openwrt/staging/hauke.git
19 months agompc85xx: add localbus frequency for WS-AP3825i
David Bauer [Sun, 2 Apr 2023 02:25:00 +0000 (04:25 +0200)]
mpc85xx: add localbus frequency for WS-AP3825i

This is normally filled by U-Boot.

Signed-off-by: David Bauer <mail@david-bauer.net>
19 months agompc85xx: add linux,stdout-path for WS-AP3825i
David Bauer [Sun, 2 Apr 2023 02:11:12 +0000 (04:11 +0200)]
mpc85xx: add linux,stdout-path for WS-AP3825i

This is normally filled by U-Boot. Prevents double-printing of early
console messages. Also enables debug-output by the zImage wrapper.

Signed-off-by: David Bauer <mail@david-bauer.net>
19 months agoramips: define remapping-range for DAP-X1860
David Bauer [Thu, 30 Mar 2023 01:51:08 +0000 (03:51 +0200)]
ramips: define remapping-range for DAP-X1860

Prevent the BBT translation layer from remapping the UBI used for
storing rootfs.

Explicitly define the number of blocks reserved for remapping.

Signed-off-by: David Bauer <mail@david-bauer.net>
19 months agoocteon: switch to Kernel 5.15 by default
Paul Spooren [Wed, 22 Feb 2023 18:59:58 +0000 (19:59 +0100)]
octeon: switch to Kernel 5.15 by default

Getting ready for the next release.

Signed-off-by: Paul Spooren <mail@aparcar.org>
19 months agokirkwood: switch to Kernel 5.15 by default
Paul Spooren [Wed, 22 Feb 2023 18:53:58 +0000 (19:53 +0100)]
kirkwood: switch to Kernel 5.15 by default

Getting ready for the next release.

Signed-off-by: Paul Spooren <mail@aparcar.org>
19 months agotegra: switch to Kernel 5.15 by default
Paul Spooren [Wed, 22 Feb 2023 19:07:30 +0000 (20:07 +0100)]
tegra: switch to Kernel 5.15 by default

Getting ready for the next release.

Acked-by: Tomasz Maciej Nowak <tmn505@gmail.com>
Signed-off-by: Paul Spooren <mail@aparcar.org>
19 months agoopenssl: fix CVE-2023-464 and CVE-2023-465
Eneas U de Queiroz [Tue, 4 Apr 2023 18:39:56 +0000 (15:39 -0300)]
openssl: fix CVE-2023-464 and CVE-2023-465

Apply two patches fixing low-severity vulnerabilities related to
certificate policies validation:

- Excessive Resource Usage Verifying X.509 Policy Constraints
  (CVE-2023-0464)
  Severity: Low
  A security vulnerability has been identified in all supported versions
  of OpenSSL related to the verification of X.509 certificate chains
  that include policy constraints.  Attackers may be able to exploit
  this vulnerability by creating a malicious certificate chain that
  triggers exponential use of computational resources, leading to a
  denial-of-service (DoS) attack on affected systems.
  Policy processing is disabled by default but can be enabled by passing
  the `-policy' argument to the command line utilities or by calling the
  `X509_VERIFY_PARAM_set1_policies()' function.

- Invalid certificate policies in leaf certificates are silently ignored
  (CVE-2023-0465)
  Severity: Low
  Applications that use a non-default option when verifying certificates
  may be vulnerable to an attack from a malicious CA to circumvent
  certain checks.
  Invalid certificate policies in leaf certificates are silently ignored
  by OpenSSL and other certificate policy checks are skipped for that
  certificate.  A malicious CA could use this to deliberately assert
  invalid certificate policies in order to circumvent policy checking on
  the certificate altogether.
  Policy processing is disabled by default but can be enabled by passing
  the `-policy' argument to the command line utilities or by calling the
  `X509_VERIFY_PARAM_set1_policies()' function.

Note: OpenSSL also released a fix for low-severity security advisory
CVE-2023-466.  It is not included here because the fix only changes the
documentation, which is not built nor included in any OpenWrt package.

Due to the low-severity of these issues, there will be not be an
immediate new release of OpenSSL.

Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
19 months agorealtek: Fix typo for EnGenius EWS2910P
Raylynn Knight [Sun, 2 Apr 2023 08:27:21 +0000 (04:27 -0400)]
realtek: Fix typo for EnGenius EWS2910P

Fix mis-typed DEVICE-MODEL in mk file for EnGenius EWS2910P.

Signed-off-by: Raylynn Knight <rayknight@me.com>
[ fix wrong SoB format and improve commit title/description ]
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
19 months agotools/zstd: update to 1.5.5
Nick Hainke [Wed, 5 Apr 2023 11:30:14 +0000 (13:30 +0200)]
tools/zstd: update to 1.5.5

Release Notes:
https://github.com/facebook/zstd/releases/tag/v1.5.5

Signed-off-by: Nick Hainke <vincent@systemli.org>
19 months agomac80211: ath11k: Fix invalid mgmt rx frame length issue
Robert Marko [Tue, 4 Apr 2023 19:49:43 +0000 (21:49 +0200)]
mac80211: ath11k: Fix invalid mgmt rx frame length issue

FW 2.9 uses multiple TLV-s for the RX mgmt even which driver currently does
not support, so import a pending upstream patch to fix that [1].

[1] https://patchwork.kernel.org/project/linux-wireless/patch/20230320133840.30162-1-quic_nmaran@quicinc.com/

Signed-off-by: Robert Marko <robimarko@gmail.com>
19 months agoath11k-firmware: update to WLAN.HK.2.9.0.1-01385-QCAHKSWPL_SILICONZ-1
Robert Marko [Tue, 4 Apr 2023 19:48:42 +0000 (21:48 +0200)]
ath11k-firmware: update to WLAN.HK.2.9.0.1-01385-QCAHKSWPL_SILICONZ-1

Current WLAN.HK.2.5.0.1 FW is quite old and buggy, but we had to hold off
from updating to 2.6.0.1 and 2.7.0.1 as they had compatibility regressions,
but now QCA finally released 2.9.0.1 FW which is working on all of the
boards.

So finally update IPQ8074 and QCN9074 FW to the latest
WLAN.HK.2.9.0.1-01385-QCAHKSWPL_SILICONZ-1 firmware.

In order to do so, we have to switch to using QCA-s QUIC repo instead of
Kalle-s.
QCA-s QUIC repo does not have BDF-s so we have to get the QCN9074 BDF from
Kalles repo.

Tested-by: Mireia Fernández Casals <meirin.f@gmail.com> # Xiaomi AX3600
Tested-by: Francisco G Luna <frangonlun@gmail.com> #Netgear WAX218
Signed-off-by: Robert Marko <robimarko@gmail.com>
19 months agokernel: backport NVMEM patch for U-Boot env data "ethaddr" cell
Rafał Miłecki [Thu, 6 Apr 2023 10:14:30 +0000 (12:14 +0200)]
kernel: backport NVMEM patch for U-Boot env data "ethaddr" cell

Adjust our local code to avoid breakage.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
19 months agokernel: backport NVMEM patches queued for the v6.4
Rafał Miłecki [Thu, 6 Apr 2023 07:15:49 +0000 (09:15 +0200)]
kernel: backport NVMEM patches queued for the v6.4

They add NVMEM layouts support. It allows handling NVMEM content
independently of NVMEM device access.

Skip U-Boot env data patch for now as it break our downstream MAC hacks.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
19 months agokernel: backport mtd fixes for nvmem
Rafał Miłecki [Thu, 6 Apr 2023 05:26:11 +0000 (07:26 +0200)]
kernel: backport mtd fixes for nvmem

They are needed for NVMEM changes pending for v6.4.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
19 months agokernel: backport of_request_module()
Rafał Miłecki [Wed, 5 Apr 2023 21:32:57 +0000 (23:32 +0200)]
kernel: backport of_request_module()

It's needed by NVMEM changes queued for 6.4.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
19 months agomediatek: introduce KERNEL_LOADADDR to Device/Default template
Daniel Golle [Wed, 5 Apr 2023 15:56:54 +0000 (16:56 +0100)]
mediatek: introduce KERNEL_LOADADDR to Device/Default template

We need to reset KERNEL_LOADADDR if we use it on a per-device base.
Otherwise the previous value will be kept in case a device doesn't
define KERNEL_LOADADDR and relies on the default.

Move initializing KERNEL_LOADADDR to target/linux/mediatek/image/Makefile,
similar to how it's done also on the ramips target.

This fixes image size related breakage on devices which rely on the
default value of KERNEL_LOADADDR.

While at it use 0x48000000 which is more common than the previous default
0x44000000 for the filogic subtarget.

Fixed: e7c399bee6 ("filogic: add support for ASUS TUF-AX4200")
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
19 months agoopenssl: add legacy provider
Eneas U de Queiroz [Fri, 10 Mar 2023 20:53:25 +0000 (17:53 -0300)]
openssl: add legacy provider

This adapts the engine build infrastructure to allow building providers,
and packages the legacy provider.  Providers are the successors of
engines, which have been deprecated.

The legacy provider supplies OpenSSL implementations of algorithms that
have been deemed legacy, including DES, IDEA, MDC2, SEED, and Whirlpool.

Even though these algorithms are implemented in a separate package,
their removal makes the regular library smaller by 3%, so the build
options will remain to allow lean custom builds.  Their defaults will
change to 'y' if not bulding for a small flash, so that the regular
legacy package will contain a complete set of algorithms.

The engine build and configuration structure was changed to accomodate
providers, and adapt to the new style of openssl.cnf in version 3.0.

There is not a clean upgrade path for the /etc/ssl/openssl.cnf file,
installed by the openssl-conf package.  It is recommended to rename or
remove the old config file when flashing an image with the updated
openssl-conf package, then apply the changes manually.

An old openssl.cnf file will silently work, but new engine or provider
packages will not be enabled.  Any remaining engine config files under
/etc/ssl/engines.cnf.d can be removed.

On the build side, the include file used by engine packages was renamed
to openssl-module.mk, so the engine packages in other feeds need to
adapt.

Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
19 months agoopenssl: make UCI config aware of built-in engines
Eneas U de Queiroz [Tue, 14 Mar 2023 21:51:45 +0000 (18:51 -0300)]
openssl: make UCI config aware of built-in engines

Engines that are built into the main libcrypto OpenSSL library can't be
disabled through UCI.  Add a 'builtin' setting to signal that the engine
can't be disabled through UCI, and show a message explaining this in
case buitin=1 and enabled=0.

Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
19 months agoopenssl: avoid OPENSSL_SMALL_FOOTPRINT, no-asm
Eneas U de Queiroz [Fri, 10 Mar 2023 20:21:11 +0000 (17:21 -0300)]
openssl: avoid OPENSSL_SMALL_FOOTPRINT, no-asm

Building openssl with OPENSSL_SMALL_FOOTPRINT yelds only from 1% to 3%
decrease in size, dropping performance from 2% to 91%, depending on the
target and algorithm.

For example, using AES256-GCM with 1456-bytes operations, X86_64 appears
to be the least affected with 2% performance penalty and 1% reduction in
size; mips drops performance by 13%, size by 3%;  Arm drops 29% in
performance, 2% in size.

On aarch64, it slows down ghash so much that I consider it broken
(-91%).  SMALL_FOOTPRINT will reduce AES256-GCM performance by 88%, and
size by only 1%.  It makes an AES-capable CPU run AES128-GCM at 35% of
the speed of Chacha20-Poly1305:

Block-size=1456 bytes   AES256-GCM   AES128-GCM  ChaCha20-Poly1305
SMALL_FOOTPRINT           62014.44     65063.23          177090.50
regular                  504220.08    565630.28          182706.16

OpenSSL 1.1.1 numbers are about the same, so this should have been
noticed a long time ago.

This creates an option to use OPENSSL_SMALL_FOOTPRINT, but it is turned
off by default unless SMALL_FLASH or LOW_MEMORY_FOOTPRINT is used.

Compiling with -O3 instead of -Os, for comparison, will increase size by
about 14-15%, with no measureable effect on AES256-GCM performance, and
about 2% increase in Chacha20-Poly1305 performance on Aarch64.

There are no Arm devices with the small flash feature, so drop the
conditional default.  The package is built on phase2, so even if we
include an Arm device with small flash later, a no-asm library would
have to be built from source anyway.

Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
19 months agoodhcpd: bump to latest git HEAD
Christian Marangi [Tue, 4 Apr 2023 04:43:23 +0000 (06:43 +0200)]
odhcpd: bump to latest git HEAD

40ab806 config: use dedicated link local function to check interface
a84bff2 netlink: add support for getting interface linklocal
2ea065f Revert "config: recheck have_link_local on interface reload if already init"
4b38e6b config: fix feature for enabling service only when interface RUNNING

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
19 months agomediatek: make TUF-AX4200 KERNEL simply expanded
David Bauer [Tue, 4 Apr 2023 00:08:32 +0000 (02:08 +0200)]
mediatek: make TUF-AX4200 KERNEL simply expanded

The KERNEL variable was recursively expanded, breaking other devices.

Signed-off-by: David Bauer <mail@david-bauer.net>
19 months agogeneric: disable SGMII in-band AN for RealTek 2.5G PHYs
Daniel Golle [Sun, 2 Apr 2023 22:21:57 +0000 (01:21 +0300)]
generic: disable SGMII in-band AN for RealTek 2.5G PHYs

MAC drivers don't use SGMII in-band autonegotiation unless told to do so
in device tree using 'managed = "in-band-status"'. When using MDIO to
access a PHY, in-band-status is unneeded as we have link-status via
MDIO. Switch off SGMII in-band autonegotiation using magic values.

Reported-by: Chen Minqiang <ptpt52@gmail.com>
Reported-by: Chukun Pan <amadeus@jmu.edu.cn>
Reported-by: Yevhen Kolomeiko <jarvis2709@gmail.com>
Tested-by: Yevhen Kolomeiko <jarvis2709@gmail.com>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
19 months agogeneric: use genphy_soft_reset for RealTek 2.5G PHYs
Daniel Golle [Sun, 2 Apr 2023 22:20:28 +0000 (01:20 +0300)]
generic: use genphy_soft_reset for RealTek 2.5G PHYs

Some vendor bootloaders do weird things with those PHYs which result in
link modes being reported wrongly. Start from a clean sheet by resetting
the PHY.

Reported-by: Yevhen Kolomeiko <jarvis2709@gmail.com>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
19 months agogeneric: interface mode switch for RTL8221B PHY
Chukun Pan [Wed, 8 Feb 2023 15:40:02 +0000 (23:40 +0800)]
generic: interface mode switch for RTL8221B PHY

Add dynamic interface mode update for the rtl8221 phy to match various
wire speeds. 10M/100M/1000M use SGMII, 2500M uses 2500Base-X.

Signed-off-by: Chukun Pan <amadeus@jmu.edu.cn>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
19 months agorealtek: refresh patches
Daniel Golle [Mon, 3 Apr 2023 01:29:17 +0000 (02:29 +0100)]
realtek: refresh patches

Refresh patches which were no longer applying cleanly after a recently
added SFP quirk.

Fixes: 658b45ce48 ("generic: add quirk for HG MXPD-483II 2500M fiber SFP")
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
19 months agoRevert "mpc85xx: remove Watchguard T10 MAC override"
David Bauer [Mon, 3 Apr 2023 01:06:45 +0000 (03:06 +0200)]
Revert "mpc85xx: remove Watchguard T10 MAC override"

This reverts commit aa4a9058fbf4d2ca7ef3af9ebb5bbec896625e93.

The assumption the bootloader fills out the MAC-address is not
correct. The MAC-address has to be set from userspace based on
information found in the device_id partition.

Signed-off-by: David Bauer <mail@david-bauer.net>
19 months agogeneric: add quirk for HG MXPD-483II 2500M fiber SFP
Daniel Golle [Sun, 2 Apr 2023 23:06:46 +0000 (00:06 +0100)]
generic: add quirk for HG MXPD-483II 2500M fiber SFP

The HG MXPD-483II 1310nm SFP module is meant to operate with 2500Base-X,
however, in their EEPROM they incorrectly specify:
    Transceiver type                          : Ethernet: 1000BASE-LX
    ...
    BR, Nominal                               : 2600MBd

Use sfp_quirk_2500basex for this module to allow 2500Base-X mode anyway.

https://forum.banana-pi.org/t/bpi-r3-sfp-module-compatibility/14573/60

X-Patchwork-Id: 13197378
X-Mailing-List: netdev@vger.kernel.org
X-Patchwork-Delegate: kuba@kernel.org
Reported-by: chowtom <chowtom@gmail.com>
Tested-by: chowtom <chowtom@gmail.com>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
19 months agolantiq: fix lzma-loader for Netgear DGN 3500(B)
Maik Goette [Sun, 8 Jan 2023 17:35:10 +0000 (18:35 +0100)]
lantiq: fix lzma-loader for Netgear DGN 3500(B)

Fixes Uncompressing Kernel Image ... ERROR: LzmaDecode.c, 561

Fixes: #11701 for both 3500(B)
Signed-off-by: Maik Goette <github@beeit.de>
19 months agotools/7z: Allow building on alpine
Olliver Schinagl [Thu, 16 Feb 2023 14:25:12 +0000 (15:25 +0100)]
tools/7z: Allow building on alpine

When using alpine as host, things start to fail. Lets pull in the
upstream alpine patches to make things work. This should not affect
other hosts.

Note, that Alpine has the '_GNU_SOURCE' define in the APKBUILD file, but
here we add this flag to the needed fix flags patch, which does similar
things too.

Signed-off-by: Olliver Schinagl <oliver@schinagl.nl>
19 months agoramips: add support for OrayBox X1
Bin We [Thu, 22 Sep 2022 17:04:01 +0000 (01:04 +0800)]
ramips: add support for OrayBox X1

Add support for OrayBox X1. It is a 802.11n router, based on MediaTek MT7628N.

Specifications:
SoC: MediaTek MT7628N (580MHz)
RAM: 64 MiB
Flash: 16 MiB NOR (Winbond W25Q128JVSIQ)
Wireless: 802.11b/g/n 2x2 2.4GHz (Built In)
Ethernet: 1x 100Mbps only
USB: 1x USB Type-A 2.0 Host Port
Button: 1x "Reset" button
LED: 1x Blue LED + 1x Red LED + 1x White LED
Power: 5V Micro-USB input

Manufacturer Page:
https://pgy.oray.com/router/x1.html/parameter

Flash Layout:
0x000000000000-0x000000030000 : "u-boot"
0x000000030000-0x000000040000 : "kpanic"
0x000000040000-0x000000050000 : "factory"
0x000000050000-0x000000fe0000 : "firmware"
0x000000fe0000-0x000000ff0000 : "bdinfo"
0x000000ff0000-0x000001000000 : "reserve"

Install via SSH:
Original firmware is based on OpenWRT, but SSH is not start by default,
You should enable it first

1. Login into web admin (10.168.1.1), default password is 'admin'
2. Open the following link, and the result should be {"code":0};
SSH is now started, username is root, password is same as web admin password
http://10.168.1.1/cgi-bin/oraybox?_api=ssh_set&enabled=1
4. You can flash firmware via mtd: mtd write /tmp/firmware_image.bin firmware

Signed-off-by: Bin We <me@udp.pw>
19 months agouqmi: explicitly disconnect IPv6 address family
Lech Perczak [Tue, 15 Mar 2022 18:27:48 +0000 (19:27 +0100)]
uqmi: explicitly disconnect IPv6 address family

Some modems (namely, Telit LE910C4) require the IPv6 connection state to
be cleared explicitly, to avoid reporting "no effect" if IPv6
connection is already connected through autoconnect mechanism, or during
LTE default bearer attach, which would lead to established session, but
without a way to inform protocol handler of the status.

Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
19 months agouqmi: set IPv6 family explicitly in status check
Lech Perczak [Tue, 15 Mar 2022 18:29:59 +0000 (19:29 +0100)]
uqmi: set IPv6 family explicitly in status check

Some modems require CID to be set explicitly during IPv6 connection
status check, others require IPv6 address family to be checked explicitly
after establishing connection, in order to provide correct status.
Set both fields in the request to satisfy them.

Fixes: c8a88118af46 ("uqmi: set CID during 'query-data-status' operation")
Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
19 months agotools/ccache: bump to 4.8
Linhui Liu [Fri, 17 Mar 2023 14:35:16 +0000 (22:35 +0800)]
tools/ccache: bump to 4.8

Release Notes:
https://ccache.dev/releasenotes.html#_ccache_4_8

Signed-off-by: Linhui Liu <liulinhui36@gmail.com>
19 months agotools/ccache: always disable documentation compilation
Linhui Liu [Wed, 22 Mar 2023 05:20:23 +0000 (13:20 +0800)]
tools/ccache: always disable documentation compilation

Speed up compilation.

Also remove useless include target.mk
(b492e69bd2d09aaa2c3c76b52203cbb3527eb2f7).

Signed-off-by: Linhui Liu <liulinhui36@gmail.com>
19 months agolantiq: nand: don't yield while holding spinlock
Thomas Nixon [Sun, 26 Mar 2023 10:19:21 +0000 (10:19 +0000)]
lantiq: nand: don't yield while holding spinlock

The nand driver normally while waiting for the device to become ready;
this is normally fine, but xway_nand holds the ebu_lock spinlock, and
this can cause lockups if other threads which use ebu_lock are
interleaved. Fix this by waiting instead of polling.

This mainly showed up as crashes in ath9k_pci_owl_loader (see
https://github.com/openwrt/openwrt/issues/9829 ), but turning on
spinlock debugging shows this happening in other places too.

This doesn't seem to measurably impact boot time.

Tested on bt_homehub-v5a with 5.10 and 5.15.

Signed-off-by: Thomas Nixon <tom@tomn.co.uk>
[Add commit description into patch]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
19 months agotools/coreutils: update to 9.2
Georgi Valkov [Tue, 21 Mar 2023 09:39:21 +0000 (11:39 +0200)]
tools/coreutils: update to 9.2

This resolves an error when building toolchain/musl on macOS due to
improper hole-detection caused by a bug in macOS/APFS [1].

As long as we don't reconfigure, 001-m4.patch is not needed.
If we keep it, it will force reconfigure the project,
since m4 files are changed. This works, but may not be optimal,
because the build should use files from coreutils/m4, but
OpenWRT uses legacy files from staging_dir/host/share/aclocal [2].

backport a couple of upstream patches
date: diagnose -f read errors
copy: fix --reflink=auto to fallback in more cases

[1] https://debbugs.gnu.org/cgi/bugreport.cgi?bug=61386
[2] https://github.com/openwrt/openwrt/pull/12233#issuecomment-1481097456

Co-developed-by: Michael Pratt <mcpratt@pm.me>
Signed-off-by: Georgi Valkov <gvalkov@gmail.com>
19 months agoipq806x: Fix Linksys upgrade, restore config step
Jacob Aharon [Mon, 27 Mar 2023 01:12:48 +0000 (12:12 +1100)]
ipq806x: Fix Linksys upgrade, restore config step

It appears that the refactor of the upgrade process for NAND devices
resulted in the nand_do_upgrade_success step not being called for
devices using the linksys.sh script. As a result, configuration
was not preserved over sysupgrade steps.

This restores the preservation of configs for ipq806x devices using the
linksys.sh script. Other devices and targets have not been examined.

This commit uses the same functionality and terminology used in commit
8634c10 ("ipq40xx: Fix Linksys upgrade, restore config step")

Fixes: e25e6d8 ("base-files: fix and clean up nand sysupgrade code")
Tested-on: EA8500

Signed-off-by: Jacob Aharon <ah.jacob@gmail.com>
19 months agolibnl-tiny: update to the latest version
Hauke Mehrtens [Sun, 2 Apr 2023 00:24:15 +0000 (02:24 +0200)]
libnl-tiny: update to the latest version

f5d9b7e libnl-tiny: fix duplicated branch in family.h
11b7c5f attr: add NLA_S* definitions

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
19 months agonetifd: strip mask from IP address in DHCP client params
Andrey Erokhin [Tue, 7 Mar 2023 11:52:58 +0000 (16:52 +0500)]
netifd: strip mask from IP address in DHCP client params

ipaddr option can be in CIDR notation,
but udhcp wants just an IP address

Signed-off-by: Andrey Erokhin <a.erokhin@inango-systems.com>
19 months agokernel: backport upstreamed DSA B53 MMAP patch
Álvaro Fernández Rojas [Sat, 1 Apr 2023 18:58:54 +0000 (20:58 +0200)]
kernel: backport upstreamed DSA B53 MMAP patch

Move the patch from pending to backport, since it has been accepted upstream:
https://github.com/torvalds/linux/commit/45977e58ce65ed0459edc9a0466d9dfea09463f5

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
19 months agodnsmasq: configure dynamic dhcp6 and dhcp4 independently
Ian Dall [Fri, 10 Mar 2023 06:32:34 +0000 (17:02 +1030)]
dnsmasq: configure dynamic dhcp6 and dhcp4 independently

Given ipv6 has SLAAC it is quite plausible to wish to use dynamic
dhcp4 but static dhcp6. This patch keeps dynamicdhcp as the default
option for both, but is overridden by dynamicdhcpv6 or dynamicdhcpv4

Signed-off-by: Ian Dall <ian@beware.dropbear.id.au>
19 months agoramips: mt7621: add migration script for WLAN PCI paths
Felix Baumann [Sat, 18 Mar 2023 05:55:12 +0000 (06:55 +0100)]
ramips: mt7621: add migration script for WLAN PCI paths

PCI paths of the WLAN devices have changed between kernel 5.10 and 5.15;
migrate config so existing wifi-iface definitions don't break.

This is implemented as a hotplug handler rather than a uci-defaults script
as the migration script must run before the 10-wifi-detect hotplug handler.

based on b452af23a8602ebf7bfb0eb084383ecd595face5

migration was forgotten when device trees were adjusted in
688697889c9ceb9d2d8d413d1abcc5fa7edb402b
c77913be5be5f11a0f20dba1b45b96470099e8a8

fixes #9374

affected devices:
Netgear R6220
Netgear WAC104
Netgear WNDR3700 v5
Zbtlink ZBT-WE1326
Wiflyer WF3526-P
Arcadyan WE420223-99
Beeline Smartbox Flash (Arcadyan WG443223)
MTS WG430223 (Arcadyan WG430223)

Tested-by: Maximilian Baumgartner <aufhaxer@googlemail.com>
Tested-by: Mikhail Zhilkin <csharper2005@gmail.com>
Signed-off-by: Felix Baumann <felix.bau@gmx.de>
19 months agodnsmasq: add dhcphostsfile to ujail sandbox
Ruben Jenster [Thu, 23 Mar 2023 10:15:52 +0000 (11:15 +0100)]
dnsmasq: add dhcphostsfile to ujail sandbox

The dhcphostsfile must be mounted into the (ujail) sandbox.
The file can not be accessed without this mount.

Signed-off-by: Ruben Jenster <rjenster@gmail.com>
19 months agoramips: Add support for CJ-Hello HYC-G920
Muhammad AL-Qadhy [Tue, 21 Mar 2023 06:49:22 +0000 (08:49 +0200)]
ramips: Add support for CJ-Hello HYC-G920

Hanyang Digitech Co., Ltd.
MSIP-CMM-HYD-HYC-G920
CJ-Hello HYC-G920

SoC : MediaTek MT7621AT
RAM : 256M (SK hynix H5TQ2G63FFR)
FLASH : 16MB (Winbond W25Q128BV)
WiFi : MediaTek MT7602EN bgn 2SS
WiFi : MediaTek MT7612EN nac 2SS
BTN : Reset
LED : - Power RED
  - WAN Green
  - LAN {1-4}
  - WiFi 2.4 GHz Blue
  - WiFi 5 GHz Blue
  - USB Green

**For MT7621 stage1 DDR Test**

UART : J4 GND - 3V3 - TX - RX - GND / 57600-8N1

```
                MT7621   stage1 code 10:33:55 (ASIC)
                CPU=500000000 HZ BUS=166666666 HZ
```

**For u boot environment**

UART : J4 GND - 3V3 - TX - RX - GND / 115200-8N1

**UART Menu**

```
Please choose the operation:
1: Load system code to SDRAM via TFTP.
2: Load system code then write to Flash via TFTP.
3: Boot system code via Flash (default).
4: Entr boot command line interface.
7: Load Boot Loader code then write to Flash via Serial.
9: Load Boot Loader code then write to Flash via TFTP.
```

**Steps**

Press 4: Entr boot command line interface.

On the pormpt enter.
`setenv firmware_size 0xf60000`
Then enter.
`saveenv`
Then enter.
`reset`

**Device will reboot**

Set your IP 192.168.100.100/24
Connect your lan cable to wan port.

**On the UART Menu**

Press 2: Load system code then write to Flash via TFTP.

 Warning!! Erase Linux in Flash then burn new one. Are you sure?(Y/N) **enter** `Y`
 Please Input new ones /or Ctrl-C to discard
        Input device IP (192.168.100.55) ==:`192.168.100.55`
        Input server IP (192.168.100.100) ==:`192.168.100.100`
        Input Linux Kernel filename () ==:`openwrt-22.03.0-ramips-mt7621-hanyang_hyc-g920-squashfs-sysupgrade.bin`

After uploading firmware image, device will boot Openwrt.

Signed-off-by: Muhammad AL-Qadhy <m.ismael@gmail.com>
19 months agokernel: modules: tg3: limit to devices with pci support
Aleksander Jan Bajkowski [Fri, 2 Dec 2022 22:38:51 +0000 (23:38 +0100)]
kernel: modules: tg3: limit to devices with pci support

Kmod-tg3 supports Ethernet adapters over PCIe bus. On targets without
PCI support, this package is empty. Symbol CONFIG_TIGON3 depends on
CONFIG_PCI.

Signed-off-by: Aleksander Jan Bajkowski <olek2@wp.pl>
19 months agokernel: modules: hfcpci: limit to devices with pci support
Aleksander Jan Bajkowski [Fri, 2 Dec 2022 22:12:23 +0000 (23:12 +0100)]
kernel: modules: hfcpci: limit to devices with pci support

Kmod-hfcpci and kmod-hfcmulti supports ISDN adapters over PCI. On targets
without PCI support, this package is empty. Symbol CONFIG_MISDN_HFCMULTI
and CONFIG_MISDN_HFCPCI depends on CONFIG_PCI.

Signed-off-by: Aleksander Jan Bajkowski <olek2@wp.pl>
19 months agolibtraceevent: update to 1.7.2
Nick Hainke [Tue, 28 Mar 2023 22:29:47 +0000 (00:29 +0200)]
libtraceevent: update to 1.7.2

Changes:
1c6f0f3 libtraceevent: version 1.7.2
73f6a8a libtraceevent: Fix some missing commas in big endian blocks
da2ea6b libtraceevent: Rename "ok" to "token_has_paren" in process_sizeof()
e6f7cfa libtraceevent: No need for testing ok in else if (!ok) in process_sizeof()
a4b1ba5 libtraceevent: Fix double free in parsing sizeof()

Signed-off-by: Nick Hainke <vincent@systemli.org>
19 months agotools/squashfs4: bump to 4.6.1
Linhui Liu [Fri, 31 Mar 2023 13:42:09 +0000 (21:42 +0800)]
tools/squashfs4: bump to 4.6.1

Changelogs:
https://github.com/plougher/squashfs-tools/blob/4.6.1/CHANGES

Removed upstreamed patches:
- 001-Unsquashfs-Add-and-make-some-header-includes-conditi.patch
- 002-Mksquashfs-Make-sysinfo-conditional.patch
- 003-Only-use-available-CPUs.patch
- 004-action-rework-strdupa-with-POSIX-strdup-and-free.patch
- 005-Don-t-use-sigwaitinfo-sigtimedwait-if-not-supported.patch
- 006-Move-sysinfo.h-into-the-linux-only-section-should-fi.patch
- 007-Unsquashfs-fix-compilation-error-for-missing-sysctl..patch
- 100-portability.patch

Signed-off-by: Linhui Liu <liulinhui36@gmail.com>
19 months agotools/squashfs4: enable parallel builds
Linhui Liu [Fri, 31 Mar 2023 13:34:01 +0000 (21:34 +0800)]
tools/squashfs4: enable parallel builds

Run `time make tools/squashfs4/{clean,compile} -j$(nproc)`
Before:
real    0m8.803s
user    0m12.415s
sys     0m1.317s
After:
real    0m13.781s
user    0m13.290s
sys     0m1.528s

Signed-off-by: Linhui Liu <liulinhui36@gmail.com>
19 months agoramips: mt7621: remove set-affinity script
Rosen Penev [Tue, 28 Mar 2023 22:29:19 +0000 (15:29 -0700)]
ramips: mt7621: remove set-affinity script

From https://github.com/openwrt/openwrt/pull/12280#issuecomment-1489279860
On Ethernet and WLAN, NAPI is threaded for all queues. This means that the
processing work is not stuck on the CPU that fired the IRQ. Under heavy
load, IRQs get disabled anyway, so it should not matter at all which CPUs
the IRQs fire on.

Basic testing indicates this to be true. There's no speedup or slowdown.

Signed-off-by: Rosen Penev <rosenp@gmail.com>
19 months agoramips: mt7621: ASUS RT-AX53U add NMBM, nest firmware
Felix Baumann [Fri, 31 Mar 2023 04:16:04 +0000 (06:16 +0200)]
ramips: mt7621: ASUS RT-AX53U add NMBM, nest firmware

Nests kernel and ubi into firmware partition in-order to be compatible
with OEM firmware. This allows restoring oem firmware from a backup of
firmware2. Add jffs2 partition which is present in the oem firmware.
Add support for mediatek NMBM (wear leveling on newer mediatek devices).
Exclude UBI partition from NMBM management.
Continues PR #10685.

Tested-by: Felix Baumann <felix.bau@gmx.de>
Signed-off-by: Felix Baumann <felix.bau@gmx.de>
19 months agokernel: bump 5.15 to 5.15.105
John Audia [Thu, 30 Mar 2023 15:39:17 +0000 (11:39 -0400)]
kernel: bump 5.15 to 5.15.105

Manually rebased:

Removed upstreamed:
backport-5.15/743-v6.3-0005-net-dsa-b53-mmap-fix-device-tree-support.patch[1]
backport-5.15/743-v6.3-0006-net-dsa-tag_brcm-legacy-fix-daisy-chained-switches.patch[2]

All other patches automatically rebased.

1. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v5.15.105&id=9bbb3d3f218f1cf51f57a43cf8fd63b05be1b821
2. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v5.15.105&id=48f52431af9980582b6faa32ff8b581edb10486c

Build system: x86_64
Build-tested: bcm2711/RPi4B, ramips/tplink_archer-a6-v3, filogic/xiaomi_redmi-router-ax6000-ubootmod
Run-tested: bcm2711/RPi4B, ramips/tplink_archer-a6-v3, filogic/xiaomi_redmi-router-ax6000-ubootmod

Signed-off-by: John Audia <therealgraysky@proton.me>
19 months agokernel: add support for XMC XM25QH64C
Joe Mullally [Sun, 26 Feb 2023 22:01:22 +0000 (22:01 +0000)]
kernel: add support for XMC XM25QH64C

The XMC XM25QH64C is a 8MB SPI NOR chip. The patch is verified on TL-WPA8631P v3.
Datasheet available at https://www.xmcwh.com/uploads/442/XM25QH64C.pdf

Signed-off-by: Joe Mullally <jwmullally@gmail.com>
19 months agomac80211: ath11k: restore 160MHz support
Robert Marko [Tue, 28 Mar 2023 11:38:57 +0000 (13:38 +0200)]
mac80211: ath11k: restore 160MHz support

Recent ath11k sync introduced a regression causing 80+80 and 160MHz to
stop being advertised and thus not selectable due to the respective feature
flags being cleared.

So, until we get answers upstream to what was the reasoning behind this and
it gets fixed, lets just remove the flag clearing to reanable 160MHz.

Fixes: 789a0bac3535 ("mac80211: ath11k: sync with ath-next")
Signed-off-by: Robert Marko <robimarko@gmail.com>
19 months agobusybox: enable taskset by default
Felix Fietkau [Sat, 1 Apr 2023 07:07:54 +0000 (09:07 +0200)]
busybox: enable taskset by default

This is useful for controlling process affinity on SMP systems

Signed-off-by: Felix Fietkau <nbd@nbd.name>
19 months agokernel: fix mtk_eth_soc flow accounting for MT7986
Felix Fietkau [Sat, 1 Apr 2023 07:07:22 +0000 (09:07 +0200)]
kernel: fix mtk_eth_soc flow accounting for MT7986

This was accidentally dropped when MT7981 support was added

Signed-off-by: Felix Fietkau <nbd@nbd.name>
19 months agoarm-trusted-firmware-sunxi: bump to 2.8
Stijn Tintel [Tue, 20 Dec 2022 18:05:40 +0000 (20:05 +0200)]
arm-trusted-firmware-sunxi: bump to 2.8

Use latest release build instead of a git snapshot. As this tarball
extracts in a trusted-firmware-a-2.8 subdirectory, we no longer need to
override the PKG_NAME defined in trusted-firmware-a.mk. The actual
package name is still the same, so we don't need to update any
dependencies.

Tested on A64-OLinuXino-1Ge16GW.

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
19 months agouboot-sunxi: bump to 2020.07
Stijn Tintel [Wed, 21 Dec 2022 20:00:49 +0000 (22:00 +0200)]
uboot-sunxi: bump to 2020.07

This is the newest release where 210-sunxi-deactivate-binman.patch still
applies.

Tested on A64-Olinuxino-eMMC.

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
19 months agoipq40xx: add LED functions for Google WiFi
Jan-Niklas Burfeind [Mon, 27 Mar 2023 05:29:39 +0000 (07:29 +0200)]
ipq40xx: add LED functions for Google WiFi

Add LED function properties for the LED controller to avoid failing
driver probe with kernel 5.15.

While at it, also define the OpenWrt LED indicator patterns for this
device.

Ref commit 583ac0e11df7 ("mpc85xx: update lp5521 led-controller node for 5.10")

Google uses white for running and red for an issue

Signed-off-by: Jan-Niklas Burfeind <git@aiyionpri.me>
Tested-by: Andrijan Möcker <amo@ct.de>
Reviewed-by: Brian Norris <computersforpeace@gmail.com>
Signed-off-by: David Bauer <mail@david-bauer.net>
19 months agoipq40xx: add reset button for Google WiFi (Gale)
Jan-Niklas Burfeind [Sun, 26 Mar 2023 20:50:42 +0000 (22:50 +0200)]
ipq40xx: add reset button for Google WiFi (Gale)

Add the external reset button for use with OpenWrt.

Co-authored-by: Brian Norris <computersforpeace@gmail.com>
Signed-off-by: Jan-Niklas Burfeind <git@aiyionpri.me>
Reviewed-by: Brian Norris <computersforpeace@gmail.com>
Tested-by: Brian Norris <computersforpeace@gmail.com>
Signed-off-by: David Bauer <mail@david-bauer.net>
19 months agobmips: bcm6362/bcm63268: enable HW RNG
Álvaro Fernández Rojas [Fri, 31 Mar 2023 15:49:58 +0000 (17:49 +0200)]
bmips: bcm6362/bcm63268: enable HW RNG

This enables the HW Random Number Generator on the BCM6362 and BCM63268 SoCs,
which is the same one used on BCM6368 SoC.

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
19 months agokernel: remove obsolete netfilter tcp window size check bypass patch
Felix Fietkau [Thu, 30 Mar 2023 12:14:11 +0000 (14:14 +0200)]
kernel: remove obsolete netfilter tcp window size check bypass patch

On any currently supported hardware, the performance impact should not
matter anymore.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
19 months agomac80211, mt76: add fixes for recently discovered security issues
Felix Fietkau [Wed, 29 Mar 2023 15:54:19 +0000 (17:54 +0200)]
mac80211, mt76: add fixes for recently discovered security issues

Fixes CVE-2022-47522

Signed-off-by: Felix Fietkau <nbd@nbd.name>
19 months agoramips: Refresh patches
Hauke Mehrtens [Wed, 29 Mar 2023 17:07:05 +0000 (19:07 +0200)]
ramips: Refresh patches

Refresh the ramips kernel patches.

Fixes: 7472476aef26 ("kernel: fix mtk_eth_soc throughput regressions on gigabit PHY ports")
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
19 months agobmips: bcm6348-enet: register emac driver from iudma
Álvaro Fernández Rojas [Wed, 29 Mar 2023 16:19:11 +0000 (18:19 +0200)]
bmips: bcm6348-enet: register emac driver from iudma

Register the ethernet driver from iudma, which avoids the attempt to probe the
emac driver before iudma and its consequent deferral.
The ethernet driver can't work without iudma anyway.

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
19 months agokernel: filter out pahole version
Robert Marko [Tue, 7 Mar 2023 14:00:28 +0000 (15:00 +0100)]
kernel: filter out pahole version

Pahole version is being autodetected during runtime since kernel 5.15.96
via in-kernel scripts/pahole-version.sh so add CONFIG_PAHOLE_VERSION to
kernel filter in order to prevent it from being added to target configs.

Signed-off-by: Robert Marko <robimarko@gmail.com>
19 months agocomgt: ncm: support Mikrotik R11e-LTE6 modem
Szabolcs Hubai [Fri, 24 Mar 2023 23:00:27 +0000 (00:00 +0100)]
comgt: ncm: support Mikrotik R11e-LTE6 modem

The Mikrotik R11e-LTE6 modem is similar to ZTE MF286R modem, added
earlier: it has a Marvel chip, able to work in ACM+RNDIS mode, knows ZTE
specific commands, runs OpenWrt Barrier Breaker fork.
While the modem is able to offer IPv6 address, the RNDIS setup is unable
to complete if there is an IPv6 adress.

While it works in ACM+RNDIS mode, the user experience isn't as good as
with "proto 3g": the modem happily serves a local IP (192.168.1.xxx)
without internet access. Of course, if the modem has enough time
(for example at the second dialup), it will serve a public IP.

Modifing the DHCP Lease (to a short interval before connect and back to
default while finalizing) is a workaround to get a public IP at the
first try.

A safe workaround for this is to excercise an offline script of the
pingcheck program: simply restart (ifdown - ifup) the connection.

Another pitfall is that the modem writes a few messages at startup,
which confuses the manufacturer detection algorithm and got disabled.

    daemon.notice netifd: Interface 'mikrotik' is setting up now
    daemon.notice netifd: mikrotik (2366): Failed to parse message data
    daemon.notice netifd: mikrotik (2366): WARNING: Variable 'ok' does not exist or is not an array/object
    daemon.notice netifd: mikrotik (2366): Unsupported modem
    daemon.notice netifd: mikrotik (2426): Stopping network mikrotik
    daemon.notice netifd: mikrotik (2426): Failed to parse message data
    daemon.notice netifd: mikrotik (2426): WARNING: Variable '*simdetec:1,sim' does not exist or is not an array/object
    daemon.notice netifd: mikrotik (2426): Unsupported modem
    daemon.notice netifd: Interface 'mikrotik' is now down

A workaround for this is to use the "delay" option in the interface
configuration.

I want to thank Forum members dchard (in topic Adding support for
MikroTik hAP ac3 LTE6 kit (D53GR_5HacD2HnD)) [1]
and mrhaav (in topic OpenWrt X86_64 + Mikrotik R11e-LTE6) [2]
for sharing their experiments and works.
Another information page was found at eko.one.pl [3].

[1]: https://forum.openwrt.org/t/137555
[2]: https://forum.openwrt.org/t/151743
[3]: https://eko.one.pl/?p=modem-r11elte

Signed-off-by: Szabolcs Hubai <szab.hu@gmail.com>
19 months agocomgt: add quirk for Mikrotik modems based on Mikrotik R11e-LTE6
Szabolcs Hubai [Fri, 24 Mar 2023 21:24:26 +0000 (22:24 +0100)]
comgt: add quirk for Mikrotik modems based on Mikrotik R11e-LTE6

The MikroTik R11e-LTE6 modem goes into flight mode (CFUN=4) at startup
and the radio is off (*RADIOPOWER: 0):

    AT+RESET
    OK

    OK

    *SIMDETEC:2,NOS

    *SIMDETEC:1,SIM

    *ICCID: 8936500119010596302

    *EUICC: 1

    +MSTK: 11, D025....74F3

    *ADMINDATA: 0, 2, 0

    +CPIN: READY

    *EUICC: 1

    *ECCLIST: 5, 0, 112, 0, 000, 0, 08, 0, 118, 0, 911

    +CREG: 0

    $CREG: 0

    +CESQ: 99,99,255,255,255,255

    *CESQ: 99,99,255,255,255,255,0

    +CGREG: 0

    +CEREG: 0

    +CESQ: 99,99,255,255,255,255

    *CESQ: 99,99,255,255,255,255,0

    *RADIOPOWER: 0

    +MMSG: 0, 0

    +MMSG: 0, 0

    +MMSG: 1, 0

    +MPBK: 1

While the chat script is able to establish the PPP connection,
it's closed instantly by the modem: LCP terminated by peer.

    local2.info chat[7000]: send (ATD*99***1#^M)
    local2.info chat[7000]: expect (CONNECT)
    local2.info chat[7000]: ^M
    local2.info chat[7000]: ATD*99***1#^M^M
    local2.info chat[7000]: CONNECT
    local2.info chat[7000]:  -- got it
    local2.info chat[7000]: send ( ^M)
    daemon.info pppd[6997]: Serial connection established.
    kern.info kernel: [  453.659146] 3g-mikrotik: renamed from ppp0
    daemon.info pppd[6997]: Renamed interface ppp0 to 3g-mikrotik
    daemon.info pppd[6997]: Using interface 3g-mikrotik
    daemon.notice pppd[6997]: Connect: 3g-mikrotik <--> /dev/ttyACM0
    daemon.info pppd[6997]: LCP terminated by peer
    daemon.notice pppd[6997]: Connection terminated.
    daemon.notice pppd[6997]: Modem hangup
    daemon.info pppd[6997]: Exit.
    daemon.notice netifd: Interface 'mikrotik' is now down

Sending "AT+CFUN=1" to modem deactivates the flight mode and
solves the issue:

    daemon.notice netifd: Interface 'mikrotik' is setting up now
    daemon.notice netifd: mikrotik (7051): sending -> AT+CFUN=1
    daemon.notice pppd[7137]: pppd 2.4.9 started by root, uid 0
    local2.info chat[7140]: abort on (BUSY)
    local2.info chat[7140]: abort on (NO CARRIER)
    local2.info chat[7140]: abort on (ERROR)
    local2.info chat[7140]: report (CONNECT)
    local2.info chat[7140]: timeout set to 10 seconds
    local2.info chat[7140]: send (AT&F^M)
    local2.info chat[7140]: expect (OK)
    local2.info chat[7140]: ^M
    local2.info chat[7140]: +CESQ: 99,99,255,255,255,255^M
    local2.info chat[7140]: ^M
    local2.info chat[7140]: *CESQ: 99,99,255,255,255,255,0^M
    local2.info chat[7140]: AT&F^MAT&F^M^M
    local2.info chat[7140]: OK
    local2.info chat[7140]:  -- got it
    ...
    local2.info chat[7140]: send (ATD*99***1#^M)
    local2.info chat[7140]: expect (CONNECT)
    local2.info chat[7140]: ^M
    local2.info chat[7140]: ATD*99***1#^M^M
    local2.info chat[7140]: CONNECT
    local2.info chat[7140]:  -- got it
    local2.info chat[7140]: send ( ^M)
    daemon.info pppd[7137]: Serial connection established.
    kern.info kernel: [  463.094254] 3g-mikrotik: renamed from ppp0
    daemon.info pppd[7137]: Renamed interface ppp0 to 3g-mikrotik
    daemon.info pppd[7137]: Using interface 3g-mikrotik
    daemon.notice pppd[7137]: Connect: 3g-mikrotik <--> /dev/ttyACM0
    daemon.warn pppd[7137]: Could not determine remote IP address: defaulting to 10.64.64.64
    daemon.notice pppd[7137]: local  IP address 100.112.63.62
    daemon.notice pppd[7137]: remote IP address 10.64.64.64
    daemon.notice pppd[7137]: primary   DNS address 185.29.83.64
    daemon.notice pppd[7137]: secondary DNS address 185.62.131.64
    daemon.notice netifd: Network device '3g-mikrotik' link is up
    daemon.notice netifd: Interface 'mikrotik' is now up

To send this AT command to the modem the "runcommand.gcom" script
dependency is moved from comgt-ncm to comgt.
As the comgt-ncm package depends on comgt already, this change
is a NOOP from that point of view.
But from the modem's point it is a low hanging fruit as the modem
is usable with installing comgt and kmod-usb-ncm packages.

Signed-off-by: Szabolcs Hubai <szab.hu@gmail.com>
19 months agomediatek: re-add wireless driver for TUF-AX4200
Andrew Sim [Wed, 29 Mar 2023 07:48:30 +0000 (09:48 +0200)]
mediatek: re-add wireless driver for TUF-AX4200

The wireless driver package was incorrectly removed from the TUF-AX4200
device-packages, resulting in images without wireless functionality.

Fixes: d98c4fb8bfab ("mediatek: broaden filogic target description")
Signed-off-by: Andrew Sim <andrewsimz@gmail.com>
[rework commit description]
Signed-off-by: David Bauer <mail@david-bauer.net>
19 months agobmips: bcm6348-enet: add PHY support
Álvaro Fernández Rojas [Wed, 29 Mar 2023 14:17:04 +0000 (16:17 +0200)]
bmips: bcm6348-enet: add PHY support

We should ensure that the PHY is properly configured.
This is specially needed in devices using the internal PHY for ethernet0.

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
19 months agokernel: fix mtk_eth_soc throughput regressions on gigabit PHY ports
Felix Fietkau [Wed, 29 Mar 2023 09:38:30 +0000 (11:38 +0200)]
kernel: fix mtk_eth_soc throughput regressions on gigabit PHY ports

Fixes issues on MT7621 and MT7623 switch ports, and PHY ports on all SoC

Signed-off-by: Felix Fietkau <nbd@nbd.name>
19 months agobmips: add basic BCM6358 ethernet support
Álvaro Fernández Rojas [Tue, 24 May 2022 09:14:21 +0000 (11:14 +0200)]
bmips: add basic BCM6358 ethernet support

The MDIO bus is supported but there are errors when trying to probe and
configure the external BCM5325E switch through B53 DSA.
Therefore, let's add basic ethernet (but working) support for now.

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
19 months agobmips: bcm6368-enetsw: refactor probe
Álvaro Fernández Rojas [Tue, 28 Mar 2023 17:36:59 +0000 (19:36 +0200)]
bmips: bcm6368-enetsw: refactor probe

- Switch to devm functions where possible.
- Fix probing when no resets provided.
- Properly check pointers (IS_ERR_OR_NULL/PTR_ERR).
- Add probe info.

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
19 months agobmips: bcm6368-enetsw: remove unneeded variables
Álvaro Fernández Rojas [Tue, 28 Mar 2023 16:47:17 +0000 (18:47 +0200)]
bmips: bcm6368-enetsw: remove unneeded variables

Change the code to use defines instead of variables.

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
19 months agobmips: bcm6368-enetsw: harmonize dev variables
Álvaro Fernández Rojas [Tue, 28 Mar 2023 16:34:46 +0000 (18:34 +0200)]
bmips: bcm6368-enetsw: harmonize dev variables

The current implementation of bcm6368-enetsw is a mess of dev, ndev and kdev
variables, which have refer to different things depending on the function.
This commit harmonizes it and resolves the issue.

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
19 months agobmips: sercomm,h500-s: add Quantenna gpio switch
Álvaro Fernández Rojas [Sun, 26 Mar 2023 12:38:41 +0000 (14:38 +0200)]
bmips: sercomm,h500-s: add Quantenna gpio switch

Sercomm H500-s has a Quantenna SoC for external wifi which can be activated or
deactivated through GPIO #20.

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
19 months agoncm: add error check and retry mechanism for gcom call
Mike Wilson [Mon, 20 Jul 2020 20:25:33 +0000 (22:25 +0200)]
ncm: add error check and retry mechanism for gcom call

This patch solves the problem of receiving "error" responses when
initially calling gcom. This avoids unnecessary NO_DEVICE failures.

A retry loop retries the call after an "error" response within the
specified delay. A successful response will continue with the connection
immediately without waiting for max specified delay, bringing the
interface up sooner.

Signed-off-by: Mike Wilson <mikewse@hotmail.com>
19 months agokernel: remove unnecessary change from backlog threading fix
Felix Fietkau [Tue, 28 Mar 2023 11:13:53 +0000 (13:13 +0200)]
kernel: remove unnecessary change from backlog threading fix

Fixes: f1c9bd6f6f41 ("kernel: fix kernel warning and network issue in the backlog threading patch")
Signed-off-by: Felix Fietkau <nbd@nbd.name>
19 months agokernel: fix kernel warning and network issue in the backlog threading patch
Felix Fietkau [Tue, 28 Mar 2023 10:46:11 +0000 (12:46 +0200)]
kernel: fix kernel warning and network issue in the backlog threading patch

Fixes enabling it early during the boot

Signed-off-by: Felix Fietkau <nbd@nbd.name>
19 months agomediatek: broaden filogic target description
Daniel Golle [Fri, 10 Mar 2023 20:05:18 +0000 (20:05 +0000)]
mediatek: broaden filogic target description

The filogic subtarget now also supports MT7981 and will in future
also support MT7988. Reflect that in the target description.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
19 months agomediatek: add device tree include for the MT7981 SoC
Daniel Golle [Sat, 11 Mar 2023 03:55:52 +0000 (03:55 +0000)]
mediatek: add device tree include for the MT7981 SoC

Add mt7981.dtsi so we can start adding MT7981 devices.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
19 months agogeneric: fix Macronix SPI-NAND driver
Daniel Golle [Sat, 11 Mar 2023 02:38:59 +0000 (02:38 +0000)]
generic: fix Macronix SPI-NAND driver

Use scratch buffer for DMA operetations. Passing a pointer to a stack
variable won't work and results in bogus bit flips being reported.

Patch was submitted upstream and is part of Linux 6.3.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
19 months agomediatek: add support for SPI calibration
Daniel Golle [Fri, 10 Mar 2023 23:53:54 +0000 (23:53 +0000)]
mediatek: add support for SPI calibration

Newer MediaTek's SoCs need SPI calibration routines for SPI to work
reliably. Import patches for that from MediaTek's SDK.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
19 months agomediatek: add PWM support for MT7981 SoC
Daniel Golle [Fri, 10 Mar 2023 19:57:18 +0000 (19:57 +0000)]
mediatek: add PWM support for MT7981 SoC

Add patch to support PWM on the MT7981 SoC.
This patch will also be submitted to upstream Linux soon.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
19 months agomediatek: add I2C for MT7981 SoC
Daniel Golle [Fri, 10 Mar 2023 19:55:09 +0000 (19:55 +0000)]
mediatek: add I2C for MT7981 SoC

Add patch to support I2C on the MT7981 SoC.
This change will also be submitted to upstream Linux soon.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
19 months agomediatek: simplify MaxLinear GPY PHY driver
Daniel Golle [Sat, 11 Mar 2023 03:51:00 +0000 (03:51 +0000)]
mediatek: simplify MaxLinear GPY PHY driver

The mxl-gpy driver apparently was built in the assumption that SGMII
auto-negotiation is always switched on at the MAC. This may be true for
few rather recent drivers (why?), but certainly isn't for most drivers
unless 'managed = "in-band-status"' is set in device tree. Add patch to
the mediatek target which reduces mxl-gpy to behave more like an
ordinary PHY driver using out-of-band status.

This allows to use these PHYs without rate-adaptation which seems to be
at least partially broken/racy in some revisions of the PHY and/or
internal PHY firmware.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
19 months agogeneric: dsa: make use of pcs-mtk-lynxi also for mt7530
Daniel Golle [Sat, 11 Mar 2023 17:58:12 +0000 (17:58 +0000)]
generic: dsa: make use of pcs-mtk-lynxi also for mt7530

The MT7531 switch IC comes with SerDes ports with PCS identical to
what is also used in MediaTek's SoCs. Make use of the shared driver
to ease maintainance and reduce code duplication.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
19 months agogeneric: backport regmap reg_base and downshift features
Daniel Golle [Sat, 11 Mar 2023 17:55:56 +0000 (17:55 +0000)]
generic: backport regmap reg_base and downshift features

Backport new regmap features from Linux 5.18 which are needed to let
mt7530 use pcs-mtk-lynxi.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
19 months agogeneric: mtk_eth_soc: switch to external PCS driver
Daniel Golle [Mon, 27 Mar 2023 17:15:13 +0000 (18:15 +0100)]
generic: mtk_eth_soc: switch to external PCS driver

Backport patch to make use of the new PCS driver in mtk_eth_soc.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
19 months agogeneric: pcs-mtk-lynxi: add hack to use 2500Base-X without AN
Daniel Golle [Mon, 27 Mar 2023 17:41:54 +0000 (18:41 +0100)]
generic: pcs-mtk-lynxi: add hack to use 2500Base-X without AN

Using 2500Base-T SFP modules e.g. on the BananaPi R3 requires manually
disabling auto-negotiation, e.g. using ethtool. While a proper fix
using SFP quirks is being discussed upstream, bring a work-around to
restore user experience to what it was before the switch to the
dedicated SGMII PCS driver.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
19 months agogeneric: net: pcs: add driver for MediaTek SGMII PCS
Daniel Golle [Mon, 27 Mar 2023 17:14:21 +0000 (18:14 +0100)]
generic: net: pcs: add driver for MediaTek SGMII PCS

Backport dedicated PCS driver for MediaTek LynxI SGMII/SerDes unit.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
19 months agogeneric: mtk_eth_soc: allow setting MDIO clock frequency
Daniel Golle [Mon, 27 Mar 2023 16:50:54 +0000 (17:50 +0100)]
generic: mtk_eth_soc: allow setting MDIO clock frequency

Backport patch allowing to set the MDIO bus clock frequency.
By default the MDIO bus clock runs on 2.5 MHz, allow increasing it
up to 25 MHz.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
19 months agomediatek: add support for built-in GE PHYs of newer SoCs
Daniel Golle [Fri, 10 Mar 2023 19:44:52 +0000 (19:44 +0000)]
mediatek: add support for built-in GE PHYs of newer SoCs

MT7981 and the upcoming MT7988 have built-in Gigabit Ethernet PHYs.
While they share some design properties with the PHYs present in
MT753x, they do need calibration data from the SoC's efuse.

Add driver to support them. Upstreaming it is planned, but there are
still some ongoing discussions with MediaTek.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
19 months agogeneric: backport support for MT7981 Ethernet
Daniel Golle [Fri, 10 Mar 2023 19:41:51 +0000 (19:41 +0000)]
generic: backport support for MT7981 Ethernet

Backport support for the MT7981 SoC to mtk_eth_soc.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
19 months agomediatek: backport pinctrl driver for MT7981 SoC
Daniel Golle [Fri, 10 Mar 2023 19:40:37 +0000 (19:40 +0000)]
mediatek: backport pinctrl driver for MT7981 SoC

Backport the pinctrl driver for the MT7981 SoC. The driver has also
been submitted upstream and is part of Linux 6.3.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
19 months agomediatek: backport clk driver for MT7981 SoC
Daniel Golle [Fri, 10 Mar 2023 19:36:12 +0000 (19:36 +0000)]
mediatek: backport clk driver for MT7981 SoC

Backport driver for common clocks in MT7981 SoC. The driver has also
been submitted upstream and became part of Linux 6.3.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
19 months agogeneric: replace mtk_sgmii patches with versions accepted upstream
Daniel Golle [Mon, 27 Mar 2023 15:02:44 +0000 (16:02 +0100)]
generic: replace mtk_sgmii patches with versions accepted upstream

Replace patches for MediaTek Ethernet driver SGMII/SerDes unit with
their corresponding upstream patches. Not all of the patches in our
tree went upstream as-is, some are slightly different implementations,
and they require the phylink_pcs helpers now made available.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
19 months agogeneric: backport some phylink helper functions
Daniel Golle [Sat, 11 Mar 2023 03:44:41 +0000 (03:44 +0000)]
generic: backport some phylink helper functions

It isn't feasible to literally backport all upstream phylink_pcs changes
down to Linux 5.15: It's just too many patches, and many downstream
drivers and hacks are likely to break. We are too close to branching off
to risk this, and it's also just too much work.
Instead just add helper functions used by modern PCS drivers while keeping
the original functions instact as well. While this may add a kilobyte or
two of extra kernel size, it has the advantage that we get the best of both
worlds: None of the existing codepaths are touched, but yet we have the
option to backport singular improvements to Ethernet drivers where needed.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
19 months agoipq-wifi: bump to latest git HEAD
Christian Marangi [Mon, 27 Mar 2023 16:15:13 +0000 (18:15 +0200)]
ipq-wifi: bump to latest git HEAD

ccd7e46 ipq40xx: add support for Wallystech DR40x9
2ce60e1 Revert "ipq40xx: add support for Wallystech DR40x9"
ea962ca ipq40xx: add Emplus WAP551 BDF

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
19 months agogeneric: 5.15: refresh kernel patches
Christian Marangi [Sat, 25 Mar 2023 16:24:27 +0000 (17:24 +0100)]
generic: 5.15: refresh kernel patches

Refresh kernel patches for generic kernel 5.15 due to new backport
version of MGLRU patchset.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>