Before commit 60fb4c92 ("hostapd: add ubus reload") netifd was
tracking hostapd/wpa_supplicant and restarting wifi in case of a
process crash. Restore this behaviour by tracking the PIDs of
hostapd and wpa_supplicant.
Also make sure hostapd and/or wpa_supplicant have been started before
emmitting ubus calls to them using ubus wait_for.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>

In addition to wpa_supplicant and hostapd managed interfaces, also
track unmanaged interfaces. This is used to make sure that running
'wifi' always returns into a clean state regardless of what the user
did before.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>

Allows graceful restart of crashing hostapd/wpa_supplicant instances

Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
[daniel@makrotopia.org: attempt to launch only present services]

Fixes: 60fb4c92 ("hostapd: add ubus reload")
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
[daniel@makrotopia.org: polish commit message]

The "/dts-v1/;" identifier is supposed to be put once at the beginning
of a device tree file. Thus, it makes no sense to provide it a second
time in to-be-included DTSI files.

This removes the identifier from all DTSI files in /target/linux.

Most of the DTS files in OpenWrt do contain the "/dts-v1/;". It is
missing for most of the following targets, though:
mvebu, ipq806x, mpc85xx, ipq40xx

This does not touch ipq806x for now, as the bump to 4.19 is close.

Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>

This enables the CTRL_IFACE_MIB symbol for wpad-full and hostapd-full.
If it is not enabled, statistic outputs such as "hostapd_cli all_sta"
are empty.

Signed-off-by: David Bauer <mail@david-bauer.net>

Netgear WNR3500L is an already supported device, but out of the
box, the device has no switch configuration and there is no wan.
The correct configuration for this specific model is similar to
some other models. This simple commit adds the correct switch
and the out-of-the-box experience is improved.

Experimentally determined:

Port 0 => WAN
Port 1..4 => LAN
Port 5..7 => unused
Port 8 => CPU

Signed-off-by: Olli Asikainen <olli.asikainen@gmail.com>
Tested-by: Fabian Zaremba <fabian@youremail.eu>
[added port mapping to commit message]
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>

A small subset of devices uses decimal notation for mediatek,mtd-eeprom
in DTS files. Convert to hexadecimal notation to be consistent with
all the rest.

Also change "0" to "0x0" in the same files for consistency.

Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>

The Mikrotik RBM33G has only 2 LAN ports.

Signed-off-by: Martin Schiller <ms@dev.tdt.de>
[moved node in 02_network to maintain alphabetic sorting]
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>

Physical port order watched from the backside of the C20i
(from left to right) is: Internet / 1 / 2 / 3 / 4

Physical Port	Switch port
WAN             0
LAN 3           1
LAN 4           2
LAN 1           3
LAN 2           4
(not used)      5
CPU             6

Signed-off-by: Walter Sonius <walterav1984@gmail.com>
[commit message/title improvements]
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>

"[...] the size component shall be zero."
(See "PCI Bus Binding to: IEEE Std 1275-1994 Rev 2.1"
section "4.1.1 Open Firmware-defined Properties for Child Nodes")

Signed-off-by: Christian Lamparter <chunkeey@gmail.com>

Update iproute2 to latest stable version, see https://lwn.net/Articles/805654/


for the changes in 5.4.0

Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>

Signed-off-by: John Crispin <john@phrozen.org>

* wg-quick: linux: suppress error when finding unused table

This fixes a spurious warning messages seen with recent versions of iproute2
and kernels.

* wg-quick: linux: ensure postdown hooks execute
* wg-quick: linux: have remove_iptables return true
* wg-quick: linux: iptables-* -w is not widely supported

Adding in iptables had some hiccups. For the record, I'm very unhappy about
having to put any firewalling code into wg-quick(8). We'll of course need to
support nftables too at some point if this continues. I'm investigating with
upstream the possibility of adding a sysctl to patch the issue that iptables
is handling now, so hopefully at somepoint down the line we'll be able to shed
this dependency once again.

* send: use kfree_skb_list
* device: prepare skb_list_walk_safe for upstreaming
* send: avoid touching skb->{next,prev} directly

Suggestions from LKML.

* ipc: make sure userspace communication frees wgdevice

Free things properly on error paths.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>

Hardware:
* SoC: Atheros AR9342-BL1A
* RAM: 64MB DDR2 (Winbond W9751G6KB-25)
* Flash: 16MB SPI NOR (Macronix MX25L12835FZ2I-10G)
* Ethernet: 1x 10/100/1000 Mbps (Atheros AR8035-A) with 24V PoE support
* Wifi 2.4GHz: Atheros AR9340 v2
* WiFi 5GHz: Ubiquiti U-AME-G1-BR4A (rebranded QCA988X v2)
* LEDs: 1x Power, 1x Ethernet
* Buttons: 1x Reset
* UART: 1x TTL 115200n8, 3.3V RX TX GND, 3.3V pin closest to RJ45 port

The LEDs do not seem to be connected to any GPIO, so there is currently
no way to control them.

Installation via U-Boot, TFTP and serial console:
* Configure your TFTP server with IP 192.168.1.254
* Connect serial console and power up the device
* Hit any key to stop autoboot
* tftpboot 0x81000000 openwrt-ath79-generic-ubnt_litebeam-ac-gen2-initramfs-kernel.bin
* bootm 0x81000000
* copy openwrt-ath79-generic-ubnt_litebeam-ac-gen2-squashfs-sysupgrade.bin
  to /tmp
* sysupgrade /tmp/openwrt-ath79-generic-ubnt_litebeam-ac-gen2-squashfs-sysupgrade.bin

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
Acked-by: Petr Štetiar <ynezz@true.cz>
Acked-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>

The sysupgrade image contains OpenWrt specific metadata. Having this
metadata in the factory images makes no sense. Drop IMAGE/factory.bin
from Device/ubnt-wa and use the default from Device/ubnt instead.

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
Acked-by: Petr Štetiar <ynezz@true.cz>
Acked-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>

It turns out my 4.14 testing had a rather large flaw in it and the
'extack' mechanism isn't quite ready.  Remove the extack stuff from this
backport.

Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>

Since the original backports from kernel 5.3 a few things have been
tweaked by kernel bumps & other upstream changes.  Update the backport
to reflect upstream as closely as possible and remove the bitrot.

Functions remain the same, error reporting improved.

Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>

In d421a8b9 ("ath79: read label MAC address from flash instead
of using phy0/phy1") the source of the label MAC address was changed
for devices just reading it from phy0. To get rid of the dependency
from phy startup, addresses were read directly from the flash
locations that are used to initialize the phy MAC addresses.

Unfortunately, it turned out that Ubiquiti XM devices seem to have
different flash locations than expected, and also seem to have
specific locations for different devices (all in art/EEPROM):

0xe012 AR9280 Nanostation M2 - 0x120c
0xe035 AR9280 Nanostation M3 - 0x120c
0xe1b2 AR9280 Rocket M2 - 0x120c
0xe1c3 AR9280 Rocket M3 - 0x120c
0xe1b5 AR9280 Rocket M5 - 0x120c
0xe2d5 AR9280 Bullet M2 Titanium - 0x120c
0xe2b5 AR9280 Nanobridge M5 - 0x120c
0xe202 AR9280 Bullet M2 - 0x120c
0xe232 AR9287 Nanobridge M2 - 0x110c
0xe4a2 AR9285 AirRouter - 0xa0bf
Picostation M2 - 0x120c and 0xa0bf
Nanostation Loco M2 - not in 0x120c, other locations not checked

An additional problem of the Ubiquiti device support in OpenWrt is
that we provide images that match several subvariants of the devices,
which might have different MAC address locations.

Given that reading the address from phy0 in 02_network _is_ working
for the ath79 target in general, it does not seem reasonable to
rebuild a complex MAC address retrieval mechanism which is already
present in the ath9k driver.

So, this patch reverts the label MAC address source for Ubiquiti XM
devices (and the Unifi AP) to /sys/class/ieee80211/phy0/macaddress.

This doesn't affect XW and Unifi AC devices, where the label MAC
address source is defined via device tree.

For alfa-network,ap121f the location 0x1002 is kept, as this has
been verified during device support preparation in PR #2199.

Fixes: d421a8b9 ("ath79: read label MAC address from flash
instead of using phy0/phy1")

Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>

IPv6 protocol is enabled on all gre interfaces, but gre(v6)tap
interfaces are usually added to a bridge interface, in which case
IPv6 should be enabled only on the bridge interface.

Signed-off-by: Alin Nastac <alin.nastac@gmail.com>
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com> [PKG_RELEASE increase]

For changes in 7.67.0; see https://curl.haxx.se/changes.html#7_67_0



Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>

Sysupgrade was failing due to incorrect replacement of backticks:
/sbin/sysupgrade: /lib/upgrade/platform.sh: line 101: syntax error: missing '))'

Fixes: 0bbfc3df ("octeon: replace backticks by $(...)")
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>

The Atheros AR8327 and AR8337 have (according to their datasheet) a
VLAN table with a maximum of 4096 entries.

Currently, there's a hard limit of 128 VLANs, which is the same as
for most other Atheros switches. Increase this limit only for the
AR83x7 series and modify some shared functions to allow them to work
with a variable max VLAN count.

Signed-off-by: David Bauer <mail@david-bauer.net>

The Owl Loader (named after the codename that Atheros gave
these devices back in the day) has been accepted upstream.

This patch removes the "misc" driver OpenWrt had and adds
the remaining differences against the version that ships
with 5.4-rc1 into a separate "120-owl-loader-compat.patch"
file that can be cut down once AR71XX is being dealt with.

Note: I decided to keep the existing (kmod-)owl-loader
package name around for now. The kernel module file in
the kmod package will be called ath9k_pci_owl_loader.ko
though.

Acked-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>

Build with musl libc fails with BUSYBOX_DEFAULT_FEATURE_VI_REGEX_SEARCH
enabled. Enabling BusyBox's vi regex search option depends upon GNU
regex.  Musl libc does not support GNU regex[1].

So this patch adds explicit dependency on GNU libc and while at it
remove the FIXME comment.

1. https://wiki.musl-libc.org/functional-differences-from-glibc.html

Ref: https://dev.archive.openwrt.org/ticket/21741.html
Ref: https://forum.openwrt.org/t/busybox-not-compiling/
Ref: https://github.com/openwrt/packages/issues/4453


Signed-off-by: Satadru Pramanik <satadru@umich.edu>
[commit subject/description tweaks, From: fix, USE_GLIBC fix, removed comments]
Signed-off-by: Petr Štetiar <ynezz@true.cz>

Problem found by AddressSanitizer[1]:

 Latest `grep` (git commit 1019e6e) compiled with asan may cause a
 heap-buffer-overflow when `-i` is specified.

     ./grep -i '\(\(\)*.\)*\(\)\(\)\1' /bin/chvt

 =================================================================
 ==16206==ERROR: AddressSanitizer: heap-buffer-overflow on address

1. https://debbugs.gnu.org/34140

Ref: https://sourceware.org/bugzilla/show_bug.cgi?id=24114


Signed-off-by: Alin Nastac <alin.nastac@gmail.com>
[commit title and description facelift]
Signed-off-by: Petr Štetiar <ynezz@true.cz>

This reverts commit 81764319 ("mt76: probe load mt7615 driver
asynchronously").  After said commit, users report that MT7615 no longer
works on boot and they have to manually enable WiFi (via command "wifi") to
make it working again.

Fixes: FS#2546
Ref: https://forum.openwrt.org/t/xiaomi-r3p-no-wifi-on-boot/45509


Signed-off-by: DENG Qingfang <dengqf6@mail2.sysu.edu.cn>
[commit description facelift]
Signed-off-by: Petr Štetiar <ynezz@true.cz>

MeshPoint.One is Wi-Fi hotspot and smart IoT gateway (based upon
Jalapeno module from 8Devices).

MeshPoint.One (https://meshpointone.com

) is a unique Wi-Fi hotspot and
smart city gateway that can be installed and powered from street
lighting (even solar power in the future).  MeshPoint provides up to 27
hours of interrupted Wi-Fi and IoT services from internal battery even
when external power is not available.  MeshPoint.One can be used for
disaster relief efforts in order to provide instant Wi-Fi coverage that
can be easily expanded by just adding more devices that create wide area
mesh network.  MeshPoint.One devices have standard Luci UI for
management.

Features:
- 1x 1Gpbs WAN
- 1x 1Gbps LAN
- POE input (eth0)
- POE output (eth1)
- Sensor for temperature, humidity and pressure (Bosch BME280)
- current, voltage and power measurement via TI INA230
- Hardware real time clock
- optional power via Li-Ion battery
- micro USB port with USB to serial chip for easy OpenWrt terminal
  access
- I2C header for connecting additional sensors

Installation:
-------------
Simply flash the sysupgrade image from stock firmware.

Or use the built in Web recovery into bootloader:
Hold Reset button for 5 to 20 seconds or use UART and httpd command.
Web UI will appear on 192.168.2.100 by default.
For web recovery use the factory.ubi image.

Signed-off-by: Damir Samardzic <damir.samardzic@sartura.hr>
Signed-off-by: Damir Franusic <damir.franusic@sartura.hr>
Signed-off-by: Valent Turkovic <valent@meshpoint.me>
Signed-off-by: Robert Marko <robert@meshpoint.me>
[commit description long line wrap, usb->USB]
Signed-off-by: Petr Štetiar <ynezz@true.cz>

Lets move common code for Jalapeno into DTSI, this way Jalapeno based
boards don't introduce duplicate code.

While at it, lets also fix some style issues and update to current DTS
style.

Signed-off-by: Robert Marko <robert@meshpoint.me>
[commit description long line wrap]
Signed-off-by: Petr Štetiar <ynezz@true.cz>

e45b140 interface: warn if ip6hint is truncated

Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>

This also fixes mac80211_prepare_vif iw set channel in monitor or
mesh mode.

Signed-off-by: Santiago Piccinini <spiccinini@altermundi.net>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
[daniel@makrotopia.org: fixed commit message]

Refreshed all patches.

Fixes:
- CVE-2019-18660

Compile-tested on: none
Runtime-tested on: none

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>

Refreshed all patches.

Altered patches:
- 950-0064-mfd-Add-Raspberry-Pi-Sense-HAT-core-driver.patch
- 0005-mtd-physmap_of-Move-custom-initialization.patch

Remove upstreamed:
- 0001-pinctrl-gemini-Mask-and-set-properly.patch
- 0002-pinctrl-gemini-Fix-up-TVC-clock-group.patch

Compile-tested on: cns3xxx
Runtime-tested on: cns3xxx

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>

Refreshed all patches.

Compile-tested on: cns3xxx
Runtime-tested on: cns3xxx

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>

e4bd927 cast ucert_argv to proper type when passing to execv

Fixes warnings:

warning: passing argument 2 of 'execv' from incompatible pointer type
[-Wincompatible-pointer-types]
  254 |       execv(usign_argv[0], usign_argv)

Signed-off-by: Daniel Golle <daniel@makrotopia.org>

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>

By adding the vendor to the Makefile device definition node name,
one can derive the standard compatible used in SUPPORTED_DEVICES
instead of having to specify it manually.

Despite, this moves the naming scheme closer to what is used for
other targets (ath79, ramips).

Build-tested on all subtargets.
Run-tested on TP-Link TL-WDR4900 v1.

Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>

This replaces tabs by spaces when preceding an equal sign. This improves
consistency in the file and makes the indent look correct on all platforms.

While at it, also fix one case of inconsistent leading spaces.

Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>

This patch is based on #1689 and adds support for TP-Link Archer
C6 v2 (US) and A6 (US/TW).

The hardware is the same as EU and RU variant, except for GPIOs
(LEDS/Buttons), flash(chip/partitions) and UART being available
on the board.

- SOC: Qualcomm QCA9563 @ 775MHz
- Flash: GigaDevice GD25Q127CS1G (16MiB)
- RAM: Zentel A3R1GE40JBF (128 MiB DDR2)
- Ethernet: Qualcomm QCA8337N: 4x 1Gbps LAN + 1x 1Gbps WAN
- Wireless:
  - 2.4GHz (bgn) QCA9563 integrated (3x3)
  - 5GHz (ac) Qualcomm QCA9886 (2x2)
- Button: 1x power, 1x reset, 1x wps
- LED: 6x LEDs: power, wlan2g, wlan5g, lan, wan, wps
- UART: 115200, 8n1 (header available on board)

Known issues:
 - Wireless: 5GHz is known to have lower RSSI signal, it affects speed and range.

Flash instructions:

Upload openwrt-ath79-generic-tplink_archer-c6-v2-us-squashfs-factory.bin
via the router Web interface.

Flash instruction using tftp recovery:

1. Connect the computer to one of the LAN ports of the router
2. Set the computer IP to 192.168.0.66
3. Start a tftp server with the OpenWrt factory image in the
   tftp root directory renamed to ArcherA6v2_tp_recovery.bin.
4. Connect power cable to router, press and hold the
   reset button and turn the router on
5. Keep the reset button pressed until the WPS LED lights up
6. Wait ~150 seconds to complete flashing

Flash partitioning: I've followed #1689 for defining the partition layout
for this patch. The partition named as "tplink" @ 0xfd0000 is marked
as read only as it is where some config for stock firmware are stored.
On stock firmware those stock partitions starts at 0xfd9400 however
I had not been able to make it functional starting on the same address as
on stock fw, so it has been partitioned following #1689 and not the stock
partition layout for this specific partition. Due to that firmware/rootfs
partition lenght is 0xf80000 and not 0xf89400 as stock.

According to the GPL code, the EU/RU/JP variant does have different GPIO pins
assignment to LEDs and buttons, also the flash memory layout is different.

GPL Source Code: https://static.tp-link.com/resources/gpl/gpl-A6v2_us.tar.gz



Signed-off-by: Anderson Vulczak <andi@andi.com.br>
[wrap commit message, remove soft_ver change for C6 v2 EU, move LED aliases
to DTS files, remove dts-v1 in DTSI, node/property reorder in DTSI]
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>

This renames all remaining occurrences of "EEPROM" to "art" to
further harmonize the partition labelling in ath79. This will
help to reduce the amount of user-space code and might be
beneficial when code is copy/pasted in the future.

Affected are only devices from Ubiquiti, where the XM board is
already using "art" in ath79.

Acked-by: Piotr Dymacz <pepe2k@gmail.com>
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>