: 42354
John Crispin [Mon, 1 Sep 2014 13:21:07 +0000 (13:21 +0000)]
firewall: implement selective conntrack flushing (#10225)
Utilize the new selective conntrack flushing facility to clear
out active conntrack entries referring to old IP addresses after
a firewall reload.
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Backport of r42114
SVN-Revision: 42353
John Crispin [Mon, 1 Sep 2014 13:21:05 +0000 (13:21 +0000)]
build: enable PER_FEED_REPO by default
Enable the separate feed repositories by default to align custom builds with the repositories
available on the snapshot download server.
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Backport of r42112
SVN-Revision: 42352
John Crispin [Mon, 1 Sep 2014 13:21:04 +0000 (13:21 +0000)]
hostapd: fix some whitespaces
Signed-off-by: Luka Perkov <luka@openwrt.org>
Backport of r42111
SVN-Revision: 42351
John Crispin [Mon, 1 Sep 2014 13:21:02 +0000 (13:21 +0000)]
broadcom-wl: add default tx and rx antennas
This adds default tx and rx antennas, so that LuCI shows automatic as the default radio option for the antennas.
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
Backport of r42085
SVN-Revision: 42350
Felix Fietkau [Sun, 31 Aug 2014 13:09:15 +0000 (13:09 +0000)]
netifd: update to the latest version, adds interface cleanup fixes
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r42345
SVN-Revision: 42347
Felix Fietkau [Sun, 31 Aug 2014 13:09:14 +0000 (13:09 +0000)]
netifd: update to the latest version, fixes proto-shell teardown after renew
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r42344
SVN-Revision: 42346
Felix Fietkau [Sun, 31 Aug 2014 12:04:22 +0000 (12:04 +0000)]
nettle: disable mips16 support to improve performance
Signed-off-by: Nikos Mavrogiannopoulos <nmav@gnutls.org>
Backport of r42341
SVN-Revision: 42342
Felix Fietkau [Sun, 31 Aug 2014 11:06:48 +0000 (11:06 +0000)]
imagebuilder: remove postinst files before generating image
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r42339
SVN-Revision: 42340
Jonas Gorski [Sat, 30 Aug 2014 11:04:19 +0000 (11:04 +0000)]
BB: binutils: link libbfd and libopcodes dynamically again
Backport of r42337.
Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 42338
Felix Fietkau [Fri, 29 Aug 2014 20:41:20 +0000 (20:41 +0000)]
mac80211: fix crash when unloading drivers (#17706)
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r42329
SVN-Revision: 42330
Felix Fietkau [Fri, 29 Aug 2014 11:37:50 +0000 (11:37 +0000)]
tools/libtool: add missing dependency on m4
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r42322
SVN-Revision: 42323
Felix Fietkau [Wed, 27 Aug 2014 09:52:44 +0000 (09:52 +0000)]
tools/bc: add missing dependency on bison
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r42309
SVN-Revision: 42311
Felix Fietkau [Wed, 27 Aug 2014 09:52:43 +0000 (09:52 +0000)]
tools/flex: disable building shared libfl.so
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r42308
SVN-Revision: 42310
Felix Fietkau [Tue, 26 Aug 2014 12:33:15 +0000 (12:33 +0000)]
target/toolchain: strip files, leave out the initial gcc
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r42302
SVN-Revision: 42303
Felix Fietkau [Tue, 26 Aug 2014 12:16:18 +0000 (12:16 +0000)]
build: override hardcoded paths to bison and m4 to make the SDK more relocatable
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r42300
SVN-Revision: 42301
Jo-Philipp Wich [Mon, 25 Aug 2014 12:46:35 +0000 (12:46 +0000)]
BB: uhttpd: do not configure TLS parameters if libustream-ssl is not present
A quite frequent problem after sysupgrading from an older, SSL enabled build
is that ustream-ssl is not installed so uhttpd fails to come up again due to
https listening directives in the preserved configuration.
Skip key/cert and ssl listen options when libustream-ssl.so is not present.
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Backport of r42284
SVN-Revision: 42286
Jo-Philipp Wich [Mon, 25 Aug 2014 12:46:27 +0000 (12:46 +0000)]
BB: iwinfo: properly decode SSIDs when scanning through wpa_supplicant
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Backport of r42273
SVN-Revision: 42285
Steven Barth [Mon, 25 Aug 2014 06:12:19 +0000 (06:12 +0000)]
odhcp6c: enable softwires support by default
SVN-Revision: 42275
Steven Barth [Sun, 24 Aug 2014 08:13:42 +0000 (08:13 +0000)]
odhcpd: fix static lease behavior with dhcpv4
SVN-Revision: 42271
Felix Fietkau [Thu, 21 Aug 2014 19:13:10 +0000 (19:13 +0000)]
uclient: update to latest version, fixes HTTP keepalive issues
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r42257
SVN-Revision: 42258
Jonas Gorski [Thu, 21 Aug 2014 18:28:09 +0000 (18:28 +0000)]
BB: dropbear: restore performance by disabling mips16
Disable MIPS16 to prevent it negatively affecting performance.
Observed was a increase of connection delay from ~6 to ~11 seconds
and a reduction of scp speed from 1.1MB/s to 710kB/s on brcm63xx.
Fixes #15209.
Backport of r42250.
Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 42255
Steven Barth [Wed, 20 Aug 2014 10:19:03 +0000 (10:19 +0000)]
base-files: increase igmp_max_memberships to improve multicast-proxy handling
SVN-Revision: 42228
Steven Barth [Wed, 20 Aug 2014 09:53:29 +0000 (09:53 +0000)]
dnsmasq: fix a race condition possibly leading to lockup
SVN-Revision: 42226
Steven Barth [Tue, 19 Aug 2014 06:00:09 +0000 (06:00 +0000)]
odhcpd: bump for bugfixes
SVN-Revision: 42219
Steven Barth [Tue, 19 Aug 2014 06:00:07 +0000 (06:00 +0000)]
dnsmasq: respect option dhcpv4 disabled in dhcp-config
SVN-Revision: 42218
Steven Barth [Tue, 19 Aug 2014 05:39:28 +0000 (05:39 +0000)]
bb: bump netifd
SVN-Revision: 42213
Florian Fainelli [Sat, 16 Aug 2014 17:35:13 +0000 (17:35 +0000)]
ar7: backport r42168
Signed-off-by: Florian Fainelli <florian@openwrt.org>
SVN-Revision: 42169
Hauke Mehrtens [Thu, 14 Aug 2014 21:24:14 +0000 (21:24 +0000)]
BB: brcm47xx: image: update list of Netgear WNR* firmwares
backport of r42165
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 42167
Hauke Mehrtens [Thu, 14 Aug 2014 21:23:36 +0000 (21:23 +0000)]
BB: brcm47xx: fix brcm47xx.legacy build
we should check for ssb here.
backport of r42164.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 42166
Steven Barth [Wed, 13 Aug 2014 14:04:01 +0000 (14:04 +0000)]
BB: feeds: switch telephony feed to for-14.07 branch
Signed-off-by: Jiri Slachta <slachta@cesnet.cz>
SVN-Revision: 42159
Steven Barth [Tue, 12 Aug 2014 13:30:38 +0000 (13:30 +0000)]
odhcpd: multiple DHCPv4 improvements (thx Christian Mehlis)
Signed-off-by: Steven Barth <steven@midlink.org>
SVN-Revision: 42154
Jo-Philipp Wich [Tue, 12 Aug 2014 11:15:38 +0000 (11:15 +0000)]
BB: package: fix segfault of iwinfo.scanlist("radio0").
This is a bug revealed in r41830.
First, the static variable `char nif[IFNAMSIZ]` of nl80211_phy2ifname()
would be zeroed out if the argument is "wlan0" or the like. This will
happen in the following call stack.
nl80211_get_scanlist("radio0", buf, len);
nl80211_phy2ifname("radio0") // return static var nif with content "wlan0"
nl80211_get_scanlist(nif, buf, len); // tail call
nl80211_get_mode(nif);
nl80211_phy2ifname(nif); // zero out nif
Later we try nl80211_ifadd("") which was supposed to create interface
"tmp.", but that won't happen because nl80211_msg() will put an invalid
ifidx 0 to the nlmsg.
Then iwinfo_ifup() and iwinfo_ifdown() would fail and happily
nl80211_get_scanlist() returned 0 and left *len undefined.
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
Backport of r42151
SVN-Revision: 42152
Jo-Philipp Wich [Mon, 11 Aug 2014 18:17:09 +0000 (18:17 +0000)]
BB: base-files: /lib/functions/network.sh: fix network_get_ipaddr6
When looking for the first ipaddr also consider the current prefix just
like network_get_ipaddrs6 does. If ipv6-address was empty the function
did not return the first ipaddr even if the list was non-empty.
fixes commit
83e9122f88a002871d5cdf421cf6aa6052b7e006
Signed-off-by: Henning Schild <henning@hennsch.de>
Backport of r42139
SVN-Revision: 42140
Jo-Philipp Wich [Sun, 10 Aug 2014 10:17:03 +0000 (10:17 +0000)]
BB: netfilter: consider reply direction on selective flushing
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Backport of r42095
SVN-Revision: 42096