openwrt/openwrt.git
4 years agoomap24xx: Add missing kconfig options
Michael Büsch [Wed, 7 Oct 2015 09:40:42 +0000 (09:40 +0000)]
omap24xx: Add missing kconfig options

SVN-Revision: 47155

4 years agoomap24xx: KERNEL_PATCHVER:=4.1
Michael Büsch [Wed, 7 Oct 2015 08:30:58 +0000 (08:30 +0000)]
omap24xx: KERNEL_PATCHVER:=4.1

SVN-Revision: 47154

4 years agoomap24xx: Remove 4.0
Michael Büsch [Wed, 7 Oct 2015 08:29:06 +0000 (08:29 +0000)]
omap24xx: Remove 4.0

SVN-Revision: 47153

4 years agolantiq: bugfix luci lantiq_dsl status
John Crispin [Wed, 7 Oct 2015 08:00:54 +0000 (08:00 +0000)]
lantiq: bugfix luci lantiq_dsl status

Signed-off-by: Florian Eckert <Eckert.Florian@googlemail.com>
Signed-off-by: Helge Mader <hmader@tdt.de>
Reported-by: Jo-Philipp Wich <jow@openwrt.org>
SVN-Revision: 47152

4 years agoramips: fix /target/linux/ramips/dts/HC5XXX.dtsi
John Crispin [Wed, 7 Oct 2015 08:00:40 +0000 (08:00 +0000)]
ramips: fix /target/linux/ramips/dts/HC5XXX.dtsi

dtb complained about those lines

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 47151

4 years agohostapd: wait longer for inactive client probe (empty data frame)
Felix Fietkau [Tue, 6 Oct 2015 12:33:10 +0000 (12:33 +0000)]
hostapd: wait longer for inactive client probe (empty data frame)

One second is not enough for some devices to ackowledge null data frame
which is sent at the end of ap_max_inactivity interval. In particular,
this causes severe Wi-Fi instability with Apple iPhone which may take
up to 3 seconds to respond.

Signed-off-by: Dmitry Ivanov <dima@ubnt.com>
SVN-Revision: 47149

4 years agokernel: drop the now unused kmod-oprofile package
Felix Fietkau [Tue, 6 Oct 2015 12:33:04 +0000 (12:33 +0000)]
kernel: drop the now unused kmod-oprofile package

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47148

4 years agooprofile: drop kmod-oprofile dependency, it uses the perf API now
Felix Fietkau [Tue, 6 Oct 2015 12:32:59 +0000 (12:32 +0000)]
oprofile: drop kmod-oprofile dependency, it uses the perf API now

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47147

4 years agoramips: enable mt76 by default for the mt7620 subtarget (for devices with mt7620...
Felix Fietkau [Tue, 6 Oct 2015 12:32:31 +0000 (12:32 +0000)]
ramips: enable mt76 by default for the mt7620 subtarget (for devices with mt7620 + mt7612)

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47146

4 years agomt76: update to the latest version, adds a few minor fixes
Felix Fietkau [Mon, 5 Oct 2015 22:30:01 +0000 (22:30 +0000)]
mt76: update to the latest version, adds a few minor fixes

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47143

4 years agomac80211: add a fix to initialize the tid field in the tx queue struct
Felix Fietkau [Mon, 5 Oct 2015 15:43:14 +0000 (15:43 +0000)]
mac80211: add a fix to initialize the tid field in the tx queue struct

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47142

4 years agobuild: include homebrew include/library directory in cflags/ldflags on mac os x
Felix Fietkau [Mon, 5 Oct 2015 14:11:30 +0000 (14:11 +0000)]
build: include homebrew include/library directory in cflags/ldflags on mac os x

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47139

4 years agobuild: use host cflags/ldflags for openssl prereq check
Felix Fietkau [Mon, 5 Oct 2015 14:11:25 +0000 (14:11 +0000)]
build: use host cflags/ldflags for openssl prereq check

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47138

4 years agoramips: improve read performance of the m25p80 chunked io patch
Felix Fietkau [Mon, 5 Oct 2015 11:34:16 +0000 (11:34 +0000)]
ramips: improve read performance of the m25p80 chunked io patch

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47137

4 years agolldpd: wrap procd command args in separate quotes
John Crispin [Mon, 5 Oct 2015 10:29:01 +0000 (10:29 +0000)]
lldpd: wrap procd command args in separate quotes

Seems the match pattern was being adapted from 'eth0' to ' eth0'
because of the way I added the procd command args.

This did not seem to be a problem when there were multiple interfaces,
just on devices with single interfaces for lldpd to listen on.

Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
SVN-Revision: 47136

4 years agobase-files: add URL option for OEM manufacturer info
John Crispin [Mon, 5 Oct 2015 10:28:53 +0000 (10:28 +0000)]
base-files: add URL option for OEM manufacturer info

This URL can be embedded e.g. within UPnP announcements where a link
to the manufacturer's homepage is desired.

Signed-off-by: Michael Heimpold <mhei@heimpold.de>
SVN-Revision: 47135

4 years agoopenvpn: add handling for route-pre-down option
John Crispin [Mon, 5 Oct 2015 10:28:47 +0000 (10:28 +0000)]
openvpn: add handling for route-pre-down option

OpenVPN 2.3 added a route-pre-down option, to run a command before
routes are removed upon disconnection.

Signed-off-by: Jeffery To <jeffery.to@gmail.com>
SVN-Revision: 47134

4 years agouboot_envtools: add P2812HNUF1
John Crispin [Mon, 5 Oct 2015 10:28:39 +0000 (10:28 +0000)]
uboot_envtools: add P2812HNUF1

Add generation of /etc/fw_env.config for P2812HNUF1

Signed-off-by: Florian Eckert <Eckert.Florian@googlemail.com>
SVN-Revision: 47133

4 years agolantiq: several ARV7519RW22 fixes
John Crispin [Mon, 5 Oct 2015 10:28:34 +0000 (10:28 +0000)]
lantiq: several ARV7519RW22 fixes

Fixes ARV7519RW22 flash register (only 1 flash).
Power LED is green, not red.
Rename message LEDs to internet.
Add LEDs configuration (lan, wan, power).
Enable switch VLANs.
Fix secondary USB.

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
Signed-off-by: José Vázquez Fernández <ppvazquezfer@gmail.com>
SVN-Revision: 47132

4 years agolantiq: make v4.1 the default kernel
John Crispin [Mon, 5 Oct 2015 10:28:26 +0000 (10:28 +0000)]
lantiq: make v4.1 the default kernel

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 47131

4 years agolantiq: Add target TP-Link TD-W8980
John Crispin [Mon, 5 Oct 2015 10:28:19 +0000 (10:28 +0000)]
lantiq: Add target TP-Link TD-W8980

The device is similar to the TD-W8970, beside a different Atheros 2.4 GHz
wireless chip and the additional, PCI connected, WAVE300 5 GHz wireless.

Signed-off-by: Mathias Kresin <openwrt@kresin.me>
SVN-Revision: 47130

4 years agolantiq: fix ath_pci_fixup hang on bridged devices
John Crispin [Mon, 5 Oct 2015 10:28:12 +0000 (10:28 +0000)]
lantiq: fix ath_pci_fixup hang on bridged devices

Port of r41856.

In contrast to the brcm63xx target, it isn't sufficient to enable/disable
the bridge. The device needs to be enabled/disabled to fix the hang. The
bridge will be automatically enabled by the time the connected device is
enabled.

Fixes boot on TD-W8980.

Signed-off-by: Mathias Kresin <openwrt@kresin.me>
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
SVN-Revision: 47129

4 years agolantiq: Change the data-type of mac address in ETOP driver
John Crispin [Mon, 5 Oct 2015 10:28:07 +0000 (10:28 +0000)]
lantiq: Change the data-type of mac address in ETOP driver

This adds the changes from r46219 to the linux 4.1 patches as well.

Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
SVN-Revision: 47128

4 years agolantiq: bugfix uninitialized variables
John Crispin [Mon, 5 Oct 2015 10:27:55 +0000 (10:27 +0000)]
lantiq: bugfix uninitialized variables

Signed-off-by: Florian Eckert <Eckert.Florian@googlemail.com>
Reported-by: Sylwester Petela <sscapi@gmail.com>
SVN-Revision: 47127

4 years agobrcm2708: keep boot partition mounted to allow users to change config.txt
John Crispin [Mon, 5 Oct 2015 10:27:49 +0000 (10:27 +0000)]
brcm2708: keep boot partition mounted to allow users to change config.txt

This way it's easier to configure device tree overlays, customize other
parameters...

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
SVN-Revision: 47126

4 years agobrcm2708: add support for device tree overlays
John Crispin [Mon, 5 Oct 2015 10:27:41 +0000 (10:27 +0000)]
brcm2708: add support for device tree overlays

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
SVN-Revision: 47125

4 years agobrcm2708: switch device tree compilation to linux
John Crispin [Mon, 5 Oct 2015 10:27:34 +0000 (10:27 +0000)]
brcm2708: switch device tree compilation to linux

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
SVN-Revision: 47124

4 years agobrcm2708: fix kmod-sound-soc-bcm2708-i2s config
John Crispin [Mon, 5 Oct 2015 10:27:18 +0000 (10:27 +0000)]
brcm2708: fix kmod-sound-soc-bcm2708-i2s config

Right now, selecting kmod-sound-soc-bcm2708-i2s causes build errors due to
missing configs.
kmod-regmap enables I2C and SPI, causing build errors due to depending
variables not defined.

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
SVN-Revision: 47123

4 years agoramips: improve systick timer
John Crispin [Mon, 5 Oct 2015 10:27:09 +0000 (10:27 +0000)]
ramips: improve systick timer

when sleep mode is disable use MIPS as clocksource and clockevent instead of systick.
because MIPS timer has higher resolution 5ns less than systick 20us and
larger counter bits 32 > 16.
clean interrupt by write compare register at isr.
fix typo cause sleep mode not enable.

Signed-off-by: Michael Lee <igvtee@gmail.com>
SVN-Revision: 47122

4 years agoramips: fix use remove_irq to release irqaction resource
John Crispin [Mon, 5 Oct 2015 10:27:04 +0000 (10:27 +0000)]
ramips: fix use remove_irq to release irqaction resource

Signed-off-by: Michael Lee <igvtee@gmail.com>
SVN-Revision: 47121

4 years agoramips: add support for 16MB VoCore
John Crispin [Mon, 5 Oct 2015 10:27:00 +0000 (10:27 +0000)]
ramips: add support for 16MB VoCore

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
SVN-Revision: 47120

4 years agoramips: fix size-cells for spi nodes
John Crispin [Mon, 5 Oct 2015 10:26:54 +0000 (10:26 +0000)]
ramips: fix size-cells for spi nodes

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
SVN-Revision: 47119

4 years agoramips: Ralink RT3883 RGMII pinmux fix.
John Crispin [Mon, 5 Oct 2015 10:26:40 +0000 (10:26 +0000)]
ramips: Ralink RT3883 RGMII pinmux fix.

Due to datasheet of rt3883 SoC rgmii1 port handles pins 84-95 and rgmii2 port handles pins 72-83. When this function ports gets added to rt3883_pinmux_data there's wrong pinmux bits set (RT3883_GPIO_MODE_GE1 manages 84-95 pins and RT3883_GPIO_MODE_GE2 manages 72-83). So when enabling rgmii2 as GPIO driver confuses hardware and nothing work, neither rgmii nor gpio.
Also in '0030-pinctrl-ralink-add-pinctrl-driver.patch' typo in name of rgmii2 port.

Signed-off-by: Nick Leiten <nickleiten@gmail.com>
SVN-Revision: 47118

4 years agoramips: Enable rx of 1536 bytes ethernet frames on MT7621
John Crispin [Mon, 5 Oct 2015 10:26:34 +0000 (10:26 +0000)]
ramips: Enable rx of 1536 bytes ethernet frames on MT7621

The default switch frame size (with FCS + header) is 1536 bytes. But the
GMAC only accepted frames up to 1522 bytes. Setting it to 1536 allows to
receive ethernet frames using the full of MTU 1500 + an extra VLAN header +
VLAN header added by the switch.

Signed-off-by: Sven Eckelmann <sven@open-mesh.com>
SVN-Revision: 47117

4 years agoramips: Fix too small rx buffer
John Crispin [Mon, 5 Oct 2015 10:26:28 +0000 (10:26 +0000)]
ramips: Fix too small rx buffer

The driver assumes that the maximum received buffer for non-jumbo frames is
1536 bytes. But the allocation of the rx fragment doesn't reflect that. It
currently allocates fragments which will only be large enough to be used as
rx buffer with the size of 1534 bytes. This is problematic because the GMAC
will now try to write to 2 bytes which don't belong to its receive buffer
when a large enough ethernet frame is received.

This may already be a problem on existing chips but will at least become a
problem when the 1536 byte rx modus is enabled on MT7621a. It is required
on this SoC to receive ethernet frames which use their full 1500 bytes MTU
and a VLAN header next to the switch VLAN tag.

Signed-off-by: Sven Eckelmann <sven@open-mesh.com>
SVN-Revision: 47116

4 years agoramips: Fix setting of rx buffer length
John Crispin [Mon, 5 Oct 2015 10:26:21 +0000 (10:26 +0000)]
ramips: Fix setting of rx buffer length

The length of the DMA rx buffer was always set to 0 because the function
for extracting the length was used to calculate the value for setting it.
Instead the macro has to be split in a get and set function similar to the
TX_DMA_(GET_|)PLEN(0|1) macro.

No problem was noticed on MT7621a before this was changed and thus maybe it
was hidden by different problem which is not yet fixed.

Signed-off-by: Sven Eckelmann <sven@open-mesh.com>
SVN-Revision: 47115

4 years agoramips: Allow to receive vlan over untag ports on MT7530
John Crispin [Mon, 5 Oct 2015 10:26:09 +0000 (10:26 +0000)]
ramips: Allow to receive vlan over untag ports on MT7530

The MT7530 switch driver with enable_vlan set will automatically set all
ports to the user port mode. The hardware will remove the incoming vlan tag
on these ports and use it for its internal vlan. This is usually not wanted
and makes it impossible to communicate via vlan over the switch in both
directions.

It is possible to configure a switch port to "transparent mode" when this
port is only used as untag in the switch VLANs. This will disable the VLAN
untagging of packets when they were received on this port. The tagging on
"tag" ports based on the vlan id is still working.

The transparent port mode cannot be used when a port is both used in a VLAN
as "tag" and in another one as "untag" port.

Signed-off-by: Sven Eckelmann <sven@open-mesh.com>
SVN-Revision: 47114

4 years agoramips: Add build files for HiWiFi HC5x61 models
John Crispin [Mon, 5 Oct 2015 10:26:02 +0000 (10:26 +0000)]
ramips: Add build files for HiWiFi HC5x61 models

HiWiFi HC5661/5761/5861 models are manufactured by http://www.hiwifi.com <http://www.hiwifi.com/>. These models have similar hardware specs(MT7620A + 128M DDR2 + 16M flash). This patch adds support for them.

The original author is Justin Liu (rssnsj@gmail.com). I ported the patch to trunk and submitted it here with his approval.

v3 fix
    1: Spaces -> Tabs
    2: Removed some packages

Signed-off-by: Xiaoning Kang <kangxn@163.com>
SVN-Revision: 47113

4 years agoramips: Add DTS files for HiWiFi HC5x61 models
John Crispin [Mon, 5 Oct 2015 10:25:53 +0000 (10:25 +0000)]
ramips: Add DTS files for HiWiFi HC5x61 models

HiWiFi HC5661/5761/5861 models are manufactured by http://www.hiwifi.com. These models have similar hardware specs(MT7620A + 128M DDR2 + 16M flash). This patch adds support for them.

The original author is Justin Liu (rssnsj@gmail.com). I ported the patch to trunk and submitted it here with his approval.

v3 fix
  1: Merged most stuff into dtsi file
  2: Remove unnecessary empty lines.

Signed-off-by: Xiaoning Kang <kangxn@163.com>
SVN-Revision: 47112

4 years agoramips: Add base-files for HiWiFi HC5x61 models
John Crispin [Mon, 5 Oct 2015 10:25:47 +0000 (10:25 +0000)]
ramips: Add base-files for HiWiFi HC5x61 models

HiWiFi HC5661/5761/5861 models are manufactured by http://www.hiwifi.com. These models have similar hardware specs(MT7620A + 128M DDR2 + 16M flash). This patch adds support for them.

The original author is Justin Liu (rssnsj@gmail.com). I ported the patch to trunk and submitted it here with his approval.

v3 fix
   1: Fixed model order
   2: Remove manufacturer name from model name
   3: Use a hacky but prettier way to get mac address.

Signed-off-by: Xiaoning Kang <kangxn@163.com>
SVN-Revision: 47111

4 years agoar71xx: Correct WeIO board name
John Crispin [Mon, 5 Oct 2015 10:25:40 +0000 (10:25 +0000)]
ar71xx: Correct WeIO board name

Trivial parch to correct WeIO board name in the system.

Signed-off-by: Drasko DRASKOVIC <drasko.draskovic@gmail.com>
SVN-Revision: 47110

4 years agotarget/mpc85xx: check if 'generic' subtarget for image file cuImage.tl-wdr4900-v1
John Crispin [Mon, 5 Oct 2015 10:25:28 +0000 (10:25 +0000)]
target/mpc85xx: check if 'generic' subtarget for image file cuImage.tl-wdr4900-v1

Only add 'cuImage.tl-wdr4900-v1' to KERNEL_IMAGES if
the selected subtarget is 'generic'.

Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
SVN-Revision: 47109

4 years agoiwinfo: nl80211: add support for reading TX power from netlink
Jo-Philipp Wich [Mon, 5 Oct 2015 09:10:17 +0000 (09:10 +0000)]
iwinfo: nl80211: add support for reading TX power from netlink

Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
SVN-Revision: 47108

4 years agoar71xx: rename firmware for Chinese version of TL-WR941ND v6
Rafał Miłecki [Mon, 5 Oct 2015 08:54:26 +0000 (08:54 +0000)]
ar71xx: rename firmware for Chinese version of TL-WR941ND v6

It seems there are two *different* versions on TL-WR941ND v6: Chinese
and European one. Currently supported by OpenWrt is Chinese one. Rename
firmware to make it clear.
For details see:
http://wiki.openwrt.org/toh/tp-link/tl-wr941nd
https://forum.openwrt.org/viewtopic.php?id=59295

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 47107

4 years agolantiq: fix gcc 5 build
Matteo Croce [Sun, 4 Oct 2015 22:10:18 +0000 (22:10 +0000)]
lantiq: fix gcc 5 build

SVN-Revision: 47106

4 years agolantiq: fix binutils 2.15.1 build
Matteo Croce [Sun, 4 Oct 2015 22:03:04 +0000 (22:03 +0000)]
lantiq: fix binutils 2.15.1 build

SVN-Revision: 47105

4 years agokernel: update 4.1 to 4.1.10
Hauke Mehrtens [Sun, 4 Oct 2015 11:34:39 +0000 (11:34 +0000)]
kernel: update 4.1 to 4.1.10

Fixes a regression on spi flash devices (upstream commit 1583eaece6fa).

Changelogs:
* https://www.kernel.org/pub/linux/kernel/v4.x/ChangeLog-4.1.7
* https://www.kernel.org/pub/linux/kernel/v4.x/ChangeLog-4.1.8
* https://www.kernel.org/pub/linux/kernel/v4.x/ChangeLog-4.1.9
* https://www.kernel.org/pub/linux/kernel/v4.x/ChangeLog-4.1.10

096-ipv4-off-by-one-in-continuation-handling-in-proc-net.patch was applied
upstream and dropped.

Signed-off-by: Mathias Kresin <openwrt@kresin.me>
[update from 4.1.9 to 4.1.10]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 47104

4 years agotoplevel.mk: drop LD_LIBRARY_PATH overrides
Jo-Philipp Wich [Sat, 3 Oct 2015 09:49:24 +0000 (09:49 +0000)]
toplevel.mk: drop LD_LIBRARY_PATH overrides

Overriding LD_LIBRARY_PATH for host utilities frequently leads to problems
with host executables loading shared libraries from the staging_dir, leading
to crashes due to incompatible ABIs.

Since most host utilties either embed a proper rpath or are reworked to link
statically we do not need the workaround anymore.

Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
SVN-Revision: 47103

4 years agolibnl: Install include files into libnl3
Felix Fietkau [Fri, 2 Oct 2015 16:24:15 +0000 (16:24 +0000)]
libnl: Install include files into libnl3

Install header files into same location as pkgconfig/libnl-3.0.pc says:
  Cflags: -I${includedir}/libnl3

Signed-off-by: Bruno Randolf <br1@einfach.org>
SVN-Revision: 47102

4 years agoiproute2: adapt coexistence layer to new unified path
Steven Barth [Fri, 2 Oct 2015 08:24:35 +0000 (08:24 +0000)]
iproute2: adapt coexistence layer to new unified path

Signed-off-by: Steven Barth <steven@midlink.org>
SVN-Revision: 47081

4 years agobase-files: sanitize and unify $PATH
Steven Barth [Fri, 2 Oct 2015 08:24:32 +0000 (08:24 +0000)]
base-files: sanitize and unify $PATH

Previously init and hotplug paths were different from console.

Signed-off-by: Steven Barth <steven@midlink.org>
SVN-Revision: 47080

4 years agoprocd: make nand_do_platform_check check image only
Rafał Miłecki [Fri, 2 Oct 2015 06:03:13 +0000 (06:03 +0000)]
procd: make nand_do_platform_check check image only

Recently all targets have been updated to call nand_do_upgrade directly.
It's not needed anymore to leave a magic /tmp/sysupgrade-nand-path mark
during image check.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Acked-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 47079

4 years agomusl: re-enable des crypto support, fixes pppd MPPE issues
Felix Fietkau [Thu, 1 Oct 2015 19:50:06 +0000 (19:50 +0000)]
musl: re-enable des crypto support, fixes pppd MPPE issues

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47078

4 years agotoolchain/uClibc: add support of uClibc-ng
Felix Fietkau [Thu, 1 Oct 2015 19:50:01 +0000 (19:50 +0000)]
toolchain/uClibc: add support of uClibc-ng

uClibc-ng is a spin-off of original uClibc, see http://www.uclibc-ng.org/

We try to regularly add changes from uClibc to uClibc-ng.
We even sent patches and bug reports to the uClibc mailing list.
The config file is compatible between uClibc-ng 1.0 and uClibc git master.
This might change in the future.

Our main goal is to provide regularly a stable and tested release
to make embedded system developers happy.

The main advantage of uClibc-ng over olde good uClibc is regular releases
so there's no need to keep tons of patches on top of years old
0.9.33.2

Build-tested for
 [1] ARM: Sunxi generic
 [2] MIPS: Netgear WNDR3600/3700/3800

Run-tested for [yet out of the tree] Synopsys Designware ARC AXS101.

Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Cc: Waldemar Brodkorb <wbx@uclibc-ng.org>
Cc: Mathieu Olivari <mathieu@codeaurora.org>
Cc: Felix Fietkau <nbd@openwrt.org>
Cc: John Crispin <blogic@openwrt.org>
SVN-Revision: 47077

4 years agoralink: add irq to mt7628 gpio node
John Crispin [Thu, 1 Oct 2015 19:12:14 +0000 (19:12 +0000)]
ralink: add irq to mt7628 gpio node

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 47076

4 years agorpcd: update to latest git HEAD
John Crispin [Thu, 1 Oct 2015 18:44:26 +0000 (18:44 +0000)]
rpcd: update to latest git HEAD

Provides ubus calls for sysupgrade and password changing.

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 47075

4 years agoralink: the mmc driver can now handle CD lines that are active low
John Crispin [Thu, 1 Oct 2015 18:13:47 +0000 (18:13 +0000)]
ralink: the mmc driver can now handle CD lines that are active low

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 47074

4 years agoralink: speed selection was broken in spi-mt7621
John Crispin [Thu, 1 Oct 2015 15:35:44 +0000 (15:35 +0000)]
ralink: speed selection was broken in spi-mt7621

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 47073

4 years agoquilt: add further relocatability fixes
Jo-Philipp Wich [Tue, 29 Sep 2015 11:37:27 +0000 (11:37 +0000)]
quilt: add further relocatability fixes

Prevent quilt's build system from generating useless compat wrappers and
hardcoded absolute utility paths to binaries in staging_dir/host/.

Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
SVN-Revision: 47070

4 years agoprereq-build: test availability of Perl Thread::Queue module
Jo-Philipp Wich [Tue, 29 Sep 2015 11:37:10 +0000 (11:37 +0000)]
prereq-build: test availability of Perl Thread::Queue module

This is needed for building automake.

Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
SVN-Revision: 47069

4 years agouml: backport another upstream patch to fix compile errors
Felix Fietkau [Mon, 28 Sep 2015 12:22:35 +0000 (12:22 +0000)]
uml: backport another upstream patch to fix compile errors

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47067

4 years agobase-files: add a dependency on fstools
Felix Fietkau [Mon, 28 Sep 2015 09:40:39 +0000 (09:40 +0000)]
base-files: add a dependency on fstools

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47066

4 years agohostapd: check for banned client on association event
Rafał Miłecki [Mon, 28 Sep 2015 09:09:00 +0000 (09:09 +0000)]
hostapd: check for banned client on association event

When using FullMAC drivers (e.g. brcmfmac) we don't get mgmt frames so
check for banned client in probe request handler won't ever be used.
Since cfg80211 provides us info about STA associating let's put a check
there.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 47064

4 years agomt76: update to the latest version, fixes some aggregation issues
Felix Fietkau [Sun, 27 Sep 2015 22:39:53 +0000 (22:39 +0000)]
mt76: update to the latest version, fixes some aggregation issues

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47063

4 years agouml: fix build with gcc 5
Felix Fietkau [Sun, 27 Sep 2015 17:29:48 +0000 (17:29 +0000)]
uml: fix build with gcc 5

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47062

4 years agouml: update to linux 4.1
Felix Fietkau [Sun, 27 Sep 2015 17:29:38 +0000 (17:29 +0000)]
uml: update to linux 4.1

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47061

4 years agomac80211: backport patch setting TX power in interface info
Rafał Miłecki [Sun, 27 Sep 2015 10:25:27 +0000 (10:25 +0000)]
mac80211: backport patch setting TX power in interface info

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 47059

4 years agobrcm47xx: describe 300-fork_cacheflush.patch & 310-no_highpage.patch
Rafał Miłecki [Sun, 27 Sep 2015 06:53:12 +0000 (06:53 +0000)]
brcm47xx: describe 300-fork_cacheflush.patch & 310-no_highpage.patch

These were initially commited in r9285.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 47058

4 years agobrcm47xx: describe 160-kmap_coherent.patch
Rafał Miłecki [Sun, 27 Sep 2015 06:24:33 +0000 (06:24 +0000)]
brcm47xx: describe 160-kmap_coherent.patch

It was initially commited in r6642.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 47057

4 years agobrcm47xx: describe 610-pci_ide_fix.patch
Rafał Miłecki [Sun, 27 Sep 2015 06:05:53 +0000 (06:05 +0000)]
brcm47xx: describe 610-pci_ide_fix.patch

It was initially commited in r20924.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 47056

4 years agoigmpproxy: fix spurious restarts on interface events, pass used netdevs to procd...
Felix Fietkau [Sat, 26 Sep 2015 23:27:23 +0000 (23:27 +0000)]
igmpproxy: fix spurious restarts on interface events, pass used netdevs to procd instead

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47055

4 years agonetifd: update to the latest version, fixes spurious client isolation in unbridged...
Felix Fietkau [Sat, 26 Sep 2015 23:18:40 +0000 (23:18 +0000)]
netifd: update to the latest version, fixes spurious client isolation in unbridged AP configurations (#20574)

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47053

4 years agobcm53xx: add early support for Linksys EA6300 V1
Rafał Miłecki [Sat, 26 Sep 2015 22:57:33 +0000 (22:57 +0000)]
bcm53xx: add early support for Linksys EA6300 V1

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 47052

4 years agobison: replace the yacc script with one that does not hardcode the path (fixes #20605)
Felix Fietkau [Sat, 26 Sep 2015 21:10:48 +0000 (21:10 +0000)]
bison: replace the yacc script with one that does not hardcode the path (fixes #20605)

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47050

4 years agotools/patch: enable parallel build
Felix Fietkau [Sat, 26 Sep 2015 21:10:37 +0000 (21:10 +0000)]
tools/patch: enable parallel build

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47049

4 years agofirewall: allow DHCPv6 traffic to/from fc00::/6 instead of fe80::/10
Jo-Philipp Wich [Fri, 25 Sep 2015 08:41:12 +0000 (08:41 +0000)]
firewall: allow DHCPv6 traffic to/from fc00::/6 instead of fe80::/10

There is no RFC requirement that DHCPv6 servers must reply with a link local
address and some ISP servers in the wild appear to using addresses in the ULA
range to send DHCPv6 offers.

Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
SVN-Revision: 47048

4 years agokernel: crypto: fix ubsec_ssb.c compilation with 4.0+
Rafał Miłecki [Fri, 25 Sep 2015 04:22:13 +0000 (04:22 +0000)]
kernel: crypto: fix ubsec_ssb.c compilation with 4.0+

SSB_DEVTABLE_END was removed in 4.0 by upstream commit 673e2baaa6d9
("treewide: Remove unnecessary SSB_DEVTABLE_END macro").

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 47047

4 years agokernel: backport some SPI layer improvements from 4.1 to 3.18, significantly improves...
Felix Fietkau [Thu, 24 Sep 2015 16:07:10 +0000 (16:07 +0000)]
kernel: backport some SPI layer improvements from 4.1 to 3.18, significantly improves flash speed on ramips

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47045

4 years agomt76: update to the latest version, adds performance improvements
Felix Fietkau [Thu, 24 Sep 2015 15:39:45 +0000 (15:39 +0000)]
mt76: update to the latest version, adds performance improvements

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47043

4 years agoath9k: fix handling of tx headroom for padding (fixes #20556)
Felix Fietkau [Thu, 24 Sep 2015 15:02:15 +0000 (15:02 +0000)]
ath9k: fix handling of tx headroom for padding (fixes #20556)

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47042

4 years agomac80211: fix a few issues with fast-xmit support
Felix Fietkau [Thu, 24 Sep 2015 15:02:07 +0000 (15:02 +0000)]
mac80211: fix a few issues with fast-xmit support

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47041

4 years agokernel: remove old "rootfs" splitter
Rafał Miłecki [Thu, 24 Sep 2015 11:19:12 +0000 (11:19 +0000)]
kernel: remove old "rootfs" splitter

A year ago I made this old splitter obsolete (see r42828), it's time to
completely remove it now. For all this time we got these warnings:
"Dedicated partitioner didn't create "rootfs_data" partition, please fill a bug report!"
and
"Support for built-in "rootfs_data" splitter will be removed, please use CONFIG_MTD_SPLIT_SQUASHFS_ROOT"
but I'm not aware of any bug reports related to that.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 47040

4 years agooxnas: add platform_pre_upgrade for sysupgrade
Rafał Miłecki [Thu, 24 Sep 2015 09:18:59 +0000 (09:18 +0000)]
oxnas: add platform_pre_upgrade for sysupgrade

We're in process of cleaning nand_do_platform_check. Currently it leaves
a special mark in /tmp/sysupgrade-nand-path triggering some diffent code
path in nand_upgrade_stage1. This can be a bit confusing.

The plan is to have the check function only check the image and nothing
else. Then platform code (platform_pre_upgrade) should trigger NAND
specific upgrade path. This is what this patch implements.

This follows all other targets making it finally possible to change
nand_do_platform_check.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 47039

4 years agooxnas: drop some unused sysupgrade code
Rafał Miłecki [Thu, 24 Sep 2015 09:18:47 +0000 (09:18 +0000)]
oxnas: drop some unused sysupgrade code

First of all platform_do_upgrade_phase2 doesn't seem to be defined
anywhere. Other than that platform_do_upgrade is not likely to be ever
called. There is a call to nand_do_platform_check which triggers
nand_upgrade_stage1 call with "exit 0" in it.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Acked-by: Daniel Golle <daniel@makrotopia.org>
SVN-Revision: 47038

4 years agolibnl: split libnl into smaller libraries
Felix Fietkau [Thu, 24 Sep 2015 09:08:52 +0000 (09:08 +0000)]
libnl: split libnl into smaller libraries

Some modules may require only libnl-genl, some
libnl-route and fewer would require libnl-nf.

This patch splits the entire libnl package into smaller
more granular libs that can be installed individually as required.

Also added libnl*.so symlinks for convenience.

Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
SVN-Revision: 47037

4 years agoar71xx: Add WeIO Board Support
Felix Fietkau [Thu, 24 Sep 2015 09:08:47 +0000 (09:08 +0000)]
ar71xx: Add WeIO Board Support

Add support for WeIO board (http://we-io.net).
This board is based on Carambola2 board form 8Devices.

Signed-off-by: Drasko DRASKOVIC <drasko.draskovic@gmail.com>
Signed-off-by: Felix Fietkau <nbd@openwrt.org> [some cleanups]
SVN-Revision: 47036

4 years agokernel: add missing CAN related config symbol
Felix Fietkau [Thu, 24 Sep 2015 08:41:15 +0000 (08:41 +0000)]
kernel: add missing CAN related config symbol

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47035

4 years agoar71xx: fix bogus hardcoded kernel image size for Seama images (fixes #20585)
Felix Fietkau [Thu, 24 Sep 2015 08:37:50 +0000 (08:37 +0000)]
ar71xx: fix bogus hardcoded kernel image size for Seama images (fixes #20585)

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47034

4 years agodropbear: add respawn param in case dropbear crashes
Felix Fietkau [Thu, 24 Sep 2015 08:37:40 +0000 (08:37 +0000)]
dropbear: add respawn param in case dropbear crashes

Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
SVN-Revision: 47033

4 years agoimx6: disable ext2/ext3 filesystem support (ext4 is enough)
Felix Fietkau [Thu, 24 Sep 2015 08:37:35 +0000 (08:37 +0000)]
imx6: disable ext2/ext3 filesystem support (ext4 is enough)

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47032

4 years agouml: disable ext2/ext3 filesystem support (ext4 is enough)
Felix Fietkau [Thu, 24 Sep 2015 08:37:30 +0000 (08:37 +0000)]
uml: disable ext2/ext3 filesystem support (ext4 is enough)

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47031

4 years agoarm64: disable ext2/ext3 filesystem support (ext4 is enough)
Felix Fietkau [Thu, 24 Sep 2015 08:37:23 +0000 (08:37 +0000)]
arm64: disable ext2/ext3 filesystem support (ext4 is enough)

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47030

4 years agomxs: fixup permissions of image generation scripts
Jo-Philipp Wich [Wed, 23 Sep 2015 18:40:42 +0000 (18:40 +0000)]
mxs: fixup permissions of image generation scripts

Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
SVN-Revision: 47029

4 years agokernel: add a patch to fix crashes on pppoe disconnect/reconnect
Felix Fietkau [Tue, 22 Sep 2015 11:28:28 +0000 (11:28 +0000)]
kernel: add a patch to fix crashes on pppoe disconnect/reconnect

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47026

4 years agopx5g-standalone: use /dev/urandom instead of havege (fixes #20216)
Felix Fietkau [Tue, 22 Sep 2015 09:06:00 +0000 (09:06 +0000)]
px5g-standalone: use /dev/urandom instead of havege (fixes #20216)

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47025

4 years agogcc: fix accidentally dropped chunk of the musl gcc 5.2.0 powerpc patch (fixes #20501)
Felix Fietkau [Tue, 22 Sep 2015 08:30:15 +0000 (08:30 +0000)]
gcc: fix accidentally dropped chunk of the musl gcc 5.2.0 powerpc patch (fixes #20501)

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47024

4 years agogcc: use qstrip template to fix syntax highlighting
Felix Fietkau [Tue, 22 Sep 2015 08:30:05 +0000 (08:30 +0000)]
gcc: use qstrip template to fix syntax highlighting

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47023

4 years agopackage: Remove dependencies to kmod-ipv6
Steven Barth [Mon, 21 Sep 2015 21:15:41 +0000 (21:15 +0000)]
package: Remove dependencies to kmod-ipv6

Since r46834, IPv6 support is builtin if selected. Therefor, dependencies
on kmod-ipv6 can no longer be fulfilled, since it is not a module anymore.

Signed-off-by: Arjen de Korte <arjen+openwrt@de-korte.org>
SVN-Revision: 47022

4 years agomxs: re-work SD card image generation
Zoltan Herpai [Mon, 21 Sep 2015 21:10:43 +0000 (21:10 +0000)]
mxs: re-work SD card image generation

- Duckbill uses a different partitioning approach than standard
  FSL and Olimex
- use new sdimage to integrate U-Boot into the SD card images

Signed-off-by: Michael Heimpold <mhei@heimpold.de>
SVN-Revision: 47021

4 years agotools: add sdimage for mxs
Zoltan Herpai [Mon, 21 Sep 2015 21:08:08 +0000 (21:08 +0000)]
tools: add sdimage for mxs

This tool is used for SD card generation on Freescale i.MX23/i.MX28
platforms. These CPU's ROM need a tiny header of front of a boot stream.

Signed-off-by: Michael Heimpold <mhei@heimpold.de>
SVN-Revision: 47020