If an external module uses exported symbols from another external
module, Kbuild needs to have full knowledge of all symbols to
avoid spitting out warnings about undefined symbols.

Use PKG_EXTMOD_SUBDIRS to point to the build directory which contains
the Module.symvers.

Pass KERNEL_MAKE_FLAGS to the external module build, to inject
KBUILD_EXTRA_SYMBOLS. KBUILD_EXTRA_SYMBOLS holds a space separated list
of Module.symvers, which list all exported symbols.

Signed-off-by: Mathias Kresin <dev@kresin.me>

The USB ports if a FRIZZ!Box 7320 do not supply power to connected
devices.

Add the GPIOs enabling USB power as regulator, to enable USB power
supply as soon as the USB driver is loaded.

Fixes FS#3624

Signed-off-by: Mathias Kresin <dev@kresin.me>

When recursively deleting partitions, don't acquire the masters
partition lock twice. Otherwise the process endy up in a deadlocked
state.

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

These symbols were unset when configuring for ath79.

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

Hardware
--------

MediaTek MT7622
512MB DDR3 RAM
64M SPI-NOR Flash (Winbond W25Q512JV)
MediaTek MT7622 802.11bgn 4T4R WMAC
MediaTek MT7915 802.11ax 4T4R
Marvell AQR1112 100/1000/2500 NBase-T PHY
Holtek HT32F52241 LED controller
Reset Switch

UART
----

CPU UART0 at the pinout next to the Holtek MCU.

Pinout (first pin next to SoC / MCU)

0 3V3
1 RX
2 TX
3 GND

Settings are 115200 8N1.

Opening the case
----------------

Opening the case is not a nice task, as itis glued together. Insert a
flat knife between the front and back casing below the ethernet port.
Open up a gap this way and insert a flat scredriver, remove the knife.

Work your way around the casing by applying force to seperate the front
and back casing. This losens the glue and opens the plastic clips. Be
gentle, as these clips are very cheap and break quickly.

Installation
------------

1. Connect to the booted device at 192.168.1.20 using username/password
   "ubnt".

2. Transfer the OpenWrt sysupgrade image to the device using SCP.

3. Check the mtd partition number for bs / kernel0 / kernel1

   $ cat /proc/mtd

4. Set the bootselect flag to boot from kernel0

   $ dd if=/dev/zero bs=1 count=1 of=/dev/mtdblock6

5. Write the OpenWrt sysupgrade image to both kernel0 as well as kernel1

   $ dd if=openwrt.bin of=/dev/mtdblock8
   $ dd if=openwrt.bin of=/dev/mtdblock9

6. Reboot the device. It should boot into OpenWrt.

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

Add a driver for controlling the RGB LED via Ubiquitis own "LEDBAR" LED
controller based on the Holtek HT32F52241 MCU.

This driver is initially used by the Ubiquiti UniFi 6 LR, however
judging from FCC pictures the MCU is also found on the U6-Mesh as well
as the U6-Extender.

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

Runtime-tested on Comtrend AR-5387un.

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

Compile and runtime-tested on mediatek/mt7622

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

Only use symtab.h on the final kernel link

Signed-off-by: Felix Fietkau <nbd@nbd.name>

This fixes 4 security vulnerabilities/bugs:

- CVE-2021-2839 - SSLv2 vulnerability. Openssl 1.1.1 does not support
  SSLv2, but the affected functions still exist. Considered just a bug.

- CVE-2021-2840 - calls EVP_CipherUpdate, EVP_EncryptUpdate and
  EVP_DecryptUpdate may overflow the output length argument in some
  cases where the input length is close to the maximum permissable
  length for an integer on the platform. In such cases the return value
  from the function call will be 1 (indicating success), but the output
  length value will be negative.

- CVE-2021-2841 - The X509_issuer_and_serial_hash() function attempts to
  create a unique hash value based on the issuer and serial number data
  contained within an X509 certificate. However it was failing to
  correctly handle any errors that may occur while parsing the issuer
  field (which might occur if the issuer field is maliciously
  constructed). This may subsequently result in a NULL pointer deref and
  a crash leading to a potential denial of service attack.

- Fixed SRP_Calc_client_key so that it runs in constant time. This could
  be exploited in a side channel attack to recover the password.

The 3 CVEs above are currently awaiting analysis.

Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>

Add m4 patch to avoid conflict with tools/autoconf-archive.

Add build parallel as it seems to work now.

Remove a bunch of uClibc-ng hacks as it is not in the tree anymore.

Format security patch was fixed upstream.

Refreshed other patches.

Signed-off-by: Rosen Penev <rosenp@gmail.com>

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

This patch has been added to 5.4, but not been copied to 5.10:
7495acb5 ("kernel: backport mtd commit converting partitions doc syntax")

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

This applies another patch from 5.4 to 5.10 as well:
de09355f ("kernel/hack-5.4: make UDP tunneling user-selectable")

UDP tunneling support isn't user-selectable, but it's required by WireGuard
which is, for the time being, an out-of-tree module. We currently work around
this issue by selecting an unrelated module which depends on UDP tunnelling
(VXLAN). This is inconvenient, as it implies this unrelated module needs to be
built-in when doing a monolithic build.

Fix this inconvenience by making UDP tunneling user-selectable in the kernel
configuration.

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

File extension was truncated for
pending-5.4/770-11-net-ethernet-mtk_eth_soc-avoid-rearming-interrupt-if.pa

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

Reapply changes added to 5.4 but not copied to 5.10:
3da4acaa ("kernel: fix busy wait loop in mediatek PPE code")

The intention is for the loop to timeout if the body does not succeed.
The current logic calls time_is_before_jiffies(timeout) which is false
until after the timeout, so the loop body never executes.

time_is_after_jiffies(timeout) will return true until timeout is less
than jiffies, which is the intended behavior here.

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

Kernel has changed the of_get_phy_mode API in commit 0c65b2b90d13
("net: of_get_phy_mode: Change API to solve int/unit warnings").

This is already included in kernel 5.5, so fix the version switch
(though this will not actually matter for the versions we support).

Similar driver adjustments to account for the API change will
probably be necessary to various other local drivers.

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

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

Signed-off-by: Felix Fietkau <nbd@nbd.name>

fix typo in kernel initramfs zstd compression option

Signed-off-by: Felix Fietkau <nbd@nbd.name>

This is needed for linux 5.10, where modules.builtin is generated from
vmlinux.o

Signed-off-by: Felix Fietkau <nbd@nbd.name>

After the ABI version rework, packages need to be declared in the order of
their dependencies, so that dependent packages will use the right ABI version

Signed-off-by: Felix Fietkau <nbd@nbd.name>

Let's switch to 5.10 now that mac80211 has been updated.
Runtime-tested on ipq806x (Netgear R7800).

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

This allows using the last integrated PHY (and so e.g. WAN port on the
ASUS GT-AC5300).

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>

The sender domain has a DMARC Reject/Quarantine policy which disallows
sending mailing list messages using the original "From" header.

To mitigate this problem, the original message has been wrapped
automatically by the mailing list software.
Fixes: FS#2086
The logic for IS_TTY was broken, because it was testing stdin
instead of stdout.
MAKE_TERMOUT was introduced in GNU Make version 4.1 (05 Oct 2014),
so it should be available everywhere nowadays.

Signed-off-by: R. Diez <rdiezmail-openwrt@yahoo.com>

FS#2086 "IS_TTY in the makefile is broken" reports flawed detection of
stdout piping to a file. The issue describes how e.g. terminal color
codes and up in log files if running make like `make > log.txt`.

The proposed solution uses the make variable "MAKE_TERMOUT", which was
introduced in make 4.1. All major distributions seem to updated to 4.1
or later, so this ideally dosen't break anything.

Signed-off-by: Paul Spooren <mail@aparcar.org>

After the ABI version rework, packages need to be declared in the order of
their dependencies, so that dependent packages will use the right ABI version

Signed-off-by: Felix Fietkau <nbd@nbd.name>

We can't rely on the .version file being created yet, so use package variables
to get it

Signed-off-by: Felix Fietkau <nbd@nbd.name>

The removed patches were applied upstream.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>

A wrong quilt configuration was used last time.

Fixes: ed1e234d ("mac80211: refresh patches")
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>

There are 2 new patches:
1. Netgear R8000P switch ports
2. Netgear R8000P LEDs

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>

This was accidentally dropped in 27a4a71c
("metadata: handle ABI version rebuild tracking for transient dependencies")

Signed-off-by: Felix Fietkau <nbd@nbd.name>

Signed-off-by: Felix Fietkau <nbd@nbd.name>

2537be018587 cmake: add a possibility to set library version

Signed-off-by: Felix Fietkau <nbd@nbd.name>

Should avoid some spurious rebuilds

Signed-off-by: Felix Fietkau <nbd@nbd.name>

This reverts commit b12288fa.
The patchelf approach is too fragile, and the only users of this have been
converted to make patching unnecessary
Leave the abi_version_str variable in place in rules.mk

Signed-off-by: Felix Fietkau <nbd@nbd.name>

Makes it unnecessary to patch .so files after build

Signed-off-by: Felix Fietkau <nbd@nbd.name>

Use the version from PKG_ABI_VERSION

Signed-off-by: Felix Fietkau <nbd@nbd.name>

This removes the need to patch it afterwards

Signed-off-by: Felix Fietkau <nbd@nbd.name>