openwrt/staging/robimarko.git
7 years agoar71xx: add support for Nokia WI2A-AC200i
Felix Fietkau [Tue, 27 Jun 2017 21:26:03 +0000 (23:26 +0200)]
ar71xx: add support for Nokia WI2A-AC200i

Specifications:
 - SoC: Qualcomm QCA9558 (720 MHz)
 - RAM: 256MB
 - Storage: 1MB NOR, 128 MB NAND flash
 - Ethernet: 1x1000M

Installation:
 1. Connect to serial console on the board
 2. Boot initramfs image over u-boot
 3. Copy image to the device and run sysupgrade

Installation without serial console is not supported at this time

Signed-off-by: Felix Fietkau <nbd@nbd.name>
7 years agobrcm47xx: use kernel 4.9 by default
Hauke Mehrtens [Wed, 18 Oct 2017 19:53:31 +0000 (21:53 +0200)]
brcm47xx: use kernel 4.9 by default

Kernel 4.9 is now working on the brcm47xx boards, we just recently fixed
the problem that some boards did not boot at all, by changing the memory
regions used to relocate the kernel to in the loader.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
7 years agobroadcom-wl: fix compile with kernel 4.9
Hauke Mehrtens [Wed, 18 Oct 2017 20:55:59 +0000 (22:55 +0200)]
broadcom-wl: fix compile with kernel 4.9

ENOENT could not be found by the compiler when compiling again kernel
4.9.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
7 years agokernel: fix ftrace support on 4.9
Felix Fietkau [Wed, 18 Oct 2017 20:46:29 +0000 (22:46 +0200)]
kernel: fix ftrace support on 4.9

When porting the kernel patches from 4.4 to 4.9, they were missing a
small chunk that ensures that ftrace sections are kept in the vmlinux
image, even when linked with --gc-sections

Signed-off-by: Felix Fietkau <nbd@nbd.name>
7 years agolantiq: ARV752DPW22: fix wireless mac address
Mathias Kresin [Wed, 18 Oct 2017 04:59:38 +0000 (06:59 +0200)]
lantiq: ARV752DPW22: fix wireless mac address

The ARV752DPW22 has the same generic mac address in the EEPROM as it
was already noticed for other lantiq boards using a ralink wireless.

Use the base mac address from the boardconfig partition as it is done
by the stock firmware.

Signed-off-by: Mathias Kresin <dev@kresin.me>
7 years agolantiq: ARV752DPW22: set correct wireless led trigger
Mathias Kresin [Mon, 16 Oct 2017 20:36:35 +0000 (22:36 +0200)]
lantiq: ARV752DPW22: set correct wireless led trigger

The ARV752DPW22 has a ralink based wireless and can not use the ath9k
only phy0tpt trigger.

Signed-off-by: Mathias Kresin <dev@kresin.me>
7 years agoomap: clean up configuration
Hauke Mehrtens [Sun, 15 Oct 2017 19:58:22 +0000 (21:58 +0200)]
omap: clean up configuration

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Acked-by: Alexander Couzens <lynxis@fe80.eu>
7 years agoomap: Add support for kernel 4.9
Hauke Mehrtens [Sun, 15 Oct 2017 19:43:59 +0000 (21:43 +0200)]
omap: Add support for kernel 4.9

This adds support for kernel 4.9 and replaces the kernel 4.4 support.

These are lynxis test results:
panda-board a3 - works, but no network, but master/4.4 doesn't have network either.
panda-board-a4 - u-boot SPL refuse to boot.
beaglebone-black - works
beagle-board - usb attached network doesn't come up and I doesn't have a serial around.
beagle-board-xm - ToDo: image code is missing.

Kernel 4.4 does not look better, so we merge this anyway.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Acked-by: Alexander Couzens <lynxis@fe80.eu>
7 years agokernel: bump 4.4 to 4.4.93
Kevin Darbyshire-Bryant [Wed, 18 Oct 2017 12:27:48 +0000 (13:27 +0100)]
kernel: bump 4.4 to 4.4.93

No patch refresh required.

Compile-tested for ar71xx - Archer C7 v2
Runtime-tested on  ar71xx - Archer C7 v2

Fixes the following CVEs:

- CVE-2017-15265
- CVE-2017-0786

Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
7 years agokernel: bump 4.9 to 4.9.57
Kevin Darbyshire-Bryant [Wed, 18 Oct 2017 12:19:10 +0000 (13:19 +0100)]
kernel: bump 4.9 to 4.9.57

Refresh patches.
Compile-tested for ar71xx - Archer C7 v2
Runtime-tested on  ar71xx - Archer C7 v2

Fixes the following CVEs:

- CVE-2017-7518
- CVE-2017-0786
- CVE-2017-1000255
- CVE-2017-12188
- CVE-2017-15265

Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
7 years agohostapd: bump PKG_RELEASE
Stijn Tintel [Wed, 18 Oct 2017 10:01:23 +0000 (13:01 +0300)]
hostapd: bump PKG_RELEASE

The previous commit did not adjust PKG_RELEASE, therefore the
hostapd/wpad/wpa_supplicant packages containing the AP-side workaround
for KRACK do not appear as opkg update.

Bump the PKG_RELEASE to signify upgrades to downstream users.

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
7 years agowireguard: version bump to 0.0.20171017
Jason A. Donenfeld [Tue, 17 Oct 2017 17:34:20 +0000 (19:34 +0200)]
wireguard: version bump to 0.0.20171017

This is a simple version bump. Changes:

  * noise: handshake constants can be read-only after init
  * noise: no need to take the RCU lock if we're not dereferencing
  * send: improve dead packet control flow
  * receive: improve control flow
  * socket: eliminate dead code
  * device: our use of queues means this check is worthless
  * device: no need to take lock for integer comparison
  * blake2s: modernize API and have faster _final
  * compat: support READ_ONCE
  * compat: just make ro_after_init read_mostly

  Assorted cleanups to the module, including nice things like marking our
  precomputations as const.

  * Makefile: even prettier output
  * Makefile: do not clean before cloc
  * selftest: better test index for rate limiter
  * netns: disable accept_dad for all interfaces

  Fixes in our testing and build infrastructure. Now works on the 4.14 rc
  series.

  * qemu: add build-only target
  * qemu: work on ubuntu toolchain
  * qemu: add more debugging options to main makefile
  * qemu: simplify shutdown
  * qemu: open /dev/console if we're started early
  * qemu: phase out bitbanging
  * qemu: always create directory before untarring
  * qemu: newer packages
  * qemu: put hvc directive into configuration

  This is the beginning of working out a cross building test suite, so we do
  several tricks to be less platform independent.

  * tools: encoding: be more paranoid
  * tools: retry resolution except when fatal
  * tools: don't insist on having a private key
  * tools: add pass example to wg-quick man page
  * tools: style
  * tools: newline after warning
  * tools: account for padding being in zero attribute

  Several important tools fixes, one of which suppresses a needless warning.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
7 years agohostapd: add wpa_disable_eapol_key_retries option
Stijn Tintel [Tue, 17 Oct 2017 13:35:03 +0000 (16:35 +0300)]
hostapd: add wpa_disable_eapol_key_retries option

Commit 2127425434046ae2b9f02fdbbdd37cac447af19c introduced an AP-side
workaround for key reinstallation attacks. This option can be used to
mitigate KRACK on the station side, in case those stations cannot be
updated. Since many devices are out there will not receive an update
anytime soon (if at all), it makes sense to include this workaround.

Unfortunately this can cause interoperability issues and reduced
robustness of key negotiation, so disable the workaround by default, and
add an option to allow the user to enable it if he deems necessary.

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
7 years agohostapd: backport extra changes related to KRACK
Stijn Tintel [Tue, 17 Oct 2017 13:24:14 +0000 (16:24 +0300)]
hostapd: backport extra changes related to KRACK

While these changes are not included in the advisory, upstream
encourages users to merge them.
See http://lists.infradead.org/pipermail/hostap/2017-October/037989.html

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
7 years agoRevert "ar71xx: Add GRO support to ag71xx"
Felix Fietkau [Tue, 17 Oct 2017 13:52:08 +0000 (15:52 +0200)]
Revert "ar71xx: Add GRO support to ag71xx"

This reverts commit 13e5e473699b92f171205e0f5c57c9ebe7922492.
This commit causes a severe regression in LAN->WAN routing performance
for several devices. This appears to be caused by the extra requirement
to validate the SKB checksum early in the rx path, which the ethernet
hardware does not do

Signed-off-by: Felix Fietkau <nbd@nbd.name>
7 years agohostapd: bump PKG_RELEASE
Stijn Tintel [Mon, 16 Oct 2017 23:13:34 +0000 (02:13 +0300)]
hostapd: bump PKG_RELEASE

The previous CVE bugfix commit did not adjust PKG_RELEASE, therefore the
fixed hostapd/wpad/wpa_supplicant packages do not appear as opkg update.

Bump the PKG_RELEASE to signify upgrades to downstream users.

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
7 years agomac80211: backport kernel fix for CVE-2017-13080
Stijn Tintel [Mon, 16 Oct 2017 22:49:58 +0000 (01:49 +0300)]
mac80211: backport kernel fix for CVE-2017-13080

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
7 years agoat91: fix legacy build
Hauke Mehrtens [Mon, 16 Oct 2017 19:36:17 +0000 (21:36 +0200)]
at91: fix legacy build

The build system took the DTB_SIZE definition from Default and not from
production-dtb under some conditions. Move the size definitions to
Default now as it is only used in production-dtb anyway.

Thanks Mathias Kresin for helping me with this.

Fixes: c2f052acaeb ("at91: convert boards to generic build target")
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
7 years agoppp: make the patches apply correctly again
Hauke Mehrtens [Mon, 16 Oct 2017 18:08:56 +0000 (20:08 +0200)]
ppp: make the patches apply correctly again

This fixes a compile problem recently introduced by me.

Fixes: f40fd43ab2f ("ppp: fix compile warning")
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
7 years agowireguard: add wireguard to base packages
Jason A. Donenfeld [Fri, 13 Oct 2017 15:05:18 +0000 (17:05 +0200)]
wireguard: add wireguard to base packages

Move wireguard from openwrt/packages to base a package.

This follows the pattern of kmod-cake and openvpn. Cake is a fast-moving
experimental kernel module that many find essential and useful. The
other is a VPN client. Both are inside of core. When you combine the two
characteristics, you get WireGuard. Generally speaking, because of the
extremely lightweight nature and "stateless" configuration of WireGuard,
many view it as a core and essential utility, initiated at boot time
and immediately configured by netifd, much like the use of things like
GRE tunnels.

WireGuard has a backwards and forwards compatible Netlink API, which
means the userspace tools should work with both newer and older kernels
as things change. There should be no versioning requirements, therefore,
between kernel bumps and userspace package bumps.

Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Acked-by: Jo-Philipp Wich <jo@mein.io>
Acked-by: Felix Fietkau <nbd@nbd.name>
7 years agoramips: fix typo in MT7621 NAND driver
Felix Fietkau [Mon, 16 Oct 2017 10:15:08 +0000 (12:15 +0200)]
ramips: fix typo in MT7621 NAND driver

Signed-off-by: Felix Fietkau <nbd@nbd.name>
7 years agohostapd: merge fixes for WPA packet number reuse with replayed messages and key reins...
Felix Fietkau [Mon, 16 Oct 2017 09:39:23 +0000 (11:39 +0200)]
hostapd: merge fixes for WPA packet number reuse with replayed messages and key reinstallation

Fixes:
- CERT case ID: VU#228519
- CVE-2017-13077
- CVE-2017-13078
- CVE-2017-13079
- CVE-2017-13080
- CVE-2017-13081
- CVE-2017-13082
- CVE-2017-13086
- CVE-2017-13087
- CVE-2017-13088

For more information see:
https://w1.fi/security/2017-1/wpa-packet-number-reuse-with-replayed-messages.txt

Signed-off-by: Felix Fietkau <nbd@nbd.name>
7 years agomalta: activate some more standard kernel features
Hauke Mehrtens [Sun, 15 Oct 2017 12:33:56 +0000 (14:33 +0200)]
malta: activate some more standard kernel features

These options where deactivated in the malta kernel, take the default
options form the generic kernel configuration now to better match the
other targets.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
7 years agomalta: upgrade to kernel 4.9
Hauke Mehrtens [Sun, 15 Oct 2017 10:42:33 +0000 (12:42 +0200)]
malta: upgrade to kernel 4.9

This brings the MIPS malta target to kernel 4.9.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
7 years agomalta: add 64 bit qemu commands to README
Hauke Mehrtens [Sun, 15 Oct 2017 12:18:40 +0000 (14:18 +0200)]
malta: add 64 bit qemu commands to README

This shows how to boot up the 64 bit images.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
7 years agoppp: fix compile warning
Hauke Mehrtens [Sun, 15 Oct 2017 12:19:20 +0000 (14:19 +0200)]
ppp: fix compile warning

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
7 years agolantiq: xrx200: rename nas0/ptm0 to dsl0
Martin Schiller [Mon, 9 Oct 2017 06:26:01 +0000 (08:26 +0200)]
lantiq: xrx200: rename nas0/ptm0 to dsl0

This change makes it possible to configure the wan/dsl ppp interface
settings independantly from the used TC-Layer (ATM/PTM).

Now you can move a device from an ADSL/ATM port to an VDSL/PTM port
without any configuration changes for example.

Signed-off-by: Martin Schiller <ms@dev.tdt.de>
[use the dsl0 interface name for the default netdev trigger in 01_led,
add ip dependency]
Signed-off-by: Mathias Kresin <dev@kresin.me>
7 years agomac80211: ath6kl: add missing usb-core dependency to kmod-ath6kl-usb
Hauke Mehrtens [Sun, 15 Oct 2017 08:53:46 +0000 (10:53 +0200)]
mac80211: ath6kl: add missing usb-core dependency to kmod-ath6kl-usb

This fixes a build problem with many targets.

Fixes 618ed77a17422a ("mac80211: add ath6kl kernel modules")
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
7 years agobinutils: add version 2.29.1
Hauke Mehrtens [Sat, 14 Oct 2017 19:28:11 +0000 (21:28 +0200)]
binutils: add version 2.29.1

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
7 years agokernel: kmod-macsec module for 4.9
Christian Lamparter [Sat, 14 Oct 2017 15:57:19 +0000 (17:57 +0200)]
kernel: kmod-macsec module for 4.9

MACsec/IEEE 802.1AE is useful to secure communication to and
from endpoints at Layer 2.

Starting with 4.6, the linux kernel provides a universal
macsec driver for authentication and encryption of traffic
in a LAN, typically with GCM-AES-128, and optional replay
protection.

http://standards.ieee.org/getieee802/download/802.1AE-2006.pdf

Note:
LEDE can utilize MACsec with a static connectivity association
key (static PSK) with the ip-full package installed.
<http://man7.org/linux/man-pages/man8/ip-macsec.8.html>

Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
7 years agolibs/libnl: Update to 3.3.0
Daniel Engberg [Sat, 13 May 2017 14:38:56 +0000 (16:38 +0200)]
libs/libnl: Update to 3.3.0

Update libnl to 3.3.0
Import patches to fix compilation
Source: https://git.busybox.net/buildroot/tree/package/libnl
Source: https://gitweb.gentoo.org/proj/musl.git/diff/dev-libs/libnl/files/libnl-3.3.0_rc1-musl.patch?id=48d2a287
Use more automatic toolchain logic

Signed-off-by: Daniel Engberg <daniel.engberg.lists@pyret.net>
7 years agopackage/kernel/leds-apu2: add apu3 board detection
Florian Eckert [Thu, 6 Jul 2017 09:20:20 +0000 (11:20 +0200)]
package/kernel/leds-apu2: add apu3 board detection

Signed-off-by: Florian Eckert <fe@dev.tdt.de>
7 years agopackage/kernel/leds-apu2: fix whitespaces
Florian Eckert [Thu, 6 Jul 2017 09:32:44 +0000 (11:32 +0200)]
package/kernel/leds-apu2: fix whitespaces

Signed-off-by: Florian Eckert <fe@dev.tdt.de>
7 years agotoolchain/glibc: Update to 2.26
Daniel Engberg [Sat, 12 Aug 2017 18:02:20 +0000 (20:02 +0200)]
toolchain/glibc: Update to 2.26

Update glibc to 2.26

Signed-off-by: Daniel Engberg <daniel.engberg.lists@pyret.net>
7 years agomvebu: clean up ClearFog Base package selection
Ryan Mounce [Sun, 1 Oct 2017 09:44:02 +0000 (20:14 +1030)]
mvebu: clean up ClearFog Base package selection

It is unclear why so many packages are selected for ClearFog Base compared
to its big brother, and there is no reason to not append metadata for Base.

Tidy this up as the only hardware difference between Base/Pro is the
presence of a switch and a different board name / device tree.

Signed-off-by: Ryan Mounce <ryan@mounce.com.au>
7 years agomvebu: Fix ClearFog sysupgrade board definitions
Ryan Mounce [Thu, 13 Jul 2017 09:11:59 +0000 (18:41 +0930)]
mvebu: Fix ClearFog sysupgrade board definitions

Remove redundancy for platform_do_upgrade_clearfog
Fix platform_copy_config_clearfog to reflect -base/-pro split

Signed-off-by: Ryan Mounce <ryan@mounce.com.au>
7 years agomvebu: Sort 02_network alphabetically
Ryan Mounce [Wed, 12 Jul 2017 02:21:09 +0000 (11:51 +0930)]
mvebu: Sort 02_network alphabetically

Signed-off-by: Ryan Mounce <ryan@mounce.com.au>
7 years agowpan-tools: add the wpan-ping to test the 6LoWPAN network
Yunhui Fu [Fri, 13 Oct 2017 18:32:40 +0000 (14:32 -0400)]
wpan-tools: add the wpan-ping to test the 6LoWPAN network

This patch adds the help tool wpan-ping to test the 6LoWPAN
network to help the user debug network problem.

Signed-off-by: Yunhui Fu <yhfudev@gmail.com>
7 years agotoolchain/gcc: update 5.x to 5.5.0
Ryan Mounce [Thu, 12 Oct 2017 01:21:36 +0000 (11:51 +1030)]
toolchain/gcc: update 5.x to 5.5.0

This is the final bugfix release in the gcc-5 series.

Compile and run tested on macOS 10.13 (Xcode 9), mvebu/ar71xx.

Removed redundant patch for macOS (backported upstream by yours truly)

Signed-off-by: Ryan Mounce <ryan@mounce.com.au>
7 years agoat91: add support for the WB50N module from Laird
Ben Whitten [Wed, 28 Jun 2017 10:28:12 +0000 (11:28 +0100)]
at91: add support for the WB50N module from Laird

This module from Laird includes the following:
  - CPU Atmel SoC SAMA5D31
  - Wifi QCA6004
  - Bluetooth CSR8811
  - RAM 64MB LPDDR
  - FLASH 128MB

The flash is a dual image layout, kernel a/b, rootfs a/b, and a user
partition.

Signed-off-by: Ben Whitten <ben.whitten@gmail.com>
7 years agoat91: add support for the WB45N module from Laird
Ben Whitten [Fri, 10 Jun 2016 19:07:32 +0000 (20:07 +0100)]
at91: add support for the WB45N module from Laird

This module from Laird includes the following:
  - CPU Atmel SoC ARM926EJS
  - Wifi AR6003
  - Bluetooth CSR8510
  - RAM 64MB LPDDR
  - FLASH 128MB

The flash is a dual image layout, kernel a/b, rootfs a/b, and a user
partition.

Signed-off-by: Ben Whitten <ben.whitten@gmail.com>
7 years agomac80211: add ath6kl kernel modules
Ben Whitten [Tue, 1 Dec 2015 17:42:39 +0000 (17:42 +0000)]
mac80211: add ath6kl kernel modules

Allow board to include the ath6kl kernel modules.

Signed-off-by: Ben Whitten <ben.whitten@gmail.com>
7 years agolinux-firmware: add ath6k firmware to package
Ben Whitten [Sun, 19 Jun 2016 22:49:16 +0000 (23:49 +0100)]
linux-firmware: add ath6k firmware to package

Systems which include the ath6k chipset need to have the firmware included
in the image.

Signed-off-by: Ben Whitten <ben.whitten@gmail.com>
7 years agoat91: separate MKUBIFS opts to defaults in the sub target
Ben Whitten [Thu, 1 Jun 2017 22:04:52 +0000 (23:04 +0100)]
at91: separate MKUBIFS opts to defaults in the sub target

Instead of applying global defaults based on selected board, transition
to using a per board setting for UBIFS and UBINIZE.

Signed-off-by: Ben Whitten <ben.whitten@gmail.com>
7 years agoat91: refresh kernel config, enable UBI block and DMA
Ben Whitten [Mon, 18 Sep 2017 12:09:21 +0000 (13:09 +0100)]
at91: refresh kernel config, enable UBI block and DMA

The platform generates squashfs images in a UBI block but misses the
kernel module to be able to mount the block.
DMA is also enabled to allow systems which include them in the DTS to
use it.

Signed-off-by: Ben Whitten <ben.whitten@gmail.com>
7 years agosunxi: backport support for Allwinner Security System PRNG
Lucian Cristian [Sun, 1 Oct 2017 15:29:22 +0000 (18:29 +0300)]
sunxi: backport support for Allwinner Security System PRNG

Signed-off-by: Lucian Cristian <lucian.cristian@gmail.com>
7 years agobusybox: provide "ip"
Hans Dedecker [Sat, 14 Oct 2017 18:13:29 +0000 (20:13 +0200)]
busybox: provide "ip"

Let busybox provide "ip" as it supports the ip applets link, address,
route, rule and neighbor

Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
7 years agoar71xx: add support for Mikrotik RB750P-PBr2
Robert Marko [Thu, 5 Oct 2017 12:26:23 +0000 (14:26 +0200)]
ar71xx: add support for Mikrotik RB750P-PBr2

Specifications:
- SoC: Qualcomm QCA9531 (650MHz)
- RAM: 64MB
- Storage: 16MB NOR SPI flash
- Ethernet: 5x100M (1 PoE in, 4 PoE out)
- Outdoor use ready

This ethernet router is based on the same platform as the hEX PoE lite.

Installation

1. login to the Mikrotik WebUI to backup your licence keys
2. setup a DHCP/BOOTP Server with:
     * DHCP-Option 66 (TFTP server name) pointing to a local TFTP
       Server within the same subnet of the DHCP range
     * DHCP-Option 67 (Bootfile-Name) matching the initramfs filename
       of the to be booted image
3. connect the port labled internet to your local network
4. keep the reset button pushed down and power on the board

The board should load and start the initramfs image from the TFTP
Server. Login as root/without password to the started LEDE via ssh
listing on IPv4 address 192.168.1.1. Use sysupgrade to install LEDE.

Revert to RouterOS

Use the "rbcfg" package on in LEDE:
  * rbcfg set boot_protocol bootp
  * rbcfg set boot_device ethnand
  * rbcfg apply

Open Netinstall and reboot routerboard. Now netinstall sees routerboard
and you can install RouterOS. If NetInstall gets stuck on Sending offer
just wait for it to timeout and then close and open Netinstall again.

Click on install again.

In order for RouterOS to function properly, you need to restore license
for the device. You can do that by including license in NetInstall

Signed-off-by: Robert Marko <robimarko@gmail.com>
7 years agoramips: add support for Kimax U25AWF-H1
Daniel Kucera [Sun, 19 Feb 2017 08:46:44 +0000 (09:46 +0100)]
ramips: add support for Kimax U25AWF-H1

Kimax U-25AWF-H1 is is a 2,5" HDD Enclosure with Wi-Fi/Eth conection
and battery, based on MediaTek MT7620A.

Patch rewritten from: https://forum.openwrt.org/viewtopic.php?pid=305643

Specification:

- MT7620A CPU
- 64 MB of RAM
- 16 MB of FLASH
- 802.11bgn WiFi
- 1x 10/100 Mbps Ethernet
- USB 2.0 Host
- UART for serial console

Flash instruction:
1. Download lede-ramips-mt7620-u25awf-h1-squashfs-sysupgrade.bin
2. Open webinterface a upgrade
3. After boot connect via ethernet to ip 192.168.1.1

Signed-off-by: Daniel Kucera <daniel.kucera@gmail.com>
[fix reset button gpio, don't add a lan/wan vlan config for single
port board, add -H1 suffix do make sure that this revision of the
board is supported/tested]
Signed-off-by: Mathias Kresin <dev@kresin.me>
7 years agompc85xx: Add Aerohive HiveAP-330 Access Point
Chris Blake [Sat, 30 Sep 2017 16:14:58 +0000 (11:14 -0500)]
mpc85xx: Add Aerohive HiveAP-330 Access Point

The following adds the Aerohive HiveAP-330 Access Point to LEDE under
the mpc85xx/p1020 subtarget.

Hardware:
- SoC: Freescale P1020NSE2DFB
- NAND: Intel JS28F512M29EWH 64MB
- Memory: 2x ProMOS V59C1G01168QBJ3 128MB (Total of 256MB)
- 2.4GHz WiFi: Atheros AR9390-AL1A
- 5.0GHz WiFi: Atheros AR9390-AL1A
- Eth1: Atheros AR8035-A PoE
- Eth2: Atheros AR8035-A
- TPM: Atmel AT97SC3204
- LED Driver: TI LP5521

Flashing:
1. Hook into UART (9600 baud) and enter U-Boot. You may need to enter a
password of administrator or AhNf?d@ta06 if prompted.
2. Once in U-Boot, tftp boot the initramfs image:
    dhcp;
    tftpboot 0x1000000 192.168.1.101:lede-
mpc85xx-p1020-hiveap-330-initramfs.zImage;
    tftpboot 0x6000000 192.168.1.101:lede-mpc85xx-p1020-hiveap-330.fdt;
    bootm 0x1000000 - 0x6000000;
3. Once booted, scp over the sysupgrade file and sysupgrade the device
to flash LEDE to the NAND.
    sysupgrade /tmp/lede-mpc85xx-p1020-hiveap-330-sysupgrade.img

Signed-off-by: Chris Blake <chrisrblake93@gmail.com>
7 years agompc85xx: Add cmdline override patch
Chris Blake [Sat, 30 Sep 2017 16:14:20 +0000 (11:14 -0500)]
mpc85xx: Add cmdline override patch

This patch adds a new kernel option called CONFIG_CMDLINE_OVERRIDE. This
setting is for devices with locked down u-boot environments, where users
are unable to change the default bootargs. When set, the fdt driver will
propagate the cmdline for the kernel from chosen/bootargs-override
instead of chosen/bootargs as long as it exists within the DTB.

Signed-off-by: Chris Blake <chrisrblake93@gmail.com>
7 years agompc85xx: Enable initramfs for p1020 subtarget
Chris Blake [Sat, 30 Sep 2017 16:13:35 +0000 (11:13 -0500)]
mpc85xx: Enable initramfs for p1020 subtarget

The following patch enables building of initramfs images by default for
the P1020 subtarget in mpc85xx.

Signed-off-by: Chris Blake <chrisrblake93@gmail.com>
7 years agoat91: Add ext4 filesystem
Sandeep Sheriker Mallikarjun [Fri, 13 Oct 2017 17:21:51 +0000 (10:21 -0700)]
at91: Add ext4 filesystem

Add ext4 filesystem for creating sdcard image with ext4 rootfs and
removing ext2 as it superset of ext4.

Signed-off-by: Sandeep Sheriker Mallikarjun <sandeepsheriker.mallikarjun@microchip.com>
7 years agoat91: Enable SDHCI for sama5 in default config
Sandeep Sheriker Mallikarjun [Fri, 13 Oct 2017 17:21:50 +0000 (10:21 -0700)]
at91: Enable SDHCI for sama5 in default config

Enabled SDHCI for sama5 in kernel default config and this is needed
to mount sdcard rootfs partition during boot.

Signed-off-by: Sandeep Sheriker Mallikarjun <sandeepsheriker.mallikarjun@microchip.com>
7 years agoat91bootstrap: remove manual copy of binaries to BIN_DIR
Sandeep Sheriker Mallikarjun [Wed, 4 Oct 2017 23:14:40 +0000 (16:14 -0700)]
at91bootstrap: remove manual copy of binaries to BIN_DIR

removed copying of binaries to BIN_DIR during install and using
default/install to install binaries to BIN_DIR folder.

Signed-off-by: Sandeep Sheriker Mallikarjun <sandeepsheriker.mallikarjun@microchip.com>
7 years agoat91bootstrap: Add BUILD_SUBTARGET variable
Sandeep Sheriker Mallikarjun [Wed, 4 Oct 2017 23:14:39 +0000 (16:14 -0700)]
at91bootstrap: Add BUILD_SUBTARGET variable

Added sama5 to BUILD_SUBTARGET variable.This will populate at91bootstrap
menu options in bootloader menu only when SAMA5 devices are selected as
SUBTARGET and to avoid showing up this menu when legacy device is
selected as SUBTARGET and fixed typo mistake: sama5d3 -> sama5d2.

Signed-off-by: Sandeep Sheriker Mallikarjun <sandeepsheriker.mallikarjun@microchip.com>
7 years agoat91: refresh kernel configuration
Hauke Mehrtens [Fri, 13 Oct 2017 22:00:43 +0000 (00:00 +0200)]
at91: refresh kernel configuration

This was done by running "make kernel_oldconfig" and
"make kernel_oldconfig CONFIG_TARGET=subtarget"

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
7 years agouboot-at91: multiple build fixes
Hauke Mehrtens [Fri, 13 Oct 2017 21:40:03 +0000 (23:40 +0200)]
uboot-at91: multiple build fixes

This fixes the following problems:
 * Add BUILD_DEVICES for legacy subtarget
 * Use features from u-boot.mk for sama5 subtarget This is mainly done
   by changing the prefix from uboot to U-Boot. This makes them depend
   on the sama5 subtarget and not selectable for the legacy subtarget
   any more

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
7 years agokernel: bump 4.4 to 4.4.92
Kevin Darbyshire-Bryant [Thu, 12 Oct 2017 19:06:34 +0000 (20:06 +0100)]
kernel: bump 4.4 to 4.4.92

No patch refresh changes required.

Compile tested for: ar71xx Archer C7 v2
Run tested: ar71xx Archer C7 v2

Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
7 years agoiproute2: bump to 4.13
Hans Dedecker [Sun, 1 Oct 2017 19:46:27 +0000 (21:46 +0200)]
iproute2: bump to 4.13

Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
7 years agomac80211: fix tx power regression
Felix Fietkau [Fri, 13 Oct 2017 18:27:33 +0000 (20:27 +0200)]
mac80211: fix tx power regression

Revert an accidental change that was introduced by having an old version
of the patch in my git tree, which was merged in 609208597b6d

Signed-off-by: Felix Fietkau <nbd@nbd.name>
7 years agodnsmasq: add listen_address parameter
Christian Lamparter [Thu, 12 Oct 2017 21:16:06 +0000 (23:16 +0200)]
dnsmasq: add listen_address parameter

This patch adds a parser for the uci representation of
dnsmasq's "-a | --listen-address" option.

In summary, this option forces dnsmasq to listen on the
given IP address(es). Both interface and listen-address
options may be given, in which case the set of both
interfaces and addresses is used.

Note that if no interface option is given, but listen_address is,
dnsmasq will not automatically listen on the loopback interface.
To achieve this, the loopback IP addresses, 127.0.0.1 and/or ::1
must be explicitly added.

This option is useful for ujailed dnsmasq instances, that would
otherwise fail to work properly, because listening to the
"This host on this network" address (aka 0.0.0.0 see rfc1700 page 4)
may not be allowed.

Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com> (PKG_RELEASE increase)
7 years agokernel: add fix for bgmac with B50212E B1 PHY
Rafał Miłecki [Fri, 13 Oct 2017 08:22:41 +0000 (10:22 +0200)]
kernel: add fix for bgmac with B50212E B1 PHY

This PHY requires some extra programming to work reliably with all
devices. Backport upstream fix for it.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
7 years agomt76: update to the latest version
Felix Fietkau [Thu, 12 Oct 2017 11:39:40 +0000 (13:39 +0200)]
mt76: update to the latest version

e781569 update to latest mac80211/cfg80211 API changes
37654d7 mt76x2: fix tx status ampdu length corner case

Signed-off-by: Felix Fietkau <nbd@nbd.name>
7 years agomac80211: update to wireless-testing 2017-10-06
Felix Fietkau [Sat, 7 Oct 2017 10:52:56 +0000 (12:52 +0200)]
mac80211: update to wireless-testing 2017-10-06

Rework the code to get rid of some extra kernel module dependencies
introduced in the last update.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
7 years agoath10k-ct: activate user space firmware loading again
Hauke Mehrtens [Thu, 12 Oct 2017 20:44:06 +0000 (22:44 +0200)]
ath10k-ct: activate user space firmware loading again

This backports a patch from kernel 4.14 to the ath10k-ct version based
on kernel 4.13.
Some devices are using a user space script to load the calibration data
from the flash and this was not trigged any more.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
7 years agoath10k-ct driver: use dma_alloc_coherent, 4.13 based driver
Ben Greear [Tue, 3 Oct 2017 22:45:37 +0000 (15:45 -0700)]
ath10k-ct driver: use dma_alloc_coherent, 4.13 based driver

This should help ath10k work on systems with little or no IOMMU
memory.  apu2 can boot two 9888 NICs now, for instance.  From
upstream patch by Adrian Chadd.

And, start building the 4.13 based CT ath10k driver.

Signed-off-by: Ben Greear <greearb@candelatech.com>
7 years agoath10k-ct firmware: Tx-hang and EAPOL handling fixes for wave-2 firmware.
Ben Greear [Mon, 2 Oct 2017 19:57:59 +0000 (12:57 -0700)]
ath10k-ct firmware: Tx-hang and EAPOL handling fixes for wave-2 firmware.

Changes since last LEDE release include:

  *  Fix key-setting bug that broke sending the EAPOL 2/4 in some cases.  This was a
     bug I introduced some time back while trying to fix .11r and simplify the key
     handling logic.  (Patch to wpa_supplicant fixed the race with sending the 4/4
     and setting the key...un-patched supplicant will still have this race and the 4-way
     auth will not work as reliably.)

  *  Increase amount of active-tids that can be scheduled.  This fixes a tx-stall
     seen with many station vdevs.

  *  Fix bug in upstream code that would cause the maximum peer to never be scheduled
     for tx.

Signed-off-by: Ben Greear <greearb@candelatech.com>
7 years agolayerscape: only support 64-bit for ls1088ardb/ls2088ardb
Yangbo Lu [Tue, 10 Oct 2017 11:17:37 +0000 (19:17 +0800)]
layerscape: only support 64-bit for ls1088ardb/ls2088ardb

This is no requirement and plan to support 32-bit for ls1088ardb
and ls2088ardb. Current 32-bit firmware for them couldn't work,
so only keep 64-bit support for these two boards in menuconfig.

Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
7 years agolayerscape: fix compile issue for usb ehci-fsl driver
Yangbo Lu [Tue, 10 Oct 2017 08:25:22 +0000 (16:25 +0800)]
layerscape: fix compile issue for usb ehci-fsl driver

When build firmware for layerscape target with CONFIG_ALL_NONSHARED=y,
there would be a compile issue of usb ehci-fsl driver. Actually this
driver was for PPC platforms initially and was not ready for non-PPC
now, but a kernel kconfig patch removed PPC dependency for it. So that
kernel patch should be reverted.

Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
7 years agoar71xx: fix section mismatches
Hauke Mehrtens [Sat, 7 Oct 2017 15:54:12 +0000 (17:54 +0200)]
ar71xx: fix section mismatches

Fix some section mismatches found with kernel 4.9.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
7 years agoar71xx: Add kernel 4.9 support
Hauke Mehrtens [Tue, 3 Oct 2017 16:02:59 +0000 (18:02 +0200)]
ar71xx: Add kernel 4.9 support

This add support for kernel 4.9 to the ar71xx target.
It was compile tested with the generic, NAND and mikrotik subtarget.
Multiple members of the community tested it on their boards and did not
report any major problem so far.

Especially the NAND part received some changes to adapt to the new
kernel APIs. The serial driver hack used for the Arduino Yun was not
ported because the kernel changed there a lot.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
7 years agobrcm47xx: relocate the stack in loader
Hauke Mehrtens [Sun, 8 Oct 2017 16:31:03 +0000 (18:31 +0200)]
brcm47xx: relocate the stack in loader

By default we are reusing the stack provided by CFE, like it is intended
by CFE. On my WRT54GS it is located at 0x8043BF30, so a big kernel image
could overwrite it. Relocate it to a different memory region which is
still under the 8MB RAM, but in the higher area. We only need this
memory region for the stack of the loader, Linux will set up this
for its own.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
7 years agobrcm47xx: relocate loader to higher address
Hauke Mehrtens [Sun, 8 Oct 2017 14:52:28 +0000 (16:52 +0200)]
brcm47xx: relocate loader to higher address

The boot process on a WRT54GL works the following way:
1. CFE gets loaded by the boot rom from flash
2. CFE loads the loader from the flash and gzip uncompresses it
3. CFE starts the loader
4. The loader stores the FW arguments and relocates itself to
   BZ_TEXT_START (now 0x80600000)
5. The loader reads the Linux image from flash
6. The loader lzma decompresses the Linux image to LOADADDR (0x80001000)
7. The loader executes the uncompress Linux image at LOADADDR

The BZ_TEXT_START was set to 0x80400000 before. When the kernel gets
uncompressed and is bigger than BZ_TEXT_START - LOADADDR it overwrote
the loader which was currently uncompressing it and made the board
crash. Increase the BZ_TEXT_START my 2 MB to have more space for the
kernel. Even on 16MB RAM devices the memory goes till 0x80FFFFFF so this
should not be a problem.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
7 years agobcm53xx: backport DTS changes up to the first 4.15 queued commits
Rafał Miłecki [Tue, 10 Oct 2017 07:19:10 +0000 (09:19 +0200)]
bcm53xx: backport DTS changes up to the first 4.15 queued commits

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
7 years agoramips/mt76x8: add initramfs support for tplink 841n v13
Alexander Couzens [Mon, 9 Oct 2017 11:51:21 +0000 (13:51 +0200)]
ramips/mt76x8: add initramfs support for tplink 841n v13

The tplink 841n v13 requires an tplink v2 image header in
front of an initramfs image.

To boot an initramfs image:
- break the uboot by holding the '4' key
- setup your tftp server 192.168.0.255
- tftp 81000000 lede-ramips-mt76x8-tl-wr841n-v13-initramfs-kernel.bin
- bootm

Signed-off-by: Alexander Couzens <lynxis@fe80.eu>
7 years agoimage-commands: tplink-v2-header: pass kernel loadaddr and entry
Alexander Couzens [Mon, 9 Oct 2017 11:50:03 +0000 (13:50 +0200)]
image-commands: tplink-v2-header: pass kernel loadaddr and entry

Initramfs images won't boot if the default loadaddr and entrypont is
different.

Signed-off-by: Alexander Couzens <lynxis@fe80.eu>
7 years agonet: uqmi: fix blocking in endless loops when unplugging device
Alexandru Ardelean [Fri, 29 Sep 2017 13:35:30 +0000 (16:35 +0300)]
net: uqmi: fix blocking in endless loops when unplugging device

If you unplug a QMI device, the /dev/cdc-wdmX device
disappears but uqmi will continue to poll it endlessly.

Then, when you plug it back, you have 2 uqmi processes,
and that's bad, because 2 processes talking QMI to the
same device [and the same time] doesn't seem to work well.

Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
7 years agoconntrack-tools: switch to git
Stijn Tintel [Wed, 20 Sep 2017 12:58:54 +0000 (15:58 +0300)]
conntrack-tools: switch to git

There have been a number of interesting fixes in conntrack-tools since
the current latest release. Most notable is that this fixes IPv6
conntrack table syncing when cross-compiling conntrack-tools.

7e7748d src/main: refresh help message
fe32043 conntrackd.8: refresh file
47a4dda conntrackd.8: add reference to systemd
0cfe7ff doc/manual: include some bits about init systems
74a418b conntrackd: cthelper: ftp: Set match offset/len for PORT mangling
d833bed conntrackd: cthelper: ftp: Fix debug print
dd4b5a1 conntrackd: cthelper: Add new mdns helper
498d698 Link nfct and helper modules with `-z lazy`
9e94e85 sync-mode: print errno message on failure
ab81c35 log: print messages to stdout/sderr if running in console mode
631d92b log: introduce a mechanism to know if log was initialized
ccb1c8b conntrackd: replace error reporting in the config parser with dlog()
bee121e conntrackd: replace fprintf calls with dlog()
5a51b04 conntrack-tools: update Arturo Borrero Gonzalez email address
abb9984 helper: remove copy and paste from uapi kernel header
a91a004 src: add log message when resync is requested by other node
c2d8be1 systemd: fix missing log.h include
f6ca216 config: drop old/obsolete/deprecated conntrackd.conf config options
8b83771 conntrack: send mark filter to kernel iff set
1ba5e76 conntrackd: cthelper: Don't leak nat_tuple
832166d conntrackd: cthelper: Free pktb after use
ff843bc conntrackd: config: Do not strdup() tokens
b61c454 conntrackd: cthelper: ssdp: Track UPnP eventing
8ea394e conntrackd: Remove obsolete rule to catch ambiguous Checksum option
39398cd conntrackd: CommitTimeout breaks DisableExternalCache set On
29b390a conntrack: Support IPv6 NAT
381827a conntrackd: factorice tx_queue functions
131df89 conntrackd: factorize resync operations
d31bacc conntrackd: consolidate more code to use resync_send()
3d98496 conntrackd: request resync at startup
ef410bf conntrackd: remove use of HAVE_INET_PTON_IPV6
9d38445 conntrackd: evaluate configuration earlier
6feded7 conntrackd: cleanup if failed forking
dbfdea7 conntrackd: deprecate unix backlog configuration
210f542 conntrackd: make the daemon run in RT mode by default
37cc7f0 conntrackd: remove warning for -S
d2849d1 conntrack: Show multiple CPUs stats from proc
bc0b49a conntrackd: cthelper: ssdp: fix build with musl
0c77a25 tests: don't fail on modprobe since the driver might be built-in
eefe649 conntrack.8: refresh manpage

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
7 years agolibnetfilter_conntrack: switch to git
Stijn Tintel [Wed, 20 Sep 2017 12:54:30 +0000 (15:54 +0300)]
libnetfilter_conntrack: switch to git

In order to build conntrack-tools from git, a newer version of
libnetfilter_conntrack is required. As 1.0.6 is currently the latest
release, switch to git.

b0a7cf7 include: expose a copy of nf_conntrack_common.h
f68f7b3 conntrack: fix missing break in setobjopt_undo_dnat()
79dac5a conntrack: revert getobjopt_is_nat() condition
b266523 libnetfilter_conntrack: bump version to 1.0.7
e870432 labels: don't crash on NULL labelmap

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
7 years agolibnetfilter_queue: bump to git HEAD
Stijn Tintel [Sun, 8 Oct 2017 18:46:44 +0000 (21:46 +0300)]
libnetfilter_queue: bump to git HEAD

b39cac7 src: Correct typo in the location of internal.h in #include
58cb066 src: Declare the define visibility attribute together
e84b559 Revert "src: Declare the define visibility attribute together"
003c2b1 examples: set dummy connmark value to show use of NFQA_CT nested attribute
63973da doc: extend the doxygen section about NFQA_CFG_F_GSO
d7f74c7 build: bump version to 1.0.3
3f9eb57 build: bump library release version too
601abd1 doc: Add information about retrieving UID/GID/SECCTX fields

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
7 years agomediatek: fix a sleep while atomic bug inside the ethernet driver
John Crispin [Mon, 9 Oct 2017 09:29:01 +0000 (11:29 +0200)]
mediatek: fix a sleep while atomic bug inside the ethernet driver

this sometimes happens when we have hw nat enabled.

Signed-off-by: John Crispin <john@phrozen.org>
7 years agotreewide: switch git.netfilter.org to HTTPS
Stijn Tintel [Sun, 8 Oct 2017 18:09:46 +0000 (21:09 +0300)]
treewide: switch git.netfilter.org to HTTPS

As git.netfilter.org seems to support HTTPS, use that instead of HTTP
which is insecure, or GIT which is blocked on many corporate networks.

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
7 years agoipset-dns: bump to git HEAD
Stijn Tintel [Sun, 8 Oct 2017 15:03:29 +0000 (18:03 +0300)]
ipset-dns: bump to git HEAD

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
7 years agokernel: bump 4.9 to 4.9.54
Stijn Tintel [Sun, 8 Oct 2017 14:56:59 +0000 (17:56 +0300)]
kernel: bump 4.9 to 4.9.54

Refresh patches.
Remove upstreamed patches:
- ramips/0067-enable-mt7621-xhci.patch
- ramips/0085-pinmux-util.patch
- ramips/301-fix-rt3883.patch

Compile-tested on brcm2708/bcm2708, octeon, ramips/mt7621, x86/64.
Runtime-tested on brcm2708/bcm2708, octeon, ramips/mt7621, x86/64.

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
7 years agokernel: bump 4.4 to 4.4.91
Kevin Darbyshire-Bryant [Sun, 8 Oct 2017 11:41:52 +0000 (12:41 +0100)]
kernel: bump 4.4 to 4.4.91

Refresh patches.

Compile-tested for: ar71xx Archer C7 v2
Run-tested on: ar71xx Archer C7 v2

Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
7 years agosunxi: build image for the NanoPi NEO
Daniel Golle [Wed, 20 Sep 2017 17:32:01 +0000 (19:32 +0200)]
sunxi: build image for the NanoPi NEO

The NanoPi NEO is a small Allwinner H3 based board available with
different DRAM configurations.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
7 years agosunxi: kernel: add patches for NanoPi NEO
Daniel Golle [Tue, 26 Sep 2017 18:05:31 +0000 (20:05 +0200)]
sunxi: kernel: add patches for NanoPi NEO

Enable dwmac-sun8i Ethernet core to use the built-in FastEthernet port.
Enable exposed UART, USB and I2C pins.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
7 years agouboot-sunxi: build for NanoPi NEO
Daniel Golle [Sat, 23 Sep 2017 00:47:19 +0000 (02:47 +0200)]
uboot-sunxi: build for NanoPi NEO

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
7 years agolayerscape: refresh patches
Stijn Tintel [Sun, 8 Oct 2017 13:34:37 +0000 (16:34 +0300)]
layerscape: refresh patches

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
7 years agokernel: update and refresh patches
Stijn Tintel [Thu, 28 Sep 2017 07:08:59 +0000 (10:08 +0300)]
kernel: update and refresh patches

The lantiq patch 0028-NET-lantiq-various-etop-fixes.patch and sunxi
patch 0051-stmmac-form-4-11.patch no longer applied after applying the
the "generalize napi_complete_done()" patch.
Update them so they apply, and refresh patches while at it.

Fixes: 9aeb7ce8dc5 ("generic: net: generalize napi_complete_done")
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
7 years agogeneric: net: generalize napi_complete_done
Pavel Kubelun [Thu, 7 Sep 2017 09:52:43 +0000 (05:52 -0400)]
generic: net: generalize napi_complete_done

Backport upstream commit that improves ethernet performance by a
small amount.

Compile and run tested on ipq8065.

Signed-off-by: Pavel Kubelun <be.dissent@gmail.com>
7 years agogeneric: net: tcp: backport tcp tx performance patches
Pavel Kubelun [Tue, 25 Jul 2017 07:57:31 +0000 (03:57 -0400)]
generic: net: tcp: backport tcp tx performance patches

An overall throughput gain of 22 % for heavy TCP use over a single TX queue.

Original patchset comment
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?h=v4.13&id=3f4888adae7c1619b990d98a9b967536f71822b8

Signed-off-by: Pavel Kubelun <be.dissent@gmail.com>
7 years agokernel: split 82574L patch into multiple files
Stijn Tintel [Sun, 8 Oct 2017 11:58:59 +0000 (14:58 +0300)]
kernel: split 82574L patch into multiple files

When refreshing patches that cointain multiple patches including headers
in a single file, quilt will remove the headers from all but the first
patch. This makes it difficult to review commits that refresh patches.
Next to that, if only a few of the patch series are accepted in -stable,
the patch needs to be manually modified. With each patch in a separate
file, it's just a matter of git rm.

Refresh patches while at it.

Patchwork links:
[1/5] https://patchwork.kernel.org/patch/9857487/
[2/5] https://patchwork.kernel.org/patch/9857489/
[3/5] https://patchwork.kernel.org/patch/9857495/
[4/5] https://patchwork.kernel.org/patch/9857491/
[5/5] https://patchwork.kernel.org/patch/9857493/

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
7 years agokernel: Add some more generic config options
Hauke Mehrtens [Sun, 8 Oct 2017 10:40:26 +0000 (12:40 +0200)]
kernel: Add some more generic config options

These options are deactivating some kernel modules for IP blocks not
uses on this SoC. I saw the same when working with the ARM64 Marvell
board so it is better to move them to generic.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
7 years agolayerscape: remove wifi config
Hauke Mehrtens [Sun, 8 Oct 2017 10:33:59 +0000 (12:33 +0200)]
layerscape: remove wifi config

This is provided by the mac80211 package

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
7 years agokernel: add missing config options for layerscape target
Hauke Mehrtens [Sun, 8 Oct 2017 10:27:23 +0000 (12:27 +0200)]
kernel: add missing config options for layerscape target

This fixes some problems found by build bot.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
7 years agoar71xx: Fix NULL pointer dereferece in at803x_link_change_notify()
Andrey Jr. Melnikov [Sat, 7 Oct 2017 21:00:28 +0000 (00:00 +0300)]
ar71xx: Fix NULL pointer dereferece in at803x_link_change_notify()

Check pdata pointer before use to avoid OOPS if dev_get_platdata() return
NULL.

Signed-off-by: Andrey Jr. Melnikov <temnota.am@gmail.com>
7 years agolayerscape: select sdk dts for ls1043ardb/ls1046ardb
Yangbo Lu [Mon, 4 Sep 2017 06:29:17 +0000 (14:29 +0800)]
layerscape: select sdk dts for ls1043ardb/ls1046ardb

ls1043ardb/ls1046ardb should use sdk dts. This patch is
to fix this.

Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
7 years agolayerscape: fix CPU_TYPE for armv8_32b targets
Yangbo Lu [Mon, 4 Sep 2017 05:53:38 +0000 (13:53 +0800)]
layerscape: fix CPU_TYPE for armv8_32b targets

Layerscape uses Cortex-A53 or Cortex-A72 CPU. It's proper
to set below CPU_TYPE for 32-bit mode.
CPU_TYPE:=cortex-a15
CPU_SUBTYPE:=neon-vfpv4

Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>