ar71xx: enable ethernet LED of Arduino Yun

Commit 05d73a2a7379 enabled GPIO on ethernet LED, but proper LED setup was
not added then. This commit fixes it by reverting the change on the LED.

Fixes: 05d73a2a7379 ("ar71xx: Arduino Yun board 'WLAN RST' button support")
Signed-off-by: Sungbo Eo <mans0n@gorani.run>
(cherry picked from commit a5e404d1923d135d335e4ece83f87e6e891396e2)

ar71xx: fix sysupgrade for Arduino Yun

Commit bb46b635df48 changed its partition scheme, but sysupgrade image
validation still uses the old format. This commit fixes it so that
force flag is not needed for sysupgrade.

Fixes: bb46b635df48 ("ar71xx: move Arduino Yun to generic building code")
Signed-off-by: Sungbo Eo <mans0n@gorani.run>
(cherry picked from commit 58dc1d0637425cfe023192466e6212009332b677)

ramips: add kmod-usb-dwc2 to ZyXEL Keenetic image

ZyXEL Keenetic has a USB port. Thus, DWC2 USB controller driver should
be in the default image for this device.

Fixes: a7cbf59e0e04 ("ramips: add new device ZyXEL Keenetic as kn")
Signed-off-by: Alexey Dobrovolsky <dobrovolskiy.alexey@gmail.com>
[fixed whitespace issue]
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(backported from commit 0a182fcba6d9cb2cf74cae9114ea4770ef928f75)
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>

ramips: remove patches for USB-dwc2

In FS#2738 we can see that patch first introduced in
e8ebcff ("ramips: add a explicit reset to dwc2")
breaks USB functionality since 18.06. Thus, this patch should be removed.

Removed:
- 0032-USB-dwc2-add-device_reset.patch

Fixes: FS#2738
Fixes: FS#2964
Signed-off-by: Alexey Dobrovolsky <dobrovolskiy.alexey@gmail.com>
(cherry picked from commit ab841b4393a4077a5819da1da040ab9a89e3b69d)

firewall: backport patch for mss clamping in both directions

Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>

ar71xx: Fix mikrotik NAND compile problem

This fixes the following compile error:
drivers/mtd/nand/rb91x_nand.c: In function 'rb91x_nand_remove':
drivers/mtd/nand/rb91x_nand.c:445:16: error: 'rbni' undeclared (first use in this function)
  nand_release(&rbni->chip);

Fixes: 0f07496f520c ("kernel: Update kernel 4.9 to version 4.9.229")
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit 66e04abbb6d0dec8642be5deb2fca4bba470f8ac)

ar71xx: Fix mikrotik NAND compile problem

There is one closing bracket too much.

Fixes: 0f07496f520c ("kernel: Update kernel 4.9 to version 4.9.229")
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit 014d3f593acea13ee6aa002d858f182806ed43f0)

wireguard: bump to 1.0.20200611

This bump fixes breakage introduced by kernel commit 8ab8786f78c3fc930f9abf6d6d85e95567de4e1f,
which is part of the 4.14.181 kernel bump, and backported ip6_dst_lookup_flow to 4.14.
This breaks the older WireGuard version currently in 19.07.

For reference, the compilation error is the one below:

build_dir/target-x86_64_musl/linux-x86_64/wireguard-linux-compat-1.0.20200506/src/compat/compat.h:104:42: error: 'const struct ipv6_stub' has no member named 'ipv6_dst_lookup'; did you mean 'ipv6_dst_lookup_flow'?
 #define ipv6_dst_lookup_flow(a, b, c, d) ipv6_dst_lookup(a, b, &dst, c) + (void *)0 ?: dst

Changelogs below taken from the official release announcements.

== Changes since v1.0.20200506 ==

  This release aligns with the changes I sent to DaveM for 5.7-rc7 and were
  pushed to net.git about 45 minutes ago.

  * qemu: use newer iproute2 for gcc-10
  * qemu: add -fcommon for compiling ping with gcc-10

  These enable the test suite to compile with gcc-10.

  * noise: read preshared key while taking lock

  Matt noticed a benign data race when porting the Linux code to OpenBSD.

  * queueing: preserve flow hash across packet scrubbing
  * noise: separate receive counter from send counter

  WireGuard now works with fq_codel, cake, and other qdiscs that make use of
  skb->hash. This should significantly improve latency spikes related to
  buffer bloat. Here's a before and after graph from some data Toke measured:
  https://data.zx2c4.com/removal-of-buffer-bloat-in-wireguard.png

  * compat: support RHEL 8 as 8.2, drop 8.1 support
  * compat: support CentOS 8 explicitly
  * compat: RHEL7 backported the skb hash renamings

  The usual RHEL churn.

  * compat: backport renamed/missing skb hash members

  The new support for fq_codel and friends meant more backporting work.

  * compat: ip6_dst_lookup_flow was backported to 4.14, 4.9, and 4.4

== Changes since v1.0.20200611 ==

  * qemu: always use cbuild gcc rather than system gcc
  * qemu: remove -Werror in order to build ancient kernels better
  * qemu: patch kernels that rely on ancient make
  * qemu: force 2MB pages for binutils 2.31
  * qemu: use cbuild gcc for avx512 exclusion
  * qemu: add extra fill in idt handler for newer binutils
  * qemu: support fetching kernels for arbitrary URLs
  * qemu: patch in UTS_UBUNTU_RELEASE_ABI for Ubuntu detection
  * qemu: work around broken centos8 kernel
  * qemu: mark per_cpu_load_addr as static for gcc-10

  Our qemu test suite can now handle more kernels and more compilers. Scroll
  down to the bottom of https://www.wireguard.com/build-status/ to see the
  expanded array of kernels we now test against, including some distro kernels.

  * compat: widen breadth of integer constants
  * compat: widen breadth of memzero_explicit backport
  * compat: backport skb_scrub_packet to 3.11
  * compat: widen breadth of prandom_u32_max backport
  * compat: narrow the breadth of iptunnel_xmit backport
  * compat: backport iptunnel_xmit to 3.11

  With the expanded qemu test suite, it was possible to expand our list of
  mainline kernels, so the backport compat layer is now more precise.

  * compat: ubuntu appears to have backported ipv6_dst_lookup_flow
  * compat: bionic-hwe-5.0/disco kernel backported skb_reset_redirect and ipv6 flow

  Ubuntu kernels changed recently, so this ensures we can compile with the
  latest Ubuntu releases.

  * compat: remove stale suse support

Signed-off-by: Stijn Segers <foss@volatilesystems.org>
(cherry picked from commit 1fd1f5e8cff18f97675ce303b05d411136b99fb0)

libubox: backport additional length-checking fixes

Fixes: FS#3177
Cc: Felix Fietkau <nbd@nbd.name>
Cc: Rafał Miłecki <rafal@milecki.pl>
Signed-off-by: Baptiste Jonglez <git@bitsofnetworks.org>

kernel: Update kernel 4.9 to version 4.9.229

Fixes:
- CVE-2020-10757

Run tested: ar71xx
Build tested: ar71xx

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

kernel: Update kernel 4.14 to version 4.14.187

Fixes:
- CVE-2020-10757

Run tested: x86
Build tested: x86

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

bcm63xx: a226m-fwb: fix linux partition offset

The Pirelli A226M-FWB has a wrong linux partition offset, caused
by a copy-paste error. As of result of this, OpenWrt is currently
broken in this unit.

Fix it.

While at it, also use generic node names and fix the addresses
there as well.

Fixes: a27d59bb4274 ("brcm63xx: switch to new partition layout
specification")

Signed-off-by: Daniel González Cabanelas <dgcbueu@gmail.com>
[also fix/update node names, extend commit message]
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
(cherry picked from commit d64d5ed1425f4397d093c8777ca03f70ff1ee90c)

ipq40xx: essedma: Disable TCP segmentation offload for IPv6

It was noticed that the the whole MAC can hang when transferring data from
one ar40xx port (WAN ports) to the CPU and from the CPU back to another
ar40xx port (LAN ports). The CPU was doing only NATing in that process.

Usually, the problem first starts with a simple data corruption:

  $ wget https://cdimage.debian.org/debian-cd/current/amd64/iso-cd/debian-10.4.0-amd64-netinst.iso -O /dev/null
  ...
  Connecting to saimei.ftp.acc.umu.se (saimei.ftp.acc.umu.se)|2001:6b0:19::138|:443... connected.
  ...
  Read  error at byte 48807936/352321536 (Decryption has failed.). Retrying.

But after a short while, the whole MAC will stop to react. No traffic can
be transported anymore from the CPU port from/to the AR40xx PHY/switch and
the MAC has to be resetted.

The whole problem can be avoided by disabling the TSO for IPv6 for this
ethernet MAC driver.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
Acked-by: John Crispin <john@phrozen.org>
(backported from commit 678569505623e50bbbbc344c7e820fb315b79ede,
with updated commit message)
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>

ca-certificates: update to version 20200601

This patch updates the ca-certificates and ca-bundle package.
This version changed the files directory again, to work/, so
PKG_BUILD_DIR was brought back.

A list of changes from Debian's change-log entry for 20200601 [0]:

  * mozilla/{certdata.txt,nssckbi.h}:
    Update Mozilla certificate authority bundle to version 2.40.
Closes: #956411, #955038
  * mozilla/blacklist.txt
    Add distrusted Symantec CA list to blacklist for explicit removal.
Closes: #911289
    Blacklist expired root certificate, "AddTrust External Root"
Closes: #961907
    The following certificate authorities were added (+):
    + "Certigna Root CA"
    + "emSign ECC Root CA - C3"
    + "emSign ECC Root CA - G3"
    + "emSign Root CA - C1"
    + "emSign Root CA - G1"
    + "Entrust Root Certification Authority - G4"
    + "GTS Root R1"
    + "GTS Root R2"
    + "GTS Root R3"
    + "GTS Root R4"
    + "Hongkong Post Root CA 3"
    + "UCA Extended Validation Root"
    + "UCA Global G2 Root"
    The following certificate authorities were removed (-):
    - "AddTrust External Root"
    - "Certinomis - Root CA"
    - "Certplus Class 2 Primary CA"
    - "Deutsche Telekom Root CA 2"
    - "GeoTrust Global CA"
    - "GeoTrust Primary Certification Authority"
    - "GeoTrust Primary Certification Authority - G2"
    - "GeoTrust Primary Certification Authority - G3"
    - "GeoTrust Universal CA"
    - "thawte Primary Root CA"
    - "thawte Primary Root CA - G2"
    - "thawte Primary Root CA - G3"
    - "VeriSign Class 3 Public Primary Certification Authority - G4"
    - "VeriSign Class 3 Public Primary Certification Authority - G5"
    - "VeriSign Universal Root Certification Authority"

[0] <https://metadata.ftp-master.debian.org/changelogs//main/c/ca-certificates/ca-certificates_20200601_changelog>

Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
(cherry picked from commit f611b014a713d82d7c7da4c171f3aa04a8984063)

musl: fix locking synchronization bug

Import proposed upstream fix [2] for the critical locking
synchronization bug recently found in musl [1].

This affects all programs that are temporarily multithreaded, but then
return to single-threaded operation.

[1] https://www.openwall.com/lists/musl/2020/05/22/3
[2] https://www.openwall.com/lists/musl/2020/05/22/10

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
(cherry picked from commit 10c211031ccd4703230493025a5a3b9d6fcad2f2)

rpcd: update to latest openwrt-18.06 Git HEAD

7be1f17 file: exec: properly free memory on error
313964c file: avoid closing stdio descriptors in rpc_file_exec_run
cd09c5f file: patch process stdin to /dev/null
efbcedb file: remove unused members from struct rpc_file_exec_context
71b00ab file: rpc_file_exec_run: fix potential memory leak and integer overflow
c7bb956 plugin: fix double free in finish callback
16de3fa plugin: do not free method name separately
29c9c11 exec: properly free memory on rpc_exec() error
5cd4f4e plugin: exec: properly free memory on parse error
d80f70e plugin: fix leaking invoked method name for exec plugins
53a0952 session: deny access if password login is disabled
662d034 uci: reset uci_ptr flags when merging options during section add
dd46d6d uci: free configs list memory on return
abbc302 uci: reset uci_ptr flags when merging set operations

Signed-off-by: Jo-Philipp Wich <jo@mein.io>

usign: update to latest git HEAD

f1f65026a941 Always pad fingerprints to 16 characters

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
(cherry picked from commit e35e40ad824eab9d51cdd690fb747e576e01412f)

usign: update to latest Git HEAD

f34a383 main: fix some resource leaks

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit 81e93fff7d867851f2fedd966a931336d4092686)

squashfs: Fix compile with GCC 10

Fixes the following build error with GCC 10:
/usr/bin/ld: read_fs.o:(.bss+0x0): multiple definition of `swap'; mksquashfs.o:(.bss+0x1b2a88): first defined here
And a compile warning.

Fixes: FS#3104, FS#3119
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit 1bbc1aa884902fd05cc579b53d68b2ba0b18683f)

build: prereq: tidy gcc version checks

There is a restriction in the number of parameters(10)  that may be passed to
the SetupHostCommand macro so continually adding explicit gcc'n' version
checks ends up breaking the compiler check for the later versions and
oddballs like Darwin as was done in 835d1c68a0 which added gcc10.

Drop all the explicitly specified gcc version checks.  If a suitable gcc
compiler is not found, it may be specified at the dependency checking
stage after which that version will be symlinked into the build staging
host directory.

eg. 'CC=gccfoo CXX=g++foo make prereq'

Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
Acked-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit 1fb3c003d68d3feaf797e8b64edccc9fa622d250)

build: add GCC 10 version detection

Lets add GCC 10 detection to the build system as distributions like Fedora 32 have started shipping with it.
Some tools like mtd-utils need work to compile under GCC10, but that will be next step.

Signed-off-by: Robert Marko <robert.marko@sartura.hr>
(cherry picked from commit 835d1c68a0f036c8b0d837a48b5a05fdfb2e8218)

build: adjust gcc/g++ version checks for newer apple compilers

Signed-off-by: Felix Fietkau <nbd@nbd.name>
(cherry-picked from commit 46a129194de0f9e2f71b6526634569e1ec802504)

ramips: drop non-existant ralink,port-map for Ravpower WD03

The property "ralink,port-map" has been obsolete long before
this device was added, and the device is a one-port anyway.
Just remove it.

Fixes: 5ef79af4f80f ("ramips: add support for Ravpower WD03")
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
(cherry picked from commit c00b2df6c8e421ea7aa96f53178dc85db99f2305)

bcm63xx: mask interrupts on init

Fixes BCM6348/BCM6358 hangs while booting:
https://bugs.openwrt.org/index.php?do=details&task_id=2202

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
(cherry picked from commit 13c33f3f121ca6fe2ab1f80e04cf2d4f2cd6abec)

bcm63xx: ext_intc: fix warning

In file included from ./arch/mips/include/asm/io.h:34,
                 from ./arch/mips/include/asm/mmiowb.h:5,
                 from ./include/linux/spinlock.h:60,
                 from ./include/linux/irq.h:14,
                 from drivers/irqchip/irq-bcm6345-ext.c:10:
drivers/irqchip/irq-bcm6345-ext.c: In function 'bcm6345_ext_intc_of_init':
./arch/mips/include/asm/mach-bcm63xx/ioremap.h:48:9: warning: 'base' may be used uninitialized in this function [-Wmaybe-uninitialized]
  return is_bcm63xx_internal_registers((unsigned long)addr);
         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/irqchip/irq-bcm6345-ext.c:255:16: note: 'base' was declared here
  void __iomem *base;
                ^~~~

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
(cherry picked from commit 29c3bb5f413f0962971c66e6516b8409a78e0135)

bcm63xx: periph_intc: fix warning

drivers/irqchip/irq-bcm6345-periph.c: In function 'bcm6345_periph_irq_handle':
drivers/irqchip/irq-bcm6345-periph.c:55:21: warning: 'block' may be used uninitialized in this function [-Wmaybe-uninitialized]
  struct intc_block *block;
                     ^~~~~

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
(cherry picked from commit f2f2cf07a61e9c170ab8d65e1e054f0131a84e2a)

bcm63xx: redboot: fix warning

drivers/mtd/parsers/redboot.c: In function 'parse_redboot_partitions':
drivers/mtd/parsers/redboot.c:194:59: warning: suggest parentheses around '-' in operand of '&' [-Wparentheses]
     fis_origin = (buf[i].flash_base & (master->size << 1) - 1);
                                       ~~~~~~~~~~~~~~~~~~~~^~~

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
(cherry picked from commit f314cbe54be4d90e748e18ae5e9fb00d5fd31c91)

bcm63xx: bcm6362: fix pinctrl bug

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
(cherry picked from commit ee6bf7e865510ff1e0645b08bec65f8212ed70d4)

bcm63xx: refresh kernel config

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

mbedtls: update to 2.16.6

Security fixes for:
* CVE-2020-10932
* a potentially remotely exploitable buffer overread in a DTLS client
* bug in DTLS handling of new associations with the same parameters

Full release announement:
https://tls.mbed.org/tech-updates/releases/mbedtls-2.16.6-and-2.7.15-released

Signed-off-by: Magnus Kroken <mkroken@gmail.com>
(cherry picked from commit 02fcbe2f3d4eaf65e90bb167aa7818eacc08c633)

mbedtls: update to version 2.16.5

Changelog:
https://tls.mbed.org/tech-updates/releases/mbedtls-2.16.5-and-2.7.14-released

Security advisory:
https://tls.mbed.org/tech-updates/security-advisories/mbedtls-security-advisory-2020-02

Signed-off-by: Josef Schlehofer <pepe.schlehofer@gmail.com>
(cherry picked from commit 36af1967f5fcfc889594a8af0f92f873f445d249)

libjson-c: backport security fixes

This backports upstream fixes for the out of bounds write vulnerability in json-c.
It was reported and patches in this upstream PR: https://github.com/json-c/json-c/pull/592

Addresses CVE-2020-12762

Signed-off-by: Robert Marko <robert.marko@sartura.hr>
Signed-off-by: Luka Perkov <luka.perkov@sartura.hr>
[bump PKG_RELEASE, rebase patches on top of json-c 0.12]
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(backported from commit bc0288b76816578f5aeccb2abd679f82bfc5738e)

kernel: bump 4.14 to 4.14.180

Refreshed all patches.

Fixed:
- CVE-2020-12114
- CVE-2020-11669

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

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

kernel: bump 4.9 to 4.9.223

Refreshed all patches.

Fixes:
- CVE-2020-12114

Compile-tested on: ar71xx
Runtime-tested on: ar71xx

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

wireguard: bump to 1.0.20200506

* compat: timeconst.h is a generated artifact

Before we were trying to check for timeconst.h by looking in the kernel
source directory. This isn't quite correct on configurations in which
the object directory is separate from the kernel source directory, for
example when using O="elsewhere" as a make option when building the
kernel. The correct fix is to use $(CURDIR), which should point to
where we want.

* compat: use bash instead of bc for HZ-->USEC calculation

This should make packaging somewhat easier, as bash is generally already
available (at least for dkms), whereas bc isn't provided by distros by
default in their build meta packages.

* socket: remove errant restriction on looping to self

It's already possible to create two different interfaces and loop
packets between them. This has always been possible with tunnels in the
kernel, and isn't specific to wireguard. Therefore, the networking stack
already needs to deal with that. At the very least, the packet winds up
exceeding the MTU and is discarded at that point. So, since this is
already something that happens, there's no need to forbid the not very
exceptional case of routing a packet back to the same interface; this
loop is no different than others, and we shouldn't special case it, but
rather rely on generic handling of loops in general. This also makes it
easier to do interesting things with wireguard such as onion routing.
At the same time, we add a selftest for this, ensuring that both onion
routing works and infinite routing loops do not crash the kernel. We
also add a test case for wireguard interfaces nesting packets and
sending traffic between each other, as well as the loop in this case
too. We make sure to send some throughput-heavy traffic for this use
case, to stress out any possible recursion issues with the locks around
workqueues.

* send: cond_resched() when processing tx ringbuffers

Users with pathological hardware reported CPU stalls on CONFIG_
PREEMPT_VOLUNTARY=y, because the ringbuffers would stay full, meaning
these workers would never terminate. That turned out not to be okay on
systems without forced preemption. This commit adds a cond_resched() to
the bottom of each loop iteration, so that these workers don't hog the
core. We don't do this on encryption/decryption because the compat
module here uses simd_relax, which already includes a call to schedule
in preempt_enable.

* selftests: initalize ipv6 members to NULL to squelch clang warning

This fixes a worthless warning from clang.

* send/receive: use explicit unlikely branch instead of implicit coalescing

Some code readibility cleanups.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
(cherry picked from commit 4f6343ffe7fe8f7018f904b153dea9fc6038daf4)

wireguard: bump to 20191226

As announced on the mailing list, WireGuard will be in Linux 5.6. As a
result, the wg(8) tool, used by OpenWRT in the same manner as ip(8), is
moving to its own wireguard-tools repo. Meanwhile, the out-of-tree
kernel module for kernels 3.10 - 5.5 moved to its own wireguard-linux-
compat repo. Yesterday, releases were cut out of these repos, so this
commit bumps packages to match. Since wg(8) and the compat kernel module
are versioned and released separately, we create a wireguard-tools
Makefile to contain the source for the new tools repo. Later, when
OpenWRT moves permanently to Linux 5.6, we'll drop the original module
package, leaving only the tools. So this commit shuffles the build
definition around a bit but is basically the same idea as before.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
(cherry picked from commit ea980fb9c6de24350976dcc6c20da2bed5fc8cb8)

relayd: bump to version 2020-04-25

f4d759b dhcp.c: further improve validation

Further improve input validation for CVE-2020-11752

Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
(cherry picked from commit 9e7d11f3e275d6f5d6b3edd7f0fa0440da43c45a)

relayd: bump to version 2020-04-20

796da66 dhcp.c: improve input validation & length checks

Addresses CVE-2020-11752

Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
(cherry picked from commit be172e663f318ec364c13f795df025bbcce9ac18)

umdns: update to version 2020-04-25

cdac046 dns.c: fix input validation fix

Due to a slight foobar typo, failing to de-reference a pointer, previous
fix not quite as complete as it should have been.

Improve CVE-2020-11750 fix

Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
(cherry picked from commit 9f7c8ed0786be97eda879e5f6681994e4de53d74)

umdns: update to version 2020-04-20

e74a3f9 dns.c: improve input validation

Addresses CVE-2020-11750

Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
(cherry picked from commit 533da61ac63079f218a9946cd8e347b880c33dc0)

umdns: update to the version 2020-04-05

ab7a39a umdns: fix unused error
45c4953 dns: explicitly endian-convert all fields in header and question

Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
(cherry picked from commit 22ae8bd50ef6d056b25a96ce6c77de0b0d53c1a1)

umdns: suppress address-of-packed-member warning

gcc 8 & 9 appear to be more picky with regards access alignment to
packed structures, leading to this warning in dns.c:

dns.c:261:2: error: converting a packed ‘struct dns_question’ pointer
(alignment 1) to a ‘uint16_t’ {aka ‘short unsigned int’} pointer
(alignment 2) may result in an unaligned pointer value
[-Werror=address-of-packed-member]

261 |  uint16_t *swap = (uint16_t *) q;

Work around what I think is a false positive by turning the warning off.
Not ideal, but not quite as not ideal as build failure.

Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
(cherry picked from commit 02640f014719a994e2e538b2cb6376a189cd39de)

ramips: remove unnecessary DEVICE_PACKAGES for Belkin F7C027

kmod-usb-dwc2 and kmod-usb-ledtrig-usbport are not target default packages, and
Belkin F7C027 does not have a USB port anyway. Just drop it.

Signed-off-by: Sungbo Eo <mans0n@gorani.run>
(cherry picked from commit 1dedad2a00d8519d4892b8529b5f2fffcb768220)

oxnas: move service file to correct place

This service file has been misplaced from the very beginning.

Fixes: dcc34574efba ("oxnas: bring in new oxnas target")
Signed-off-by: Sungbo Eo <mans0n@gorani.run>
(cherry picked from commit 01961f163d927d6b44097f48a67bbc5b4c63eaf7)

kernel: bump 4.14 to 4.14.176

Refreshed all patches.

Remove upstreamed:
- 0001-net-thunderx-workaround-BGX-TX-Underflow-issue.patch
- 600-ipv6-addrconf-call-ipv6_mc_up-for-non-Ethernet-inter.patch

Fixes:
- CVE-2020-8647
- CVE-2020-8648 (potentially)
- CVE-2020-8649

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

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

kernel: bump 4.9 to 4.9.219

Refreshed all patches.

Altered patches:
- 0026-NET-multi-phy-support.patch

Fixes:
- CVE-2020-8647
- CVE-2020-8648 (Potentially)
- CVE-2020-8649

Compile-tested on: ar71xx
Runtime-tested on: ar71xx

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

kernel: add missing symbol for Kernel 4.14

Reported by Buildbot:

x86 instruction decoder selftest (X86_DECODER_SELFTEST) [N/y/?] (NEW) aborted!

Fixes: 4eba86820fd3 ("kernel: bump 4.14 to 4.14.169")
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
[rebased onto 18.06]
Signed-off-by: David Bauer <mail@david-bauer.net>

ramips: use full 8MB flash on ZyXEL Keenetic

ZyXEL Keenetic has 8MB flash, but OpenWrt uses only 4MB.
This commit fixes the problem.

WikiDevi page [1] says that ZyXEL Keenetic has FLA1: 8 MiB, there is
an article with specs [2] (in Russian).

[1] https://wikidevi.wi-cat.ru/ZyXEL_Keenetic
[2] https://3dnews.ru/608774/page-2.html

Fixes: FS#2487
Fixes: a7cbf59e0e04 ("ramips: add new device ZyXEL Keenetic as kn")
Signed-off-by: Alexey Dobrovolsky <dobrovolskiy.alexey@gmail.com>
(cherry picked from commit fea232ae8feb6af780fd4fa78ebe9231778bf75a)

Revert "ar71xx: use status led for GL.iNet GL-AR750S"

This reverts commit c3c6cc95eeadf5230d77d5afabc9a9e0b4887ee5.

The GL.iNet GL-AR750S is not supported in 18.06.

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

ar71xx: use status led for GL.iNet GL-AR750S

Use power led for device status.

The status led behavior has already been fixed in af28d8a539fe
("ath79: add support for GL.iNet GL-AR750S") when porting the
device to ath79. This fixes it for ar71xx as well.

Signed-off-by: Jan Alexander <jan@nalx.net>
[minor commit title/message adjustments]
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
(cherry picked from commit d394c354ee0e8660f876889f6293803c581cbf85)

ar71xx: fix port order on TP-Link Archer C60 v1/v2

The labels on the LAN ports of the TP-Link Archer C60 v1/v2 are
actually inverted compared to the ports of the internal switch.

Add this information to 02_network.

Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
(cherry picked from commit 14a07fa1f000f962d55bf53cfe9b128448b1f11f)

ar71xx: remove wrong MAC address adjustment for Archer C60 v2

The adjustment of the MAC address for Archer C60 v2 in 10_fix_wifi_mac
is broken since a "mac" partition is not set up for this device on
ar71xx. Instead, the MAC address is already patched correctly in
11-ath10k-caldata.

Remove the useless adjustment.

Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
(cherry picked from commit cbdc91902466dd113b9680566abb2eb2f8022dd4)

ar71xx: fix swapped LAN/WAN MAC address for Archer C60 v1/v2

The MAC addresses for lan/wan are swapped compared to the vendor
firmware. This adjusts to vendor configuration, which is:

lan   *:7b   label
wan   *:7c   label+1
2.4g  *:7b   label
5g    *:7a   label-1

Only one address is stored in <&mac 0x8>, corresponding to the label.

This has been checked on revisions v1, v2 and v3.

Since ar71xx calculates the ath10k MAC address based on the ethernet
addresses, the number there is adjusted, too.

Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
(cherry picked from commit 14eb54938b525d4c2d2847a6fad7670ed944650c)

ar71xx: ew-dorin, fix the trigger level for WPS button

Because the WPS button had the wrong trigger level,
the failsafe mode was triggered quite often,
after this commit:
https://git.openwrt.org/?p=openwrt/openwrt.git;a=commit;h=27f3f493de

Signed-off-by: Catrinel Catrinescu <cc@80211.de>
(cherry picked from commit 3e03b7ac4ada4966d058871d0a1f67f1f0f408d4)

kernel: backport out-of-memory fix for non-Ethernet devices

Doing up & down on non-Ethernet devices (e.g. monitor mode interface)
was consuming memory.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
(cherry picked from commit ec8e8e2ef0826d82b4dfbd567a073b31dc27b764)

kernel: bump 4.14 to 4.14.172

Refreshed all patches.

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

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

kernel: bump 4.9 to 4.9.215

Refreshed all patches.

Compile-tested on: ar71xx
Runtime-tested on: ar71xx

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

OpenWrt v18.06.8: revert to branch defaults

Signed-off-by: Jo-Philipp Wich <jo@mein.io>

OpenWrt v18.06.8: adjust config defaults

Signed-off-by: Jo-Philipp Wich <jo@mein.io>

libubox: backport blobmsg_check_array() fix

Fixes: FS#2833
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(backported from commit 955634b473284847e3c8281a6ac85655329d8b06)

ppp: backport security fixes

8d45443bb5c9 pppd: Ignore received EAP messages when not doing EAP
8d7970b8f3db pppd: Fix bounds check in EAP code
858976b1fc31 radius: Prevent buffer overflow in rc_mksid()

Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit 215598fd03899c19a9cd26266221269dd5ec8cee)
Fixes: CVE-2020-8597
Signed-off-by: Jo-Philipp Wich <jo@mein.io>

Revert "ppp: backport security fixes"

This reverts commit cc78f934a9466a0ef404bb169cc42680c7501d02 since it
didn't contain a reference to the CVE it addresses. The next commit
will re-add the commit including a CVE reference in its commit message.

Signed-off-by: Jo-Philipp Wich <jo@mein.io>

uhttpd: update to latest Git HEAD

2ee323c file: poke ustream after starting deferred program

Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit 04069fde19e86af7728111814afadf780bf08018)

kernel: bump 4.14 to 4.14.171

Refreshed all patches.

Fixes:
- CVE-2013-1798

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

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

kernel: bump 4.9 to 4.9.214

Refreshed all patches.

Fixes:
- CVE-2013-1798

Compile-tested on: ar71xx
Runtime-tested on: ar71xx

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

ppp: backport security fixes

8d45443bb5c9 pppd: Ignore received EAP messages when not doing EAP
8d7970b8f3db pppd: Fix bounds check in EAP code
858976b1fc31 radius: Prevent buffer overflow in rc_mksid()

Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit 215598fd03899c19a9cd26266221269dd5ec8cee)

hostapd: remove erroneous $(space) redefinition

The $(space) definition in the hostapd Makefile ceased to work with
GNU Make 4.3 and later, leading to syntax errors in the generated
Kconfig files.

Drop the superfluous redefinition and reuse the working $(space)
declaration from rules.mk to fix this issue.

Fixes: GH#2713
Ref: https://github.com/openwrt/openwrt/pull/2713#issuecomment-583722469
Reported-by: Karel Kočí <cynerd@email.cz>
Suggested-by: Jonas Gorski <jonas.gorski@gmail.com>
Tested-by: Shaleen Jain <shaleen@jain.sh>
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(backported from commit 766e778226f5d4c6ec49ce22b101a5dbd4306644)

kernel: add support for GD25D05 SPI NOR

This chip is used on newer RB912UAG-5HPnD r2 boards:

Before:

[    0.642553] m25p80 spi0.0: unrecognized JEDEC id bytes: c8, 40, 10
[    0.649381] NAND flash driver for the RouterBOARD 91x series

After:

[    0.641714] m25p80 spi0.0: found gd25d05, expected m25p80
[    0.649916] m25p80 spi0.0: gd25d05 (64 Kbytes)
[    0.655122] Creating 4 MTD partitions on "spi0.0":
[    0.660164] 0x000000000000-0x00000000c000 : "routerboot"
[    0.667782] 0x00000000c000-0x00000000d000 : "hard_config"
[    0.675073] 0x00000000d000-0x00000000e000 : "bios"
[    0.682613] 0x00000000e000-0x00000000f000 : "soft_config"
[    0.690304] NAND flash driver for the RouterBOARD 91x series

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

kernel: bump 4.14 to 4.14.169

Refreshed all patches.

Fixes:
- CVE-2019-14896
- CVE-2019-14897

Remove upstreamed:
- 023-0007-crypto-crypto4xx-Fix-wrong-ppc4xx_trng_probe-ppc4xx_.patch

Altered patches:
- 102-MIPS-BCM63XX-move-code-touching-the-USB-private-regi.patch
- 105-MIPS-BCM63XX-add-support-for-the-on-chip-OHCI-contro.patch
- 106-MIPS-BCM63XX-register-OHCI-controller-if-board-enabl.patch
- 108-MIPS-BCM63XX-add-support-for-the-on-chip-EHCI-contro.patch
- 207-MIPS-BCM63XX-move-device-registration-code-into-its-.patch
- 350-MIPS-BCM63XX-support-settings-num-usbh-ports.patch
- 356-MIPS-BCM63XX-move-fallback-sprom-support-into-its-ow.patch
- 390-MIPS-BCM63XX-do-not-register-SPI-controllers.patch
- 391-MIPS-BCM63XX-do-not-register-uart.patch
- 392-MIPS-BCM63XX-remove-leds-and-buttons.patch
- 416-BCM63XX-add-a-fixup-for-ath9k-devices.patch
- 422-BCM63XX-add-a-fixup-for-rt2x00-devices.patch
-

Compile-tested on: brcm63xx, cns3xxx
Runtime-tested on: cns3xxx

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

kernel: bump 4.9 to 4.9.212

Refreshed all patches.

Fixes:
- CVE-2019-14896
- CVE-2019-14897

Compile-tested on: ar71xx
Runtime-tested on: ar71xx

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

OpenWrt v18.06.7: revert to branch defaults

Signed-off-by: Jo-Philipp Wich <jo@mein.io>

OpenWrt v18.06.7: adjust config defaults

Signed-off-by: Jo-Philipp Wich <jo@mein.io>

opkg: update to latest Git HEAD

80d161e opkg: Fix -Wformat-overflow warning
c09fe20 libopkg: fix skipping of leading whitespace when parsing checksums

Fixes: CVE-2020-7982
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit c69c20c6670081d1eaab000734d89de57eb64148)

libubox: backport security patches

This backports some security relevant patches from libubox master. These
patches should not change the existing API and ABI so that old
applications still work like before without any recompilation.
Application can now also use more secure APIs.

The new more secure interfaces are also available, but not used.

OpenWrt master and 19.07 already have these patches by using a more
recent libubox version.

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

lantiq: ltq-ptm: vr9: fix skb handling in ptm_hard_start_xmit()

Call skb_orphan(skb) to call the owner's destructor function and make
the skb unowned.

This is necessary to prevent sk_wmem_alloc of a socket from overflowing,
which leads to ENOBUFS errors on application level.

Signed-off-by: Martin Schiller <ms@dev.tdt.de>
(cherry picked from commit 996f02e5bafad2815e72821c19d41fb5297e4dad)

tools/expat: Update to version 2.2.9

Fixes two CVEs:
- CVE-2019-15903 (Fix heap overflow triggered by XML_GetCurrentLineNumber)
- CVE-2018-20843 (Fix extraction of namespace prefixes from XML names)

Signed-off-by: Josef Schlehofer <pepe.schlehofer@gmail.com>
(cherry picked from commit b4af2c689fc8736777940b7bbf009bb1672296ec)

mbedtls: update to 2.16.4

Fixes side channel vulnerabilities in mbed TLS' implementation of ECDSA.

Release announcement:
https://tls.mbed.org/tech-updates/releases/mbedtls-2.16.4-and-2.7.13-released

Security advisory:
https://tls.mbed.org/tech-updates/security-advisories/mbedtls-security-advisory-2019-12

Fixes:
 * CVE-2019-18222: Side channel attack on ECDSA

Signed-off-by: Magnus Kroken <mkroken@gmail.com>
(cherry picked from commit 6e96fd90471a49185bcfe9dcb4844d444674ecab)

kernel: bump 4.14 to 4.14.167

Refreshed all patches.

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

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

kernel: bump 4.9 to 4.9.211

Refreshed all patches.

Compile-tested on: ar71xx
Runtime-tested on: ar71xx

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

kirkwood: fix HDD LED labels for Zyxel NSA325 in 01_leds

Change the LED labels for hdd1/hdd2 in 01_leds to match their
counterpart in DTS.

Signed-off-by: Stephan Knauss <openwrt@stephans-server.de>
[improve commit title and message]
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
(cherry picked from commit fbf297be38a93b9ca1119e5aaffecd2299087aa5)

brcm47xx: fix switch port order for Netgear WN2500RP V1

The Netgear WN2500RP V1 switch0 already works for LAN
however the port order for the LAN ports is inverted. Correct
physical port order watched from the back of the device is:
4 / 3 / 2 / 1
WAN port is absent on this device and therefore removed
from switch config.

Signed-off-by: Walter Sonius <walterav1984@gmail.com>
[move block to maintain alphabetic sorting]
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
(cherry picked from commit 098cbc68ee23db589ed6f0d081fe26cc385462f2)

brcm47xx: fix switch port order for Netgear WNR3500 V2

The Netgear WNR3500 V2 switch0 already works for WAN/LAN
however the port order for the LAN ports is inverted. Correct
physical port order watched from the back of the device is:
Internet / 4 / 3 / 2 / 1 this resembles the Linksys E3000 V1.

Verfied with imagebuilder edit FILES=/etc/board.d/01_network

Signed-off-by: Walter Sonius <walterav1984@gmail.com>
(cherry picked from commit cf2f1fc6871da0320afeefaa799af87fc7c0d1db)

ramips: fix HiWiFi HC5962 switch configuration

HC5962 has only 3 LAN ports, switch port 0 is unused

Signed-off-by: DENG Qingfang <dengqf6@mail2.sysu.edu.cn>
(backported from commit 68f49df31507454f86b72a5c1e250505176baed7)
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>

kernel: bump 4.14 to 4.14.166

Refreshed all patches.

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

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

kernel: bump 4.14 to 4.14.165

Refreshed all patches.

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

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

kernel: bump 4.9 to 4.9.210

Refreshed all patches.

Compile-tested on: ar71xx
Runtime-tested on: ar71xx

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

kernel: bump 4.14 to 4.14.164

Refreshed all patches.

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

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

kernel: bump 4.9 to 4.9.209

Refreshed all patches.

Compile-tested on: ar71xx
Runtime-tested on: ar71xx

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

ethtool: fix PKG_CONFIG_DEPENDS

Add missing CONFIG_ prefix.

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
(cherry picked from commit 41c19dd542973dbc1336ecceaa32777506933cdf)

OpenWrt v18.06.6: revert to branch defaults

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

OpenWrt v18.06.6: adjust config defaults

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

ramips: remove duplicate dts nodes of MediaTek LinkIt Smart 7688

There are two identical wmac nodes in the dts file of MediaTek
LinkIt Smart 7688, so delete one of them.

Signed-off-by: Jack Chen <redchenjs@live.com>
(cherry picked from commit 4be271a4867dcf57a122eeb7b42407e7a506915e)

sunxi: Turn on CONFIG_PINCTRL_SUN4I_A10 for A20

CONFIG_PINCTRL_SUN4I_A10 controls both the A10 and the A20 enablong of
the pinctrl driver, this is necessary since upstream commit
5d8d349618a9464714c07414c5888bfd9416638f ("pinctrl: sunxi: add A20
support to A10 driver") which has been included in v4.13 and onwards.

Fixes: ad2b3bf310f7 ("sunxi: Add support for kernel 4.14")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
(cherry picked from commit 32e4eaef1b4c2a7fa44787813fdf715b2ba500d9)

kernel: bump 4.9 to 4.9.208

Refreshed all patches.

Compile-tested on: lantiq
Runtime-tested on: lantiq

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

kernel: bump 4.14 to 4.14.162

Refreshed all patches.

Compile-tested on: ramips
Runtime-tested on: ramips

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

ramips: fix inverted reset button for Ravpower WD03

The button events "pressed" and "released" were switched. Tested with v18.06.4.

Signed-off-by: Moritz Warning <moritzwarning@web.de>
(cherry picked from commit 3e1325b219fced91f01d5594503f61d326a93b90)

ar71xx: really fix Mikrotik board detection

commit e09da0169a08 ("ar71xx: fix Mikrotik board detection")
was generated based on testing a rb-912 board, on which detection failed.

Testing on more hardware shows something fun:

machine : MikroTik RouterBOARD 922UAGS-5HPacD
machine : Mikrotik RouterBOARD 912UAG-5HPnD

Both lowercase and uppercase are used.
So ensure we support both now ..

Fixes: e09da0169a08 ("ar71xx: fix Mikrotik board detection")
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
(cherry picked from commit 845b2a1cfe307394f3f2144cce085bbb5e171ecc)

ar71xx: fix Mikrotik board detection

Fix a typo in the machine type being extracted from /proc/cpuinfo
which causes all Mikrotik board to be undetected properly.

This lead to sysupgrade issues and probably some others too.

Fixes: acf2b6c8881b ("ar71xx: base-files: fix board detect on new MikroTik devices")
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
(cherry picked from commit e09da0169a088663f6dab1f71b2a4b690eab21d3)

ar71xx: base-files: fix board detect on new MikroTik devices

Move all MikroTik devices to new function to increase script execution
speed.

Machine name in new version of MikroTik RouterBOARD devices add "RB"
before model name:

 Old machine name: MikroTik RouterBOARD 951Ui-2nD
 New:              MikroTik RouterBOARD RB951Ui-2nD

So this patch should fix it for all currently supported MikroTik boards.

Signed-off-by: Henryk Heisig <hyniu@o2.pl>
[rebased,commit message facelift,script fixes]
Signed-off-by: Petr Štetiar <ynezz@true.cz>
[spotted missing 922UAGS-5HPacD]
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
(cherry picked from commit acf2b6c8881b432530bd98fa86753bf6a3546ff7)
[backport: do not add boards not supported in 18.06]
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>

ar71xx: fix RB941-2nD detection

Some hAP lite routers aren't detected because
/proc/cpuinfo shows "RouterBOARD RB941-2nD"
instead of "RouterBOARD 941-2nD".

Fix that.

Signed-off-by: Julien Rabier <taziden@flexiden.org>
[Alter string to include all flavours + slight rewrite of commit msg]
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
(cherry picked from commit 6570f3c93aa4110efd1466a6b89742c2e11d7c11)