openwrt/staging/xback.git
46 hours agoRevert "devel/gdb: Update to 14.1"
Koen Vandeputte [Tue, 15 Oct 2024 16:52:05 +0000 (18:52 +0200)]
Revert "devel/gdb: Update to 14.1"

This reverts commit 4a5d1d0e95d6ec6ac9c9418af9d93a0ad2e7c100.

46 hours agoRevert "toolchain: gdb: Update to version 15.2"
Koen Vandeputte [Tue, 15 Oct 2024 08:17:29 +0000 (10:17 +0200)]
Revert "toolchain: gdb: Update to version 15.2"

This reverts commit 1cd7708eb3f2569ff0c005c568ed9bbd2391a06b.

46 hours agoRevert "gdb: Update to version 15.2"
Koen Vandeputte [Tue, 15 Oct 2024 08:17:21 +0000 (10:17 +0200)]
Revert "gdb: Update to version 15.2"

This reverts commit 654983ea1166f53423896688593f6d6a966f1d15.

46 hours agoRevert "gdb: Remove upstreamed SIGINT patch"
Koen Vandeputte [Tue, 15 Oct 2024 08:17:13 +0000 (10:17 +0200)]
Revert "gdb: Remove upstreamed SIGINT patch"

This reverts commit c71e53bf9df5714e7a6efd70141c9337edb6c122.

46 hours agoRevert "gdb: Remove MIPS NSIG patch"
Koen Vandeputte [Tue, 15 Oct 2024 08:17:02 +0000 (10:17 +0200)]
Revert "gdb: Remove MIPS NSIG patch"

This reverts commit 2758e1b21c35efbc9dbf40348afb4a73d88af207.

46 hours agoRevert "gdb: Fix description for apk, remove backtick"
Koen Vandeputte [Mon, 25 Nov 2024 17:35:39 +0000 (18:35 +0100)]
Revert "gdb: Fix description for apk, remove backtick"

This reverts commit adb921c34ff220951a699665bb35dc0e3538b3b2.

46 hours agoath10k-ct: add Coverage Class setting ipq40xx and qca99xx WiFi (WIP)
Koen Vandeputte [Tue, 21 May 2024 12:49:14 +0000 (14:49 +0200)]
ath10k-ct: add Coverage Class setting ipq40xx and qca99xx WiFi (WIP)

Since QCA wireless firmware version 10.4 the Coverage Class can be set
for ipq40xx and qca99xx circuits via the wmi interface.
A timeout value adjusts protocol timing for the extra radio propagation
time that is inherent to transmission over larger than local distances.

In due time an upstream patch in mac80211 can be anticipated.
This patch enables use of ipq40xx SoC based devices in OpenWrt in the
mean time when Coverage Class setting is required.

Authors:
Sebastian Gottschall    <s.gottschall@dd-wrt.com>
William Wortel          <wwortel@dorpstraat.com>

Signed-off-by: William Wortel <wwortel@dorpstraat.com>
Signed-off-by: Koen Vandeputte <koen.vandeputte@citymesh.com>
46 hours agohostapd: sync config files with upstream
Koen Vandeputte [Tue, 24 Oct 2023 12:36:15 +0000 (14:36 +0200)]
hostapd: sync config files with upstream

Sync all config files, without changing current behaviour

This removes some upstream deprecated options while
introducing new ones.

Signed-off-by: Koen Vandeputte <koen.vandeputte@citymesh.com>
46 hours agokernel: re-add missing symbols
Koen Vandeputte [Tue, 11 Jun 2024 13:27:58 +0000 (15:27 +0200)]
kernel: re-add missing symbols

a recent cleanup commit also removed our own custom symbols
added by various in-tree drivers.

re-add these symbols as the compiler will complain about
them otherwise on specific targets during kernel build:

MikroTik RB91x NAND driver support (MTD_NAND_RB91X) [N/m/y/?] (NEW) make[8]: *** [scripts/kconfig/Makefile:77: syncconfig] Error 1

Fixes: 2d00ed7caa ("kernel: trim outdated kernel config symbols")
Signed-off-by: Koen Vandeputte <koen.vandeputte@citymesh.com>
46 hours agoath79: set rb912 back to 25.000MHz clock
Koen Vandeputte [Wed, 19 Feb 2025 12:52:02 +0000 (13:52 +0100)]
ath79: set rb912 back to 25.000MHz clock

My boards all contain a 25MHz xtal ..

Signed-off-by: Koen Vandeputte <koen.vandeputte@citymesh.com>
46 hours agoath79: rb91x: enable pcie power early (WIP)
Koen Vandeputte [Mon, 8 Jan 2024 16:25:07 +0000 (17:25 +0100)]
ath79: rb91x: enable pcie power early (WIP)

ensure PCIe devices get enumerated properly during power-on

Signed-off-by: Koen Vandeputte <koen.vandeputte@citymesh.com>
2 days agoramips: add support for Hongdian H8922 v30
Coia Prant [Fri, 3 Jan 2025 18:43:17 +0000 (02:43 +0800)]
ramips: add support for Hongdian H8922 v30

This is an industrial 4G router equipped with OpenWrt 14.07 OEM
customized version

WARNING: The original firmware device tree is common to multiple
boards, and the device tree name is H9350. This submitted device
tree is a modified version, which deletes the non-this-device parts
and adds GPIO watchdog.

Specification:
- SoC: MediaTek MT7620A
- Flash: 16 MB
- RAM: 128 MB
- Power: DC 5V-36V 1.5A
- Ethernet: 1x WAN, 4x LAN (10/100 Mbps)
- Wireless radio: 802.11n 2.4g-only
- LED:
  System/Power (RUN): GPIO/26 active-low
  Ethernet: 1x WAN, 4x LAN
  Modem 1: GPIO/66 active-low
  RF 1 (Modem 1 Signal): GPIO/67 active-low
  Modem 2: GPIO 71 active-low
  RF 2 (Modem 2 Signal): GPIO/24 active-low
  WLAN: GPIO/72 active-low
  WPS: GPIO/12 active-low
- Button:
  WPS / RESET: GPIO/34 active-low
- UART: 1x UART on PCB - 115200 8N1
- GPIO Watchdog: GPIO/62 mode=toggle timeout=1s
- PCIe: 2x miniPCIe for modem
- SIM Slots: 2x SIM Slots

Issue:
- No factory partition, eeprom is located
 at /lib/firmware/mt7620a.eeprom

Flash instruction:
Using UART:
1. Configure PC with a static IP address and setup an TFTP server.
2. Put rootfs into the tftp directory.
3. Connect the UART line as described on the PCB.
4. Power up the device and press Ctrl+C to break auto boot.
5. Use `system 6` command and follow the instruction to set device
   and tftp server IP address and input the rootfs file name.
   U-boot will then load the rootfs and write it into
   the flash.
6. Use `system 1` command and follow the instruction to set device
   and tftp server IP address and input the firmware file name.
   U-boot will then load the firmware once.
7. Login to LuCI and use LuCI upgrade firmware.

Original Firmware Dump / More details:
https://blog.gov.cooking/archives/research-hongdian-h8922-and-flash.html

Signed-off-by: Coia Prant <coiaprant@gmail.com>
Tested-by: Coia Prant <coiaprant@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/17472
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 days agomac80211: rt2x00: load the eeprom data from devicetree embedded data on Ralink SoCs
Coia Prant [Mon, 17 Feb 2025 06:22:59 +0000 (14:22 +0800)]
mac80211: rt2x00: load the eeprom data from devicetree embedded data on Ralink SoCs

It will allow loading eeprom from eeprom-data embedded in device tree.
Ported from mediatek mt76 wireless driver (drivers/net/wireless/mediatek/mt76/eeprom.c)

Signed-off-by: Coia Prant <coiaprant@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/17472
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 days agoath79: add support for Fortinet FAP-221-C
Bernardus Jansen [Sat, 8 Mar 2025 13:57:42 +0000 (14:57 +0100)]
ath79: add support for Fortinet FAP-221-C

FCC ID: TVE-121402

Fortinet FAP-221-C is an indoor access point with 1gb ethernet port,
dual-band wireless, internal antenna plates, and 802.3at PoE+.

Hardware and board design are from Senao. The device appears very
similar to the EnGenius EAP1200H, albeit with double the flash and RAM.

**Specifications:**

  - QCA9557 SOC
  - QCA9882 WLAN PCI card, 5 GHz, 2x2, 26dBm
  - AR8035-A PHY RGMII GbE with PoE+ IN
  - 40 MHz clock
  - 32 MB FLASH FL256SAIFR0
  - 2x 128 MB RAM NT5TU64M16HG
  - UART populated
  - 4 internal antenna plates
  - 5 LEDs, 1 button (power, 'warning', eth0, wifi1, wifi2) (reset)

  Amber LAN LED appears hardwired to ethernet port. Power LED is green
only. Other LEDs are amber/green.

**MAC addresses:**

1 MAC Address in flash at end of uboot
ASCII encoded, no delimiters
Labeled as "MAC Address" on case

**Serial Access:**

Pinout: (arrow) VCC GND RX TX

Pins are populated with a header and traces not blocked.
Bootloader is set to 9600 baud, 8 data, 1 stop.

**Console Access:**

Bootloader:

Interrupt boot with Ctrl+C
Press "k" and enter password "1"
OR
Hold reset button for 5 sec during power on
Interrupt the TFTP transfer with Ctrl+C

to print commands available, enter "help"

OEM:

default username is "admin", password blank
telnet is available at default address 192.168.1.2
serial is available with baud 9600

to print commands available, enter "help"
or tab-tab (busybox list of commands)

**Installation:**

Use factory.bin with OEM upgrade procedures
OR
Use initramfs.bin with uboot TFTP commands.
Then perform a sysupgrade with sysupgrade.bin

**TFTP Recovery:**

Using serial console, load initramfs.bin using TFTP
to boot openwrt without touching the flash.

**Return to OEM:**

The best way to return to OEM firmware
is to have a copy of the MTD partitions
before flashing Openwrt.

Backup copies should be made of partitions
"fwconcat0", "loader", and "fwconcat1"
which together is the same flash range
as OEM's "rootfs" and "uimage"
by loading an initramfs.bin
and using LuCI to download the mtdblocks.

It is also possible to extract from the
OEM firmware upgrade image by splitting it up
in parts of lengths that correspond
to the partitions in openwrt
and write them to flash,
after gzip decompression.

After writing to the firmware partitions,
erase the "reserved" partition and reboot.

Signed-off-by: Bernardus Jansen <bernardus@bajansen.nl>
Link: https://github.com/openwrt/openwrt/pull/18109
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 days agodnsmasq: fix handlers for options `filter_rr` and `cache_rr`
Eric ZHANG [Sun, 2 Mar 2025 07:54:37 +0000 (15:54 +0800)]
dnsmasq: fix handlers for options `filter_rr` and `cache_rr`

According to:
- https://github.com/openwrt/luci/blob/master/modules/luci-mod-network/htdocs/luci-static/resources/view/network/dhcp.js#L700
- https://github.com/openwrt/luci/blob/master/modules/luci-mod-network/htdocs/luci-static/resources/view/network/dhcp.js#L402

These two options should be of type `MultiValue` but here there're used as single value. This results in dnsmasq crashes when either of these options are set with multiple values, which leads to an invalid space-separated value.

As these options are designed to take multiple values, I think it's better to use list format eg. `list filter_rr 'AAAA'`, instead of `option filter_rr 'AAAA,HTTPS'`.

See: https://forum.openwrt.org/t/selecting-more-than-one-filter-arbitrary-rr-or-cache-arbitrary-rr-in-v24-10-0-crashes-dnsmasq/225801

Tested on 24.10.0

Signed-off-by: Eric ZHANG <ericzhangjx@outlook.com>
Link: https://github.com/openwrt/openwrt/pull/18149
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 days agokernel: Add KERNEL_DCB (Data Center Bridging)
Hauke Mehrtens [Sun, 19 Jan 2025 18:49:38 +0000 (19:49 +0100)]
kernel: Add KERNEL_DCB (Data Center Bridging)

The kmod-mlxsw-spectrum driver activated CONFIG_DCB indirectly already
on all targets which are building this driver. All other DCB capable
driver did not activate their DCB support.

CONFIG_DCB increases the uncompressed kernel size by about 7.8KB.

CONFIG_DCB is only needed some data center Ethernet cards and not used
on normal routers. Activate it only on the x86_64 and the armsr_arm64
target which are used on normal servers or in VMs.

Link: https://github.com/openwrt/openwrt/pull/17672
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 days agoramips: Add support for Cudy M1300 v2
Roland Reinl [Fri, 28 Feb 2025 05:11:01 +0000 (06:11 +0100)]
ramips: Add support for Cudy M1300 v2

The M1300 v2 is similar to the WR1300 series from Cudy. Differences:
- Only 1 LAN port
- No USB

Specifications:
- MT7621
- MT7603E (2.4G b/g/n) and MT7613BE (5G ac/n) wifi
- 128 MB RAM
- 16 MB flash

MAC Addresses:
- There is one on the label, e.g. xx:xx:xx:xx:xx:1C
- LAN (bottom connector) is the same as the label, e.g. xx:xx:xx:xx:xx:1C
- WAN (top connector) is label +2, e.g. xx:xx:xx:xx:xx:1E
- WLAN (2.4G) is the same as the label, e.g. xx:xx:xx:xx:xx:1C
- WLAN (5G) is the same as WAN, e.g. xx:xx:xx:xx:xx:1E

UART:
- is available via the pin holes on the board
- From inner to outer pin: TX, RX, GND, VCC
- Do NOT connect VCC
- Settings: 3.3V, 115200, 8N1

GPIO:
- There are two LEDs: Red (GPIO 3) and White (GPIO 4)
- There are two buttons: Reset (GPIO 8) and WPS (GPIO 10)

Migration to OpenWrt:
- Download the migration image from the Cudy website (it should be available as soon as OpenWrt officially supports the device)
- The migration image is also available here until a image is provided by Cudy: https://github.com/RolandoMagico/openwrt-build/releases/tag/M1300_Build_20240222
- File: openwrt-ramips-mt7621-cudy_m1300-v2-squashfs-flash-signed.bin
- Connect computer to LAN (bottom connector) and flash the migration image via OEM web interface
- In the migration image, LAN and WAN are swapped. Computer must be connected to the other port after flashing
- OpenWrt is now accessible via 192.168.1.1
- After flashing an up to date OpenWrt image, LAN and WAN settings are again the same as in the OEM firmware
- So use the other connector again

Revert back to OEM firmware:
- Set up a TFTP server on IP 192.168.1.88 and connect to the LAN port (lower port)
- Provide the Cudy firmware as recovery.bin in the TFTP server
- Press the reset button while powering on the device
- Recovery process is started now
- When recovery process is done, OEM firmware is accessible via 192.168.10.1 again

General information:
- No possibility to load a initramfs image via U-Boot because there is no option to interrupt U-Boot

Signed-off-by: Roland Reinl <reinlroland+github@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/18139
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 days agounetd: update to Git HEAD (2025-03-09)
Felix Fietkau [Sun, 9 Mar 2025 15:43:58 +0000 (16:43 +0100)]
unetd: update to Git HEAD (2025-03-09)

d8b43985e4d7 ubus: fix token_create policy
7326459bd743 ubus: dump service information on network_get
6c9c8fbd8128 service: add @all as alias for all members, unless defined differently

Signed-off-by: Felix Fietkau <nbd@nbd.name>
2 days agocli: fix spurious error when exiting the cli
Felix Fietkau [Sun, 9 Mar 2025 09:51:46 +0000 (10:51 +0100)]
cli: fix spurious error when exiting the cli

Signed-off-by: Felix Fietkau <nbd@nbd.name>
2 days agocli: improve error handling on editor select call
Felix Fietkau [Sun, 9 Mar 2025 09:49:52 +0000 (10:49 +0100)]
cli: improve error handling on editor select call

Use common error functions instead of explicit warn() calls

Signed-off-by: Felix Fietkau <nbd@nbd.name>
3 days agolantiq: xrx200_legacy: migrate ECI V-2FUb/R to upstream PHY LED control
Aleksander Jan Bajkowski [Sun, 2 Mar 2025 20:13:47 +0000 (21:13 +0100)]
lantiq: xrx200_legacy: migrate ECI V-2FUb/R to upstream PHY LED control

This commit switches the control of the leds connected to the Lantiq PEF7071V
PHY to an upstream solution. There should be no functional changes.

Signed-off-by: Aleksander Jan Bajkowski <olek2@wp.pl>
Link: https://github.com/openwrt/openwrt/pull/18153
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
3 days agoramips: fix reading mac address for hiwifi hc5962
Tianling Shen [Sun, 2 Mar 2025 12:06:49 +0000 (20:06 +0800)]
ramips: fix reading mac address for hiwifi hc5962

The spaces in variables have been stripped since commit 551e04f3c9c0
("base-files: strip space and tab characters from ASCII mac address"),
resulting "Vfac_mac " matches nothing. Fix the issue by removing the
space at end.

Signed-off-by: Tianling Shen <cnsztl@immortalwrt.org>
Link: https://github.com/openwrt/openwrt/pull/18150
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
3 days agoramips: mt7621: 02_network optimization
Mieczyslaw Nalewaj [Sat, 1 Mar 2025 16:41:50 +0000 (17:41 +0100)]
ramips: mt7621: 02_network optimization

Merging repeating blocks inside the ramips_setup_interfaces() function.

Signed-off-by: Mieczyslaw Nalewaj <namiltd@yahoo.com>
Link: https://github.com/openwrt/openwrt/pull/18146
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
3 days agomediatek: add support for CMCC A10
Tianling Shen [Wed, 26 Feb 2025 13:52:53 +0000 (21:52 +0800)]
mediatek: add support for CMCC A10

This board is also as known as SuperElectron ZN-M5 and ZN-M8. However,
for ZN-M5 and ZN-M8, there's another version uses ZX279128 as CPU
chip, which is unsupported.
You can check it in "高级设置" > "系统日志" > "内核日志" page from webUI.

Hardware specification:
  SoC: MediaTek MT7981B 2x A53
  Flash: 128 MB SPI-NAND
  RAM: 256MB
  Ethernet: 4x 10/100/1000 Mbps
  Switch: MediaTek MT7531AE
  WiFi: MediaTek MT7976C
  Button: Reset, WPS
  Power: DC 12V 1A

Stock layout flash instructions:
Login into webUI and upload sysupgrade firmware in "系统管理" > "升级固件" page.
Remember to unselect "保留配置" ("Keep configurations") first before doing that.

OpenWrt U-Boot layout flash instructions:
1. Flash stock layout firmware first.
2. Connect to the device via SSH, and backup everything,
   especially 'Factory' partition.
3. Unlock MTD partitions:
   apk update && apk add kmod-mtd-rw
   insmod mtd-rw i_want_a_brick=1
4. Write new BL2 and FIP:
   mtd write openwrt-mediatek-filogic-cmcc_a10-ubootmod-preloader.bin BL2
   mtd write openwrt-mediatek-filogic-cmcc_a10-ubootmod-bl31-uboot.fip FIP
5. Set static IP on your PC:
   IP 192.168.1.254/24, GW 192.168.1.1
6. Serve OpenWrt initramfs image using TFTP server.
7. Cut off the power and re-engage, wait for TFTP recovery to complete.
8. After OpenWrt has booted, perform sysupgrade.

Signed-off-by: Tianling Shen <cnsztl@immortalwrt.org>
Link: https://github.com/openwrt/openwrt/pull/18121
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
3 days agogeneric: globally enable CONFIG_PCPU_DEV_REFCNT
Christian Marangi [Thu, 6 Mar 2025 11:19:40 +0000 (12:19 +0100)]
generic: globally enable CONFIG_PCPU_DEV_REFCNT

Quoting the kconfig description for CONFIG_PCPU_DEV_REFCNT:

network device refcount are using per cpu variables if this option is
set. This can be forced to N to detect underflows (with a performance
drop).

This was introduced from kernel 5.13 and was wrongly set as disabled.
Some target actually enables it but this should be always enabled unless
refcount needs to be debugged (unlikely for production images)

Enable in generic and drop the entry in every other target.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/18174
Signed-off-by: Robert Marko <robimarko@gmail.com>
4 days agoqualcommax: fap650: fix dtc warnings on partitions
Weikai Kong [Thu, 6 Mar 2025 23:58:40 +0000 (18:58 -0500)]
qualcommax: fap650: fix dtc warnings on partitions

This commit adds the missing properties to address the following warnings:
Warning (reg_format): /soc@0/spi@78b5000/flash@0/partitions/partition@x:reg: property has invalid length (8 bytes) (#address-cells == 2, #size-cells == 1)"

Signed-off-by: Weikai Kong <priv@pppig236.com>
Link: #18180
Signed-off-by: Robert Marko <robimarko@gmail.com>
4 days agomvebu/image/cortexa53.mk: add kmod-dsa-mv88e6xxx
Jiri Kastner [Fri, 7 Mar 2025 08:08:47 +0000 (09:08 +0100)]
mvebu/image/cortexa53.mk: add kmod-dsa-mv88e6xxx

all espressobins have topaz switch as same soc, so it is safe to add
kmod-dsa-mv88e6xxx to all

Signed-off-by: Jiri Kastner <cz172638@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/18161
Signed-off-by: Robert Marko <robimarko@gmail.com>
4 days agokernel: Mediatek: set default EEE Tx LPI timer
Qingfang Deng [Fri, 7 Mar 2025 12:33:06 +0000 (20:33 +0800)]
kernel: Mediatek: set default EEE Tx LPI timer

Due to API changes during the backport, the default value of Tx LPI
timer is accidentally left unset, breaking the network if EEE is on.
Set the default timer to 1ms on init, and fix an incorrect condition.

Fixes: d8315d5358d5 ("kernel: backport Mediatek SoC EEE support")
Signed-off-by: Qingfang Deng <dqfext@gmail.com>
5 days agomediatek: filogic: enable CONFIG_MITIGATE_SPECTRE_BRANCH_HISTORY
Martin Schiller [Thu, 6 Mar 2025 08:52:27 +0000 (09:52 +0100)]
mediatek: filogic: enable CONFIG_MITIGATE_SPECTRE_BRANCH_HISTORY

The Cortex-A73 in MediaTek Filogic 880 (MT7988A) is vulnerable to
multiple Spectre variants including Spectre-BHB.

Ref: https://developer.arm.com/Arm%20Security%20Center/Speculative%20Processor%20Vulnerability
Signed-off-by: Martin Schiller <ms@dev.tdt.de>
5 days agomediatek: Refresh kernel configuration
Martin Schiller [Thu, 6 Mar 2025 08:38:12 +0000 (09:38 +0100)]
mediatek: Refresh kernel configuration

I selected one subtarget after the other and refreshed their
configuration using this command:
  make kernel_oldconfig CONFIG_TARGET=subtarget

For MT7629 I had to re-add CONFIG_LEDS_SMARTRG_LED manually.
Otherwise, building MT7629 with ALL_KMODS we get prompted for
LEDS_SMARTRG_LED and this will break CI and in future buildbot
compilation. See commit 6bdea8c7bd85 ("mediatek: mt7629: 6.6: disable
LEDS_SMARTRG_LED by default") for more details.

Signed-off-by: Martin Schiller <ms@dev.tdt.de>
5 days agokernel: usbnet: Restore usb%d naming for cdc-ethernet devices with local MAC
Ahmed Naseef [Tue, 28 Jan 2025 07:28:31 +0000 (11:28 +0400)]
kernel: usbnet: Restore usb%d naming for cdc-ethernet devices with local MAC

Prior to commit https://github.com/torvalds/linux/commit/8a7d12d674ac6f2147c18f36d1e15f1a48060edf,
cdc-ethernet USB LTE modems (e.g. Quectel EC200A) were consistently named
usb0. After 8a7d12d67, devices began renaming to eth1 due to an assumption
that local MAC addresses originate exclusively from the kernel. Some
devices provide driver-assigned local MACs, causing point-to-point
interfaces with driver-set MACs to adopt eth%d names instead of usb%d.

Restore the naming exception for point-to-point devices: interfaces
without driver MACs or with driver-provided local MACs will retain the
usb%d convention. This addresses issues reported in [1] and fixed in [2].

[1] https://lore.kernel.org/all/Z00udyMgW6XnAw6h@atmark-techno.com/
[2] https://lore.kernel.org/all/20241203130457.904325-1-asmadeus@codewreck.org/

Tested-by: Ahmed Naseef <naseefkm@gmail.com>
Signed-off-by: Ahmed Naseef <naseefkm@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/17757
Signed-off-by: Robert Marko <robimarko@gmail.com>
6 days agokernel: backport Mediatek SoC EEE support
Qingfang Deng [Wed, 5 Mar 2025 15:23:09 +0000 (23:23 +0800)]
kernel: backport Mediatek SoC EEE support

Backport Mediatek SoC EEE support from net-next upstream.

Signed-off-by: Qingfang Deng <dqfext@gmail.com>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
[refreshed patches]

6 days agolibpcap: backport support for various DSA tags
Daniel Golle [Mon, 3 Mar 2025 13:27:15 +0000 (13:27 +0000)]
libpcap: backport support for various DSA tags

Trying to tcpdump DSA conduits results in errors such as
"unsupported DSA tag: mtk".
Backport two commits adding support for various DSA tags to libpcap.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
6 days agohostapd: add missing #ifdef to fix compile error when 802.11be support is disabled
Felix Fietkau [Wed, 5 Mar 2025 08:19:12 +0000 (09:19 +0100)]
hostapd: add missing #ifdef to fix compile error when 802.11be support is disabled

Fixes: d65d546bce36 ("hostapd: add missing ctrl socket initialization on bss add")
Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 days agoucode-mod-pkgen: fix refcounting issue
Felix Fietkau [Wed, 5 Mar 2025 07:24:01 +0000 (08:24 +0100)]
ucode-mod-pkgen: fix refcounting issue

Do not call ucv_get if the reference is transferred without being used
elsewhere

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 days agoucode-mod-uline: fix refcounting errors
Felix Fietkau [Wed, 5 Mar 2025 07:13:23 +0000 (08:13 +0100)]
ucode-mod-uline: fix refcounting errors

Do not call ucv_get if the reference is transferred without being used
elsewhere

Signed-off-by: Felix Fietkau <nbd@nbd.name>
7 days agohostapd: add missing ctrl socket initialization on bss add
Felix Fietkau [Wed, 5 Mar 2025 07:00:43 +0000 (08:00 +0100)]
hostapd: add missing ctrl socket initialization on bss add

Fixes adding/removing individual bss interfaces

Signed-off-by: Felix Fietkau <nbd@nbd.name>
7 days agohostapd: get reference to object before removal
Matthew Cather [Mon, 3 Mar 2025 21:46:03 +0000 (15:46 -0600)]
hostapd: get reference to object before removal

`ucv_array_set` releases the array's reference to the object being cleared.
If this is the last reference to the object, it will be freed, making our
pointer `val` invalid.

To avoid this, we need to obtain our own reference to the object so we
can safely return `val`.

Signed-off-by: Matthew Cather <mattbob4@gmail.com>
7 days agohostapd: consistent reference counting for registry
Matthew Cather [Mon, 3 Mar 2025 20:00:35 +0000 (14:00 -0600)]
hostapd: consistent reference counting for registry

Since `wpa_ucode_registry_add` collects its own reference to the values added, the
two functions `hostapd_ucode_bss_get_uval` and `hostapd_ucode_iface_get_uval` would
sometimes return a referenced object (from `uc_resource_new`) and sometimes return
an unreferenced object (from `wpa_ucode_registry_get`). Now, both functions always
return a referenced object.

This change also indirectly fixes `hostapd_ucode_bss_get_uval`, ensuring it now
always returns a referenced object.

Signed-off-by: Matthew Cather <mattbob4@gmail.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
7 days agohostapd: clean-up references to local variables
Matthew Cather [Mon, 3 Mar 2025 19:40:39 +0000 (13:40 -0600)]
hostapd: clean-up references to local variables

Remove extra ucv_get calls when passing a referenced value to an object
without using it further.

Signed-off-by: Matthew Cather <mattbob4@gmail.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
7 days agohostapd: fix ucode memory leak with strings
Matthew Cather [Mon, 3 Mar 2025 19:22:11 +0000 (13:22 -0600)]
hostapd: fix ucode memory leak with strings

This fixes a common reference counting bug typically along the lines of:
```
uc_value_push(ucv_get(ucv_string_new(...)));
```
This would leave our new string with a reference count of 2, one from
the construction of the string, the other from `ucv_get`. This would
prevent the strings from being correctly cleaned up when it goes out
of scope.

Signed-off-by: Matthew Cather <mattbob4@gmail.com>
7 days agoairoha: drop MTD Airoha parser patch
Christian Marangi [Tue, 18 Feb 2025 19:35:00 +0000 (20:35 +0100)]
airoha: drop MTD Airoha parser patch

Drop Airoha MTD parser patch as a better solution was agreed with a
fixed partition table.

Tested-by: Aleksander Jan Bajkowski <olek2@wp.pl> # tested on Quantum W1700k
Link: https://github.com/openwrt/openwrt/pull/18112
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
7 days agoairoha: an7581: cleanup kernel config
Christian Marangi [Tue, 18 Feb 2025 19:29:26 +0000 (20:29 +0100)]
airoha: an7581: cleanup kernel config

Cleanup kernel config and drop all unrelated configs. This have the side
effect of fixing the port not going up automatically due to Bridge VLAN
Filtering disabled.

Link: https://github.com/openwrt/openwrt/pull/18112
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
7 days agoscripts/feeds: shallow clone submodules
Cedric CHEDALEUX [Mon, 17 Feb 2025 09:44:36 +0000 (10:44 +0100)]
scripts/feeds: shallow clone submodules

When a feed has submodules, all its submodules are fully cloned whereas
the feed itself is shallowed. Let's be consistent and perform shallow clones
as well for the submodules.

Signed-off-by: Cedric CHEDALEUX <cedric.chedaleux@orange.com>
Link: https://github.com/openwrt/openwrt/pull/18003
Signed-off-by: Robert Marko <robimarko@gmail.com>
7 days agoscripts/feeds: shallow clone for specific commit update
Cedric CHEDALEUX [Mon, 17 Feb 2025 09:41:32 +0000 (10:41 +0100)]
scripts/feeds: shallow clone for specific commit update

When a feed is referenced with a specific commit (i.e. <git_url>^<sha1>),
a full clone was performed and a branch was created from the sha1
and named with the sha1. Other git clones operations are shallowed.

As Git does not support clone at a specific commit, let's first perform
a shallow clone to latest commit, then fetch the relevant commit and
finally checkout it (no more 'pseudo' branch).

It saves bandwith and significantly speeds up the feed update process.

Signed-off-by: Cedric CHEDALEUX <cedric.chedaleux@orange.com>
Link: https://github.com/openwrt/openwrt/pull/18003
Signed-off-by: Robert Marko <robimarko@gmail.com>
8 days agoapk: backport patch fixing broken apk update with wget fetch
Christian Marangi [Tue, 4 Mar 2025 00:03:10 +0000 (01:03 +0100)]
apk: backport patch fixing broken apk update with wget fetch

APK update is currently broken if wget is used as a tool. This wasn't
correctly tested and cause seg fault. Backport the patch fixing this to
restore original functionality.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
9 days agoqualcommax: ipq50xx: Add support for Linksys MR5500
George Moussalem [Thu, 13 Feb 2025 05:54:44 +0000 (09:54 +0400)]
qualcommax: ipq50xx: Add support for Linksys MR5500

Add support for Linksys MR5500 (Hydra 6 Pro).

Speficiations:
* SoC: Qualcomm IPQ5018 (64-bit dual-core ARM Cortex-A53 @ 1.0Ghz)
* Memory: Kingston D2516ECMDXGJD (512 MiB)
* Serial Port: 3v3 TTL 115200n8
* Wi-Fi: IPQ5018 (2x2 2.4 Ghz 802.11b/g/n/ax)
         QCN9024 (4x4:4 5 Ghz 802.11an/ac/ax)
* Ethernet: IPQ5018 integrated virtual switch connected to an external
            QCA8337 switch (4 Ports 10/100/1000 GBASE-T)
* Flash: Gigadevice GD5F2GQ5REYIH (256 MiB)
* LEDs: 1x multi-color PWM LED
        1x blue led for USB (GPIO 19 Active High)
* Buttons: 1x WPS (GPIO 27 Active Low)
           1x Reset (GPIO 28 Acive Low)
           5x ethernet port LEDs (amber for activity & green for link up)
* Peripherals: 1x USB2 (powered by GPIO 17 Active Low)
               support for USB3 will be added in a separate PR
* FCC ID: 2AYRA-03734

Flash instructions:
1. On OEM firmware, login to the device (typically at http://192.168.1.1) and click 'CA'
in the bottom right corner -> Connectivity -> Manual Upgrade. Alternatively, browse to
http://<router IP>/fwupdate.html.
Upgrade firmware using openwrt-qualcommax-ipq50xx-linksys_mr5500-squashfs-factory.bin image.
Optionally install on second partition, after first boot check actual partition:
fw_printenv -n boot_part
and install firmware on second partition using command in case of 2:
mtd -r -e kernel -n write openwrt-qualcommax-ipq50xx-linksys_mr5500-squashfs-factory.bin kernel
and in case of 1:
mtd -r -e alt_kernel -n write openwrt-qualcommax-ipq50xx-linksys_mr5500-squashfs-factory.bin alt_kernel
2. Installation using serial connection from OEM firmware (default login: root, password: admin):
fw_printenv -n boot_part
In case of 2:
flash_erase /dev/mtd12 0 0
nandwrite -p /dev/mtd12 openwrt-qualcommax-ipq50xx-linksys_mr5500-squashfs-factory.bin
or in case of 1:
flash_erase /dev/mtd14 0 0
nandwrite -p /dev/mtd14 openwrt-qualcommax-ipq50xx-linksys_mr5500-squashfs-factory.bin
After first boot install firmware on second partition:
mtd -r -e kernel -n write openwrt-qualcommax-ipq50xx-linksys_mr5500-squashfs-factory.bin kernel
or:
mtd -r -e alt_kernel -n write openwrt-qualcommax-ipq50xx-linksys_mr5500-squashfs-factory.bin alt_kernel
3. Back to the OEM firmware.
Download firmware from OEM website:
MR5500: https://support.linksys.com/kb/article/207-en/
From serial or SSH:
fw_printenv boot_part
in case of 1:
mtd -r -e alt_kernel -n write FW_MR5500_1.1.2.209598_prod.img alt_kernel
else in case of 2:
mtd -r -e kernel -n write FW_MR5500_1.1.2.209598_prod.img kernel
4. Boot from USB
This allows you loading an OpenWrt image into RAM and is meant for recovery scenarios only.
Enable loading image from USB in u-boot. From serial or SSH:
fw_setenv bootusb 'usb start && usbboot &loadaddr && bootm $loadaddr'
fw_setenv bootcmd 'run bootusb; if test $auto_recovery = no; then bootipq; elif test $boot_part = 1; then run bootpart1; else run bootpart2; fi'
Copy OpenWrt initramfs image to USB:
dd bs=1M if=openwrt-qualcommax-ipq50xx-linksys_mr5500-initramfs-uImage.itb of=/dev/sda

Signed-off-by: George Moussalem <george.moussalem@outlook.com>
Link: https://github.com/openwrt/openwrt/pull/17958
Signed-off-by: Robert Marko <robimarko@gmail.com>
9 days agokernel: bump 6.6 to 6.6.80
John Audia [Thu, 27 Feb 2025 20:06:51 +0000 (15:06 -0500)]
kernel: bump 6.6 to 6.6.80

Changelog: https://cdn.kernel.org/pub/linux/kernel/v6.x/ChangeLog-6.6.80

Removed upstreamed:
generic/backport-6.6/819-v6.8-0002-nvmem-Create-a-header-for-internal-sharing.patch[1]
generic/backport-6.6/819-v6.8-0003-nvmem-Simplify-the-add_cells-hook.patch[2]
generic/backport-6.6/819-v6.8-0004-nvmem-Move-and-rename-fixup_cell_info.patch[3]

All other patches automatically rebased.

1. https://web.git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v6.6.80&id=39dfc17a38f77b14f7cb2619bd3488a18d797d5d
2. https://web.git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v6.6.80&id=276dae17ad9757c3813d9e736a0210f05ccdf8b7
3. https://web.git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v6.6.80&id=a0ee898a5024f12572e4ce45202df9b149dadc05

Build system: x86/64
Build-tested: bcm27xx/bcm2712, flogic/xiaomi_redmi-router-ax6000-ubootmod, ramips/tplink_archer-a6-v3
Run-tested: bcm27xx/bcm2712, flogic/xiaomi_redmi-router-ax6000-ubootmod, ramips/tplink_archer-a6-v3

Signed-off-by: John Audia <therealgraysky@proton.me>
Link: https://github.com/openwrt/openwrt/pull/18140
Signed-off-by: Robert Marko <robimarko@gmail.com>
11 days agomediatek: filogic: migrate ASUS TUF AX6000 to upstream PHY LED control
Aleksander Jan Bajkowski [Wed, 12 Feb 2025 21:12:28 +0000 (22:12 +0100)]
mediatek: filogic: migrate ASUS TUF AX6000 to upstream PHY LED control

This commit switches the control of the leds connected to the Maxlinear
GPY211C PHY to an upstream solution. There should be no functional changes.

Signed-off-by: Aleksander Jan Bajkowski <olek2@wp.pl>
11 days agomediatek: filogic: migrate Acer W6/W6d to upstream PHY LED control
Aleksander Jan Bajkowski [Sun, 16 Feb 2025 22:23:17 +0000 (23:23 +0100)]
mediatek: filogic: migrate Acer W6/W6d to upstream PHY LED control

This commit switches the control of the leds connected to the Maxlinear
GPY211C PHY to an upstream solution. There should be no functional changes.

Signed-off-by: Aleksander Jan Bajkowski <olek2@wp.pl>
11 days agomediatek: filogic: migrate Zyxel NWA50AX Pro to upstream PHY LED control
Aleksander Jan Bajkowski [Mon, 17 Feb 2025 18:25:02 +0000 (19:25 +0100)]
mediatek: filogic: migrate Zyxel NWA50AX Pro to upstream PHY LED control

This commit switches the control of the leds connected to the Maxlinear
GPY211C PHY to an upstream solution. There should be no functional changes.

Signed-off-by: Aleksander Jan Bajkowski <olek2@wp.pl>
11 days agounetd: update to Git HEAD (2025-02-28)
Felix Fietkau [Fri, 28 Feb 2025 15:27:36 +0000 (16:27 +0100)]
unetd: update to Git HEAD (2025-02-28)

75a236be122a service: add missing null pointer check
f5341f327539 ubus: add api for generating and validating security tokens
3fab99eab4d5 add udebug support
28d86bd30e97 pex: only respond to update requests when we have network data
8e6f37cc361e pex-msg: ignore no-data responses if version is zero
12e6cf7f63e1 pex: create pex host from update responses
edc8fdae463a ubus: show the local addresses in network status

Signed-off-by: Felix Fietkau <nbd@nbd.name>
11 days agoucode-mod-uline: add support for querying window size from terminal if ioctl fails
Felix Fietkau [Thu, 27 Feb 2025 10:21:22 +0000 (11:21 +0100)]
ucode-mod-uline: add support for querying window size from terminal if ioctl fails

This is useful for running the cli on a serial console

Signed-off-by: Felix Fietkau <nbd@nbd.name>
11 days agocli: use model scope for hook calls
Felix Fietkau [Wed, 26 Feb 2025 09:35:02 +0000 (10:35 +0100)]
cli: use model scope for hook calls

Make the scope consistent with other callbacks

Signed-off-by: Felix Fietkau <nbd@nbd.name>
11 days agocli: add support for create/destroy object editor with single type
Felix Fietkau [Tue, 25 Feb 2025 15:25:16 +0000 (16:25 +0100)]
cli: add support for create/destroy object editor with single type

Allows hooking create/destroy in a separate menu without having the user
specify a type argument.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
11 days agocli: add support for partial completion with separator character
Felix Fietkau [Mon, 24 Feb 2025 20:22:52 +0000 (21:22 +0100)]
cli: add support for partial completion with separator character

Useful for completing long lists of possible values with common prefix

Signed-off-by: Felix Fietkau <nbd@nbd.name>
11 days agorealtek: Drop redundant LED labels
Sander Vanheule [Fri, 28 Feb 2025 15:29:20 +0000 (16:29 +0100)]
realtek: Drop redundant LED labels

Some devices have both the color/function and label property defined.
The label can be constructed from the former properties, making it
redundant.

Signed-off-by: Sander Vanheule <sander@svanheule.net>
11 days agorealtek: Add status LED for Netgear GS310TP
Sander Vanheule [Fri, 28 Feb 2025 13:15:17 +0000 (14:15 +0100)]
realtek: Add status LED for Netgear GS310TP

Power LED is identical to GS308T.

Link: https://forum.openwrt.org/t/222970/11
Signed-off-by: Sander Vanheule <sander@svanheule.net>
12 days agorealtek: add thermal zones for SFP sensors on SKS8300-8X
Bjørn Mork [Fri, 14 Feb 2025 09:59:36 +0000 (10:59 +0100)]
realtek: add thermal zones for SFP sensors on SKS8300-8X

Create thermal zones for SFP internal sensors, enabling shutdown
on critical temperatures.

Signed-off-by: Bjørn Mork <bjorn@mork.no>
Link: https://github.com/openwrt/openwrt/pull/17967
Signed-off-by: Sander Vanheule <sander@svanheule.net>
12 days agorealtek: add thermal zones for SFP sensors on GS1900-10HP
Bjørn Mork [Fri, 14 Feb 2025 09:58:39 +0000 (10:58 +0100)]
realtek: add thermal zones for SFP sensors on GS1900-10HP

Create thermal zones for SFP internal sensors, enabling shutdown
on critical temperatures.

Signed-off-by: Bjørn Mork <bjorn@mork.no>
Link: https://github.com/openwrt/openwrt/pull/17967
Signed-off-by: Sander Vanheule <sander@svanheule.net>
12 days agorealtek: thermal driver for rtl838x and rtl930x SoCs
Bjørn Mork [Wed, 12 Feb 2025 19:45:51 +0000 (20:45 +0100)]
realtek: thermal driver for rtl838x and rtl930x SoCs

Add simple driver reading the internal temperature sensor.

Signed-off-by: Bjørn Mork <bjorn@mork.no>
Link: https://github.com/openwrt/openwrt/pull/17967
Signed-off-by: Sander Vanheule <sander@svanheule.net>
12 days agolibpcap: add missing PKG_CONFIG_DEPENDS entries
Robert Marko [Thu, 27 Feb 2025 10:05:26 +0000 (11:05 +0100)]
libpcap: add missing PKG_CONFIG_DEPENDS entries

Currently, enabling USB, BT or Netfilter support after initial compilation
will not trigger a rebuild, so add the missing PKG_CONFIG_DEPENDS so
that rebuild gets triggered.

Signed-off-by: Robert Marko <robert.marko@sartura.hr>
12 days agokernel: r8126: update to v10.015.00
Álvaro Fernández Rojas [Thu, 27 Feb 2025 07:06:27 +0000 (08:06 +0100)]
kernel: r8126: update to v10.015.00

Changelog: https://github.com/openwrt/rtl8126/compare/10.014.01...10.015.00

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
12 days agokernel: r8125: update to v9.015.00
Álvaro Fernández Rojas [Thu, 27 Feb 2025 07:04:50 +0000 (08:04 +0100)]
kernel: r8125: update to v9.015.00

Changelog: https://github.com/openwrt/rtl8125/compare/9.014.01...9.015.00

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
12 days agoomcproxy: update to latest version
Álvaro Fernández Rojas [Thu, 27 Feb 2025 07:31:12 +0000 (08:31 +0100)]
omcproxy: update to latest version

Changelog: https://github.com/openwrt/omcproxy/compare/bfba2aa75802ff1a70ef2fd3eba53409a8c6e93a...582cd8d3ae7e6af0073ce1d9198387af48affd60

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/18130
Signed-off-by: Robert Marko <robimarko@gmail.com>
13 days agokernel: kmod-btmtk: Extract btmtk.ko into own package
Hauke Mehrtens [Tue, 25 Feb 2025 20:15:19 +0000 (21:15 +0100)]
kernel: kmod-btmtk: Extract btmtk.ko into own package

btmtk.ko is used by btusb.ko and btmtkuart.ko, add it into an own
package and make both packages depend on it.

Fixes: 1c42a0be3619 ("kernel: modules: bluetooth: separating UART and USB drivers")
Link: https://github.com/openwrt/openwrt/pull/18110
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 weeks agorealtek: rtl930x: sgmii support
Bjørn Mork [Wed, 12 Feb 2025 07:56:48 +0000 (08:56 +0100)]
realtek: rtl930x: sgmii support

This makes sgmii work for 1000Base-T SFPs by stupidly adding the sgmii mode
wherever 1000base-x is accepted.  No intelligence has been used in the
process.  But it "works for me".

There is an obvious need for refactoring this code to make it more obvious
how and why we configure the mac/phy link like we do for different modes.

Signed-off-by: Bjørn Mork <bjorn@mork.no>
Link: https://github.com/openwrt/openwrt/pull/17950
Signed-off-by: Sander Vanheule <sander@svanheule.net>
2 weeks agorealtek: rtl93xx: mdio-smbus support for clause 45 and Rollball SFPs
Bjørn Mork [Thu, 6 Feb 2025 17:30:06 +0000 (18:30 +0100)]
realtek: rtl93xx: mdio-smbus support for clause 45 and Rollball SFPs

These features have been added to the mdio-i2c driver and are now used by
the sfp driver. The support is required for some newer SFPs.

Signed-off-by: Bjørn Mork <bjorn@mork.no>
Link: https://github.com/openwrt/openwrt/pull/17950
Signed-off-by: Sander Vanheule <sander@svanheule.net>
2 weeks agorealtek: rtl93xx: support SFPs with phys
Bjørn Mork [Thu, 6 Feb 2025 16:51:48 +0000 (17:51 +0100)]
realtek: rtl93xx: support SFPs with phys

This driver use "phy-handle" as a placeholder for mac configuration
data.  Such handles are therefore required for all ports - even those
connected directly to SFP slots and having a managed property set to
"in-band-status".

The DSA core will register these nodes as if they are real phys. This
prevents later attachment of pluggable phys with errors like

   sfp sfp-p8: sfp_add_phy failed: -EBUSY

Replace the virtual SFP slot handles with "pseudo-phy-handle" to keep
the driver logic as-is but hide the node from the DSA core.

Signed-off-by: Bjørn Mork <bjorn@mork.no>
Link: https://github.com/openwrt/openwrt/pull/17950
Signed-off-by: Sander Vanheule <sander@svanheule.net>
2 weeks agorealtek: sfp: add mdio bus only for sfps with a phy
Bjørn Mork [Wed, 5 Feb 2025 07:37:51 +0000 (08:37 +0100)]
realtek: sfp: add mdio bus only for sfps with a phy

The SMBus patch broke the logic and caused the driver to always
register an mdio bus, regardless of the sfp.  Restore original
logic.

Signed-off-by: Bjørn Mork <bjorn@mork.no>
Link: https://github.com/openwrt/openwrt/pull/17950
Signed-off-by: Sander Vanheule <sander@svanheule.net>
2 weeks agorealtek: sfp: prevent duplicate hwmon devices when re-probing on interface up
Bjørn Mork [Tue, 4 Feb 2025 08:05:40 +0000 (09:05 +0100)]
realtek: sfp: prevent duplicate hwmon devices when re-probing on interface up

Re-probing on interface up will register a new duplicate hwmon device. Skip
the hwmon probe if we already have a sensor device.

Signed-off-by: Bjørn Mork <bjorn@mork.no>
Link: https://github.com/openwrt/openwrt/pull/17950
Signed-off-by: Sander Vanheule <sander@svanheule.net>
2 weeks agorealtek: i2c-rtl9300: fix crash on block transfers
Bjørn Mork [Wed, 5 Feb 2025 06:19:10 +0000 (07:19 +0100)]
realtek: i2c-rtl9300: fix crash on block transfers

Fix a typo which resulted in wrong .read hooks and unset .write
hooks.  This made I2C_SMBUS_BLOCK_DATA transfers dereference the
NULL .write hook and Oops.

Signed-off-by: Bjørn Mork <bjorn@mork.no>
Link: https://github.com/openwrt/openwrt/pull/17950
Signed-off-by: Sander Vanheule <sander@svanheule.net>
2 weeks agorealtek: dsa: silence debug log noise
Bjørn Mork [Tue, 4 Feb 2025 07:43:58 +0000 (08:43 +0100)]
realtek: dsa: silence debug log noise

The log noise emmitted by this driver is overwhelming, even for developers
looking at specific issues.  Demoting to debug allows individual messages
to be dynamically enabled instead.

Signed-off-by: Bjørn Mork <bjorn@mork.no>
Link: https://github.com/openwrt/openwrt/pull/17950
Signed-off-by: Sander Vanheule <sander@svanheule.net>
2 weeks agorealtek: dsa: silence log noise on route offload
Bjørn Mork [Wed, 5 Feb 2025 06:27:02 +0000 (07:27 +0100)]
realtek: dsa: silence log noise on route offload

Adding a static IPv4 route made the driver repeatedly print

 rtl83xx_l3_nexthop_update: Setting up fwding: ip 192.168.1.42, GW mac 0000001b21a7xxxx
 Route with id 3 to 192.168.99.0 / 24
 rtl83xx_l3_nexthop_update: total packets: 0
 Warning: TEMPLATE_FIELD_RANGE_CHK: not configured

These messages are only useful to developers while debugging offloading.
Demote to debug level, which in general is more useful for developers
by allowing precise dynamic control.

Signed-off-by: Bjørn Mork <bjorn@mork.no>
Link: https://github.com/openwrt/openwrt/pull/17950
Signed-off-by: Sander Vanheule <sander@svanheule.net>
2 weeks agoRevert "tools/autoconf-archive: bump to 2024.10.16"
Robert Marko [Tue, 25 Feb 2025 16:22:01 +0000 (17:22 +0100)]
Revert "tools/autoconf-archive: bump to 2024.10.16"

This reverts commit bbb3b985b036193d814e9742d07143330d19a71a.

LuaJIT support that is included in 2024.10.16 was supposed to be optional
but unfortunately, it seems that there is a bug[1] and its now breaking FRR
host builds and more.

So, revert the update for now.

[1] https://lists.gnu.org/archive/html/autoconf-archive-maintainers/2024-11/msg00000.html

Signed-off-by: Robert Marko <robimarko@gmail.com>
2 weeks agokernel: qca-nss-dp: update to 12.5.5
Kristian Skramstad [Sun, 26 Jan 2025 08:50:23 +0000 (09:50 +0100)]
kernel: qca-nss-dp: update to 12.5.5

There is some new changes since 12.5.r2 to 12.5.5, so refresh
and update patches.

Changes:
modified:   package/kernel/qca-nss-dp/Makefile
modified:   package/kernel/qca-nss-dp/patches/0006-nss_dp_main-Use-a-phy-handle-property-to-connect-to-.patch
modified:   package/kernel/qca-nss-dp/patches/0008-nss-dp-allow-setting-netdev-name-from-DTS.patch
deleted:    package/kernel/qca-nss-dp/patches/0011-01-edma_v1-rework-hw_reset-logic-to-permit-rmmod-and-in.patch
deleted:    package/kernel/qca-nss-dp/patches/0011-02-nss_dp_switchdev-correctly-unregister-notifier-on-dp.patch
deleted:    package/kernel/qca-nss-dp/patches/0011-03-nss_dp_main-swap-dp_exit-function-call.patch
deleted:    package/kernel/qca-nss-dp/patches/0011-04-nss_dp_main-call-unregister_netdev-first-in-dp_remov.patch
deleted:    package/kernel/qca-nss-dp/patches/0011-05-nss_dp_main-use-phy_detach-instead-of-disconnect-in-.patch
deleted:    package/kernel/qca-nss-dp/patches/0011-06-edma_v1-skip-edma_disable_port-in-edma_cleanup-subse.patch

Log:
2024-10-16 |4c9f671| [qca-nss-dp] removed the calling of fal_port_autoneg_status_get
2024-10-16 |fb33119| Merge "[qca-nss-dp] Change Debug Level of TX Complete Errors"
2024-10-15 |e197b9d| [qca-nss-dp] Change Debug Level of TX Complete Errors
2024-10-03 |1b7aeb1| Merge remote-tracking branch origin/AU_LINUX_QSDK_GINGER_OPEN_TARGET_ALL.12.5.5.230.918 into HEAD
2024-10-01 |830ac73| Merge "[qca-nss-dp] Enable rmmod support for qca-nss-dp module."
2024-06-05 |aef242d| [qca-nss-dp] Enable rmmod support for qca-nss-dp module.
2024-09-18 |b1bfeb6| Merge "[qca-nss-dp] Move pr_info to pr_debug"
2024-09-12 |54aee26| Merge "[qca-nss-dp] Fix Loopback ring teardown path"
2024-09-10 |21f6567| Merge "[qca-nss-dp] Extend Tx Complete errors to track individual error type"
2024-09-10 |7fee76f| Merge "[qca-nss-dp] Destination VP xmit support"
2024-09-09 |0d4ecfd| [qca-nss-dp] Fix Loopback ring teardown path
2024-08-29 |438ab8e| [qca-nss-dp] Move pr_info to pr_debug
2024-08-29 |3f45e3d| Merge "[qca-nss-dp] Fix EDMA Recovery API to Prevent Multiple RX NAPI Add Calls"
2024-08-23 |7a792b1| [qca-nss-dp] Fix EDMA Recovery API to Prevent Multiple RX NAPI Add Calls
2024-07-08 |1227746| [qca-nss-dp] Add support for DDR extended buffer
2024-08-20 |d45bceb| [qca-nss-dp] Extend Tx Complete errors to track individual error type
2024-06-17 |a8f7155| [qca-nss-dp] Destination VP xmit support
2024-08-21 |a5d189b| Merge "[qca-nss-dp] Support for the PPE-VP Qdisc feature"
2024-08-16 |da97bc7| Merge "[qca-nss-dp] Restrict compilation of a file in DP module"
2024-06-22 |bf90fe9| [qca-nss-dp] Support for the PPE-VP Qdisc feature
2024-02-07 |6ceeb8f| [qca-nss-dp] Enable EDMA driver to allocate from beyond 4GB space.
2024-03-19 |4a7ff28| [qca-nss-dp] Restrict compilation of a file in DP module
2024-07-09 |73bad34| [qca-nss-dp] Check for DS node state before getting ppe queue
2024-05-03 |22cdbd6| [qca-nss-dp] Add PPE-DS Enqueue vp to queue mapping
2024-07-04 |5d6ef18| Merge "[qca-nss-dp] NAPI Budget change for KPI improvement"
2024-07-02 |cd0b543| Merge "[qca-nss-dp] Read fake_mac header indication from rx_desc"
2024-07-01 |160b988| [qca-nss-dp] NAPI Budget change for KPI improvement
2024-05-22 |54c2fd1| [qca-nss-dp]: Fixing SA warnings as part of qca-nss-dp module.
2024-06-22 |602534b| [qca-nss-dp] Read fake_mac header indication from rx_desc
2023-11-24 |10210e2| [qca-nss-dp] EDMA register changes for ipq54xx
2024-05-22 |1af0d03| Merge "[qca-nss-dp]: Change to support PPE-VP path for MLO Assist"
2024-05-15 |2acd9f3| [qca-nss-dp]: Change to support PPE-VP path for MLO Assist
2023-09-27 |1ca59f2| [qca-nss-dp] DP changes for ipq54xx
2024-04-18 |bc60c5a| Merge "[qca-nss-dp] Add support for the XGMAC latency computation"
2024-04-18 |6a67d6d| Merge "[qca-nss-dp] Fix the packets processing in the Rx NAPI"

Signed-off-by: Kristian Skramstad <kristian+github@83.no>
Link: https://github.com/openwrt/openwrt/pull/17731
Signed-off-by: Robert Marko <robimarko@gmail.com>
2 weeks agoqualcommax: fix upgrade function for Linksys MX dual boot devices
Paweł Owoc [Mon, 24 Feb 2025 12:18:46 +0000 (13:18 +0100)]
qualcommax: fix upgrade function for Linksys MX dual boot devices

Function remove_oem_ubi_volume was called before CI_UBIPART variable was defined.

Fixes: https://github.com/openwrt/openwrt/commit/df1f6e1e186a99b2180abac5da87f071b7f6b3fa (qualcommax: ipq807x: Remove OEM UBI volume before upgrade for Linksys MX devices)
Signed-off-by: Paweł Owoc <frut3k7@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/18090
Signed-off-by: Robert Marko <robimarko@gmail.com>
2 weeks agolinux-firmware: update to 20250211
Daniel Golle [Thu, 20 Feb 2025 17:00:56 +0000 (17:00 +0000)]
linux-firmware: update to 20250211

Changes since 20241110:
 bb591083 i915: Update Xe2LPD DMC to v2.28
 68eeb2a5 ASoC: tas2781: Add regbin firmware by index for single device
 31e5b6b0 WHENCE: qca: add missing version information
 e6e52fde WHENCE: qca: add missing version information
 920ec27d WHENCE: split generic QCA section into USB and serial sections
 bb5d129b rtl_bt: Update RTL8852B BT USB FW to 0x0474_842D
 608fdf4c iwlwifi: add Bz/gl FW for core93-123 release
 c1252c30 iwlwifi: update ty/So/Ma firmwares for core93-123 release
 69e83a8d iwlwifi: update cc/Qu/QuZ firmwares for core93-82 release
 5b217397 ASoC: tas2781: Add dsp firmware for new projects
 24625a79 amdgpu: DMCUB update for DCN401
 99720d02 ath12k: WCN7850 hw2.0: update board-2.bin
 a3f44afe ath12k: QCN9274 hw2.0: update to WLAN.WBE.1.4.1-00199-QCAHKSWPL_SILICONZ-1
 94a1e6ed ath12k: QCN9274 hw2.0: update board-2.bin
 3a44e641 ath11k: WCN6750 hw1.0: update board-2.bin
 ccc21e6c ath11k: QCN9074 hw1.0: update to WLAN.HK.2.9.0.1-02146-QCAHKSWPL_SILICONZ-1
 73718ee7 ath11k: QCA6698AQ hw2.1: add to WLAN.HSP.1.1-04479-QCAHSPSWPL_V1_V2_SILICONZ_IOE-1
 06fe00a9 ath11k: QCA6698AQ hw2.1: add board-2.bin
 bc502ac3 ath11k: QCA6390 hw2.0: update board-2.bin
 7003cdef ath11k: QCA2066 hw2.1: update to WLAN.HSP.1.1-03926.13-QCAHSPSWPL_V2_SILICONZ_CE-2.52297.6
 8cd06dd2 ath11k: QCA2066 hw2.1: update board-2.bin
 0dba9647 ath11k: IPQ8074 hw2.0: update to WLAN.HK.2.9.0.1-02146-QCAHKSWPL_SILICONZ-1
 bd0511af ath11k: IPQ6018 hw1.0: update to WLAN.HK.2.7.0.1-02409-QCAHKSWPL_SILICONZ-1
 51f77040 copy-firmware: Fix 'No such file or directory' error.
 e38da104 ath11k: add device-specific firmware for QCM6490 boards
 6b12b964 qca: add more WCN3950 1.3 NVM files
 42101764 qca: add firmware for WCN3950 chips
 93160a02 qca: move QCA6390 firmware to separate section
 03087bcd qca: restore licence information for WCN399x firmware
 61d257d5 amdgpu: DMCUB updates for various ASICs
 a204b433 amdgpu: DMCUB updates forvarious AMDGPU ASICs
 7db7f79e qca: Update Bluetooth WCN6750 1.1.0-00476 firmware to 1.1.3-00069
 bba19e40 qcom:x1e80100: Support for Lenovo T14s G6 Qualcomm platform
 b3553788 qcom:x1e80100: Support for Lenovo T14s G6 Qualcomm platform
 211fbc28 linux-firmware: Update FW files for MRVL SD8997 chips
 4f91bc37 i915: Update Xe2LPD DMC to v2.27
 0ff5d857 qca: Update Bluetooth WCN6856 firmware 2.1.0-00642 to 2.1.0-00650
 9c2bf7af rtl_bt: Update RTL8852B BT USB FW to 0x049B_5037
 8f070131 amdgpu: Update ISP FW for isp v4.1.1
 84e4027f trivial: contrib: wrap the process in try/except to catch server issues
 e75caebf trivial: contrib: use python-magic to detect encoding of emails
 52e96183 QCA: Add Bluetooth firmware for QCA6698
 0e16f416 amdgpu: revert DMCUB 3.1.4 firmware
 7c07d2d2 amlogic: update firmware for w265s2
 634d0a0a mediatek MT7925: update bluetooth firmware to 20250113153307
 3b6b1cba linux-firmware: update firmware for MT7925 WiFi device
 a6cc7472 amdgpu: update psp 13.0.10 firmware
 133355e8 amdgpu: update gc 11.0.3 firmware
 ed8de441 amdgpu: update psp 13.0.8 firmware
 11ec4d49 amdgpu: update psp 13.0.5 firmware
 78caaee2 amdgpu: update vcn 4.0.0 firmware
 d9f8cb2a amdgpu: update psp 13.0.0 firmware
 29761144 amdgpu: update gc 11.0.0 firmware
 9ac860c0 amdgpu: update beige goby firmware
 f959a13f amdgpu: update dimgrey cavefish firmware
 8e978d08 amdgpu: update navy flounder firmware
 a4b34d9c amdgpu: update sienna cichlid firmware
 7b211e6a amdgpu: update navi14 firmware
 4ea4ad03 amdgpu: update smu 14.0.3 firmware
 d3547493 amdgpu: update psp 14.0.3 firmware
 0ca6b708 amdgpu: update gc 12.0.1 firmware
 367fe221 amdgpu: update navi12 firmware
 5d6d3f3e amdgpu: update smu 14.0.2 firmware
 70482da5 amdgpu: update psp 14.0.2 firmware
 95bfb9ef amdgpu: update gc 12.0.0 firmware
 64e0a082 amdgpu: update gc 9.4.3 firmware
 dcb68fd8 amdgpu: update navi10 firmware
 7718769c amdgpu: update vcn 4.0.4 firmware
 02d16372 amdgpu: update psp 13.0.7 firmware
 f0f3895c amdgpu: update gc 11.0.2 firmware
 73c7a2bd amdgpu: update yellow carp firmware
 f07c3754 qcom: correct licence information for SA8775P binaries
 efa7a4c2 qcom: update SLPI firmware for RB5 board
 6dca3000 amdgpu: DMCUB updates for various AMDGPU ASICs
 dfc33728 qcom: add DSP firmware for SA8775p platform
 f5dae710 qcom: correct venus firmware versions
 dd5f35ba qcom: add missing version information
 dcc4a069 linux-firmware: Update firmware (v10) for mt7988 internal
 fd01e808 iwlwifi: add Bz FW for core90-93 release
 ee800d83 linux-firmware: wilc3000: add firmware for WILC3000 WiFi device
 d36d77d5 rtw89: 8852b: update fw to v0.29.29.8
 b93e274c rtw89: 8852c: update fw to v0.27.122.0
 6d24ab23 rtw89: 8922a: update fw to v0.35.54.0
 9dbec479 rtw89: 8922a: update fw to v0.35.52.1 and stuffs
 882976ea rtw89: 8852bt: update fw to v0.29.110.0
 20cace1a rtw89: 8852b: update fw to v0.29.29.7
 b3049665 amdgpu: DMCUB updates for various AMDGPU ASICs
 5d60a596 amdgpu: update sdma 6.0.3 firmware
 a071c69c amdgpu: update psp 13.0.10 firmware
 17dce2a9 amdgpu: update gc 11.0.3 firmware
 256dd4b1 amdgpu: update sdma 4.4.5 firmware
 9ecdd733 amdgpu: update psp 13.0.14 firmware
 c380f294 amdgpu: update gc 9.4.4 firmware
 df61f757 amdgpu: update psp 13.0.5 firmware
 9e98368a amdgpu: update vega20 firmware
 3f7b0107 amdgpu: update vega12 firmware
 8253e1ba amdgpu: update vega10 firmware
 f46e1c5e amdgpu: update vcn 4.0.0 firmware
 3822f749 amdgpu: update psp 13.0.0 firmware
 ca765ac8 amdgpu: update gc 11.0.0 firmware
 e3d76647 amdgpu: update picasso firmware
 5ade735e amdgpu: update beige goby firmware
 ab3f55cb amdgpu: update vangogh firmware
 dcae23bb amdgpu: update dimgrey cavefish firmware
 81afa06d amdgpu: update navy flounder firmware
 38f582b0 amdgpu: update psp 13.0.11 firmware
 2a6b035e amdgpu: update gc 11.0.4 firmware
 a3010763 amdgpu: update vcn 4.0.2 firmware
 5f1d715f amdgpu: update psp 13.0.4 firmware
 29110540 amdgpu: update gc 11.0.1 firmware
 28cfd823 amdgpu: update sienna cichlid firmware
 52d598fe amdgpu: update vcn 4.0.6 firmware
 521a4cd3 amdgpu: update psp 14.0.1 firmware
 28f84675 amdgpu: update vcn 4.0.5 firmware
 1e70894c amdgpu: update gc 11.5.0 firmware
 44e3c3ba amdgpu: update vcn 5.0.0 firmware
 f96ec199 amdgpu: update smu 14.0.3 firmware
 572e86af amdgpu: update psp 14.0.3 firmware
 ffbb5c05 amdgpu: update gc 12.0.1 firmware
 42fe6fa0 amdgpu: update navi14 firmware
 67e008dc amdgpu: update arcturus firmware
 2f09a8e7 amdgpu: update renoir firmware
 c795a0cc amdgpu: update smu 14.0.2 firmware
 b255f18f amdgpu: update psp 14.0.2 firmware
 335a3d30 amdgpu: update gc 12.0.0 firmware
 0910afb8 amdgpu: update navi12 firmware
 210ed963 amdgpu: update vcn 4.0.3 firmware
 45612237 amdgpu: update sdma 4.4.2 firmware
 e7a51c79 amdgpu: update psp 13.0.6 firmware
 32114486 amdgpu: update gc 9.4.3 firmware
 a76f74a1 amdgpu: update yellow carp firmware
 cd6cef46 amdgpu: update vcn 4.0.4 firmware
 3320119d amdgpu: update psp 13.0.7 firmware
 80802ba0 amdgpu: update gc 11.0.2 firmware
 39332647 amdgpu: update navi10 firmware
 4f47e84d amdgpu: update aldebaran firmware
 5e1e0ae5 cirrus: cs35l56: Correct some links to address the correct amp instance
 7ccc69cf linux-firmware: Update firmware file for Intel Bluetooth Magnetar core
 35f74899 linux-firmware: Update firmware file for Intel BlazarU core
 127ac452 linux-firmware: Update firmware file for Intel Bluetooth Solar core
 838bf587 cirrus: cs35l41: Add Firmware for Ayaneo system 1f660105
 af598a3f Fix has_gnu_parallel function
 e39831b1 rtl_bt: Add separate config for RLT8723CS Bluetooth part
 804b3e69 amdgpu: revert VCN 3.1.2 firmware
 599deec0 amdgpu: revert yellow carp VCN firmware
 95867e92 amdgpu: revert sienna cichlid VCN firmware
 81ca4929 amdgpu: revert navy flounder VCN firmware
 382f95f0 amdgpu: revert dimgrey cavefish VCN firmware
 0f3495ac WHENCE: Link the Raspberry Pi CM5 and 500 to the 4B
 729c0c8e copy-firmware.sh: Fix typo in error message.
 2b781910 Add support to install files/symlinks in parallel.
 c170fc69 Makefile: Remove obsolete/broken reference.
 a100ea55 check_whence.py: Use a more portable shebang.
 bf3697e4 rtl_bt: Update RTL8852B BT USB FW to 0x04BE_1F5E
 9cf329b3 cnm: update chips&media wave521c firmware.
 c92f98b8 WHENCE: Add "Info:" tag to text that's clearly not part of the license
 e1e114f8 rtl_nic: add firmware rtl8125bp-2
 891a8d4f qcom: venus-5.4: update firmware binary for sc7180 and qcs615
 bd491523 cirrus: cs35l56: Correct filenames of SSID 17aa3832
 9c46d10f cirrus: cs35l56: Add and update firmware for various Cirrus CS35L54 and CS35L56 laptops
 2cb303e7 cirrus: cs35l56: Correct SSID order for 103c8d01 103c8d08 10431f43
 16329652 rtl_nic: add firmware rtl8125d-2
 7012ac97 linux-firmware: Update firmware file for Intel BlazarU core
 209c18b0 amdgpu: update dmcub 0.0.246.0 firmware
 27272156 Add top level license file.
 f5aeeb47 amdgpu: update raven firmware
 1363a0ea amdgpu: update gc 11.0.3 firmware
 a69ccec7 amdgpu: update psp 13.0.14 firmware
 be80726a amdgpu: update vcn 3.1.2 firmware
 0269927e amdgpu: update vpe 6.1.3 firmware
 66a1d205 amdgpu: update psp 14.0.4 firmware
 929c0891 amdgpu: update gc 11.5.2 firmware
 39ba34e6 amdgpu: update vcn 4.0.0 firmware
 53c443d3 amdgpu: update gc 11.0.0 firmware
 63dcba16 amdgpu: update picasso firmware
 0985337b amdgpu: update beige goby firmware
 6c7f2c6a amdgpu: update vangogh firmware
 7f87800c amdgpu: update dimgrey cavefish firmware
 3b1a3e06 amdgpu: update navy flounder firmware
 1fa991ad amdgpu: update gc 11.0.4 firmware
 c6d3af09 amdgpu: update green sardine firmware
 32de3dbc amdgpu: update vcn 4.0.2 firmware
 d4e69c52 amdgpu: update gc 11.0.1 firmware
 9ae22572 amdgpu: update sienna cichlid firmware
 923200c8 amdgpu: update vcn 4.0.6 firmware
 0599265e amdgpu: update gc 11.5.1 firmware
 0dc221f2 amdgpu: update vcn 4.0.5 firmware
 f6295215 amdgpu: update psp 14.0.0 firmware
 dd3f6661 amdgpu: add vcn 5.0.0 firmware
 614375e1 amdgpu: add smu 14.0.3 firmware
 664fb29d amdgpu: add sdma 7.0.1 firmware
 54a0aa85 amdgpu: add psp 14.0.3 firmware
 72eda6ac amdgpu: add gc 12.0.1 firmware
 90fcc01f amdgpu: update navi14 firmware
 e264aa6f amdgpu: update renoir firmware
 141c3088 amdgpu: add smu 14.0.2 firmware
 49612d12 amdgpu: add sdma 7.0.0 firmware
 95f3ca63 amdgpu: add psp 14.0.2 firmware
 c852aecc amdgpu: add gc 12.0.0 firmware
 51744331 amdgpu: update navi12 firmware
 308d5f14 amdgpu: update psp 13.0.6 firmware
 252ffc5b amdgpu: update yellow carp firmware
 f3d0f0ba amdgpu: update vcn 4.0.4 firmware
 eb22099a amdgpu: update gc 11.0.2 firmware
 c3eccd75 amdgpu: update navi10 firmware
 f367a87d amdgpu: update aldebaran firmware
 52a3bca7 upstream amdnpu firmware
 7f116580 QCA: Add Bluetooth nvm files for WCN785x
 ea71da6f i915: Update Xe2LPD DMC to v2.24
 3f08f270 cirrus: cs35l56: Add firmware for Cirrus CS35L56 for various Dell laptops
 42b01d49 iwlwifi: add Bz-gf FW for core89-91 release
 16414d8b QCA: Update Bluetooth WCN785x firmware to 2.0.0-00515-2
 113d0f19 amdgpu: update smu 13.0.10 firmware
 a956cc9c amdgpu: update sdma 6.0.3 firmware
 72809ef1 amdgpu: update psp 13.0.10 firmware
 cce4f155 amdgpu: update gc 11.0.3 firmware
 b015f2e6 amdgpu: add smu 13.0.14 firmware
 3b20eb12 amdgpu: add sdma 4.4.5 firmware
 e736a04b amdgpu: add psp 13.0.14 firmware
 4017dd0e amdgpu: add gc 9.4.4 firmware
 ad211a73 amdgpu: update vcn 3.1.2 firmware
 821595b4 amdgpu: update psp 13.0.5 firmware
 ebdbabc2 amdgpu: update psp 13.0.8 firmware
 f665d459 amdgpu: update vega20 firmware
 d0e86e75 amdgpu: update vega12 firmware
 1c720ff4 amdgpu: update psp 14.0.4 firmware
 cc7d01ee amdgpu: update gc 11.5.2 firmware
 ac83d136 amdgpu: update vega10 firmware
 92cfa0cc amdgpu: update vcn 4.0.0 firmware
 b3848445 amdgpu: update smu 13.0.0 firmware
 ed656376 amdgpu: update psp 13.0.0 firmware
 f5e2c676 amdgpu: update gc 11.0.0 firmware
 54d70b86 amdgpu: update beige goby firmware
 7416c173 amdgpu: update vangogh firmware
 169ab504 amdgpu: update dimgrey cavefish firmware
 3e78bb66 amdgpu: update navy flounder firmware
 c770f5bb amdgpu: update psp 13.0.11 firmware
 7508bcff amdgpu: update gc 11.0.4 firmware
 95091a45 amdgpu: update vcn 4.0.2 firmware
 9ae39887 amdgpu: update psp 13.0.4 firmware
 bfca88ab amdgpu: update gc 11.0.1 firmware
 c9244523 amdgpu: update sienna cichlid firmware
 5bce792a amdgpu: update vpe 6.1.1 firmware
 398e526f amdgpu: update vcn 4.0.6 firmware
 4a172771 amdgpu: update psp 14.0.1 firmware
 d316e650 amdgpu: update gc 11.5.1 firmware
 0109c8f4 amdgpu: update vcn 4.0.5 firmware
 9fdafc63 amdgpu: update psp 14.0.0 firmware
 cca5bb4a amdgpu: update gc 11.5.0 firmware
 50bd1edb amdgpu: update navi14 firmware
 78829a1a amdgpu: update arcturus firmware
 0b20421f amdgpu: update renoir firmware
 e790f175 amdgpu: update navi12 firmware
 93116bb7 amdgpu: update sdma 4.4.2 firmware
 d9e052f0 amdgpu: update psp 13.0.6 firmware
 8d6dc052 amdgpu: update gc 9.4.3 firmware
 799542f9 amdgpu: update vcn 4.0.4 firmware
 bca51065 amdgpu: update psp 13.0.7 firmware
 ce9cd47f amdgpu: update gc 11.0.2 firmware
 f9e90ac1 amdgpu: update navi10 firmware
 9a186727 amdgpu: update aldebaran firmware
 508d770e ice: update ice DDP wireless_edge package to 1.3.20.0
 1db0dc9f ice: update ice DDP comms package to 1.3.52.0
 b0cdbcef ice: update ice DDP package to ice-1.3.41.0
 b94113ee amdgpu: update DMCUB to v9.0.10.0 for DCN314
 60f838aa amdgpu: update DMCUB to v9.0.10.0 for DCN351
 48bb90cc linux-firmware: Update AMD cpu microcode
 00643cab xe: Update GUC to v70.36.0 for BMG, LNL
 36d0c646 i915: Update GUC to v70.36.0 for ADL-P, DG1, DG2, MTL, TGL
 60cdfe18 iwlwifi: add Bz-gf FW for core91-69 release
 1e7f6588 qcom: venus-5.4: add venus firmware file for qcs615
 aeede7af qcom: update venus firmware file for SC7280
 77a11ffc QCA: Add 22 bluetooth firmware nvm files for QCA2066
 4023ed0b mediatek MT7922: update bluetooth firmware to 20241106163512
 ff344a7f mediatek MT7921: update bluetooth firmware to 20241106151414
 862cd742 linux-firmware: update firmware for MT7922 WiFi device
 45c98412 linux-firmware: update firmware for MT7921 WiFi device
 01842da4 qcom: Add QDU100 firmware image files.
 af7f6e4f qcom: Update aic100 firmware files
 af70c9d5 dedup-firmware.sh: fix infinite loop for --verbose

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
2 weeks agotools/autoconf: Fix relocatable patch
Hannu Nyman [Sat, 22 Feb 2025 16:47:52 +0000 (18:47 +0200)]
tools/autoconf: Fix relocatable patch

Fix the 000-relocatable.patch broken by e0f5ce9. The patch segment about
detecting STAGING_DIR_HOST was erroneously removed, as upstream had
deleted the previous bin/autoconf.as and had implemented it in perl
in bin/autoconf.in. Re-create the previous functionality in that.

Fixes: #18059
Fixes: e0f5ce974 ("tools/autoconf: bump to 2.72")
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
Link: https://github.com/openwrt/openwrt/pull/18073
Signed-off-by: Robert Marko <robimarko@gmail.com>
2 weeks agoucode-mod-uline: fix prompt after set_hint with string bigger than the window size
Felix Fietkau [Mon, 24 Feb 2025 12:30:36 +0000 (13:30 +0100)]
ucode-mod-uline: fix prompt after set_hint with string bigger than the window size

Redraw the prompt after the hint instead of trying to move backwards

Signed-off-by: Felix Fietkau <nbd@nbd.name>
2 weeks agotools: zlib: switch to git source
Shiji Yang [Fri, 21 Feb 2025 11:52:01 +0000 (19:52 +0800)]
tools: zlib: switch to git source

Switch to the git repository source so that we can use zstd
compression algorithm to create smaller package tarball. This
patch also corrected the license file name[1].

[1] https://github.com/madler/zlib/commit/352cb28d12baf02863ff5d4d96be0587ced419a1

Suggested-by: Christian Marangi <ansuelsmth@gmail.com>
Signed-off-by: Shiji Yang <yangshiji66@qq.com>
Link: https://github.com/openwrt/openwrt/pull/17880
Signed-off-by: Robert Marko <robimarko@gmail.com>
2 weeks agozlib: switch to git source
Shiji Yang [Fri, 21 Feb 2025 11:52:01 +0000 (19:52 +0800)]
zlib: switch to git source

Switch to the git repository source so that we can use zstd
compression algorithm to create smaller package tarball. This
patch also corrected the license file name[1] and the CPE ID[2].

[1] https://github.com/madler/zlib/commit/352cb28d12baf02863ff5d4d96be0587ced419a1
[2] https://nvd.nist.gov/products/cpe/detail/95C64A3E-A897-4D55-B74A-D2285440D164

Suggested-by: Christian Marangi <ansuelsmth@gmail.com>
Signed-off-by: Shiji Yang <yangshiji66@qq.com>
Link: https://github.com/openwrt/openwrt/pull/17880
Signed-off-by: Robert Marko <robimarko@gmail.com>
2 weeks agoRevert "tools/automake: bump to 1.17"
Robert Marko [Mon, 24 Feb 2025 11:27:47 +0000 (12:27 +0100)]
Revert "tools/automake: bump to 1.17"

This reverts commit b930ce3bf7aa350cf0252dbf95c9447e51be4de9.

Bump to 1.17 broke SDK portability as it partially reverted commit
("automake: portability fixes") which made Perl PATH "/usr/bin/env perl"
as that is portable, as otherwise automake will set the absolute path to
the Perl binary and this will then fail when using SDK on a different
system as that PATH is not present.

So, since fixing this would require backport of upstream commit
("configure: make perl path with whitespace a warning, not error.") which
requires autoreconf to be done in order for configure to get regenerated
we cannot do it because at that time we do not have automake built.

So, for now revert the bump until upstream makes a new release.

Signed-off-by: Robert Marko <robimarko@gmail.com>
2 weeks agokernel: fix IPv6 TCP GSO segmentation with NAT
Felix Fietkau [Mon, 24 Feb 2025 11:26:01 +0000 (12:26 +0100)]
kernel: fix IPv6 TCP GSO segmentation with NAT

Add missing checksum update

Fixes: https://github.com/openwrt/openwrt/issues/15857
Signed-off-by: Felix Fietkau <nbd@nbd.name>
2 weeks agoethtool: work-around ETHTOOL_GRSSH/ETHTOOL_SRSSH ABI breakage
Daniel Golle [Wed, 12 Feb 2025 04:21:22 +0000 (04:21 +0000)]
ethtool: work-around ETHTOOL_GRSSH/ETHTOOL_SRSSH ABI breakage

ethtool since version 6.9 introduced support for getting/setting RSS
input transformation supported in Linux since version 6.8.

The now changed kernel ioctl ABI, however, cannot be detected from
userland, and ethtool since version 6.9 simply assumes that a previously
reserved field is now used to set the input transformation.
Unfortunately the default value RXH_XFRM_NO_CHANGE (0xff) used by ethtool
userland creates an incompatibility with older kernels which cannot be
resolved easily without introducing even more ABI breakage.

Work-around the issue and fix support for --set-rxfh and --set-rxfh-indir
ethtool userland tool commands by making the support for input_xfrm
conditional on compile time, and keep it disabled for Linux 6.6.

Fixes: 8c2dcd1518 ("ethtool: update to 6.10")
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
2 weeks agowireless-regdb: Update to version 2025.02.20
Rudy Andram [Wed, 12 Feb 2025 19:28:49 +0000 (19:28 +0000)]
wireless-regdb: Update to version 2025.02.20

b43aeb5 wireless-regdb: assert and correct maximum bandwidth within frequency difference
68588bf wireless-regdb: Update regulatory info for Syria (SY) for 2020
0dda57e wireless-regdb: Update regulatory info for Moldova (MD) on 6GHz for 2022
b19ab0b wireless-regdb: Update regulatory info for Azerbaijan (AZ) on 6GHz for 2024
f67f40d wireless-regdb: Update regulatory info for Oman (OM)
bd70876 wireless-regdb: Update regulatory rules for Armenia (AM) on 2.4 and 5 GHz
6c7cbcc wireless-regdb: Permit 320 MHz bandwidth in 6 GHz band in ETSI/CEPT
f9f6b30 wireless-regdb: Update regulatory rules for Austria (AT)
39b47ea wireless-regdb: Update regulatory info for Cayman Islands (KY) for 2024
3dd7ceb wireless-regdb: allow NO-INDOOR flag in db.txt
4d754a1 wireless-regdb: Update regulatory rules for Iran (IR) on both 2.4 and 5Ghz for 2021
8c8308a wireless-regdb: Update frequency range with NO-INDOOR for Oman (OM)
c2f11e2 wireless-regdb: update regulatory database based on preceding changes

Signed-off-by: Rudy Andram <rmandrad@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/17957
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 weeks agoltq-vdsl-vr11-mei: remove static linking
Andre Heider [Tue, 29 Aug 2023 13:48:56 +0000 (15:48 +0200)]
ltq-vdsl-vr11-mei: remove static linking

This removes -static compile option. The -static option tells GCC to
link this statically with the libc, which we do not want in OpenWrt. We
want to link everything dynamically to the libc. This fixes a compile
problem with glibc.

References: 71bdff91 "ltq-vdsl-mei: Remove static linking"
Signed-off-by: Andre Heider <a.heider@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/18056
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 weeks agoltq-vdsl-vr11-mei: remove support for older kernel versions
Andre Heider [Sat, 25 May 2024 06:13:21 +0000 (08:13 +0200)]
ltq-vdsl-vr11-mei: remove support for older kernel versions

There's only v6.6 in-tree, remove anything older.

Signed-off-by: Andre Heider <a.heider@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/18056
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 weeks agolinux: use packages kmod-hci-uart and kmod-btusb instead of kmod-bluetooth
Paweł Owoc [Sun, 5 Jan 2025 10:35:18 +0000 (11:35 +0100)]
linux: use packages kmod-hci-uart and kmod-btusb instead of kmod-bluetooth

Use Bluetooth UART and USB packages separetly.

Signed-off-by: Paweł Owoc <frut3k7@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/15118
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 weeks agokernel: modules: bluetooth: separating UART and USB drivers
Paweł Owoc [Sat, 4 Jan 2025 21:48:01 +0000 (22:48 +0100)]
kernel: modules: bluetooth: separating UART and USB drivers

Not all devices need Bluetooth UART and USB drivers to be installed together. Three separate drivers have been created:
- kmod-bluetooth
- kmod-hci-uart
- kmod-btusb

Signed-off-by: Paweł Owoc <frut3k7@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/15118
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 weeks agokernel: modules: bluetooth: creating a new Bluetooth menu
Paweł Owoc [Tue, 9 Apr 2024 16:57:58 +0000 (18:57 +0200)]
kernel: modules: bluetooth: creating a new Bluetooth menu

Move Bluetooth modules to new Bluetooth menu from Other menu.

Signed-off-by: Paweł Owoc <frut3k7@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/15118
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 weeks agorealtek: ONTi ONT-S508CL-8S is a relabeled XikeStor SKS8300-8X
Bjørn Mork [Sat, 22 Feb 2025 10:11:20 +0000 (11:11 +0100)]
realtek: ONTi ONT-S508CL-8S is a relabeled XikeStor SKS8300-8X

Both hardware and firmware of these devices appears identical except for the
manufacturers logo and device name.  The documented XikeStor SKS8300-8X
installation method is verified to work on the ONTi ONT-S508CL-8S using
Openwrt images made for the XikeStor SKS8300-8X. This includes the OEM boot
loader magic password phrases.

Signed-off-by: Bjørn Mork <bjorn@mork.no>
Link: https://github.com/openwrt/openwrt/pull/18071
Signed-off-by: Sander Vanheule <sander@svanheule.net>
2 weeks agotools: xz: update to 5.6.4
Shiji Yang [Fri, 21 Feb 2025 13:18:22 +0000 (21:18 +0800)]
tools: xz: update to 5.6.4

The serious liblzma backdoor vulnerability (CVE-2024-3094) has
been fixed since v5.6.2. It's time to bump this tool to the
latest version. This patch also added a new GitHub package URL.

Changelogs:
https://github.com/tukaani-project/xz/releases/tag/v5.6.2
https://github.com/tukaani-project/xz/releases/tag/v5.6.3
https://github.com/tukaani-project/xz/releases/tag/v5.6.4

Signed-off-by: Shiji Yang <yangshiji66@qq.com>
Link: https://github.com/openwrt/openwrt/pull/18063
Signed-off-by: Nick Hainke <vincent@systemli.org>
2 weeks agokernel: bump 6.6 to 6.6.79
John Audia [Fri, 21 Feb 2025 14:14:57 +0000 (09:14 -0500)]
kernel: bump 6.6 to 6.6.79

Changelog: https://cdn.kernel.org/pub/linux/kernel/v6.x/ChangeLog-6.6.79

Manually rebased:
ramips/patches-6.6/810-uvc-add-iPassion-iP2970-support.patch

All other patches automatically rebased.

Build system: x86/64
Build-tested: bcm27xx/bcm2712, filogic/glinet_gl-mt6000, filogic/xiaomi_redmi-router-ax6000-ubootmod, ramips/tplink_archer-a6-v3, x86/64
Run-tested: bcm27xx/bcm2712, filogic/glinet_gl-mt6000, filogic/xiaomi_redmi-router-ax6000-ubootmod, ramips/tplink_archer-a6-v3, x86/64

Signed-off-by: John Audia <therealgraysky@proton.me>
Link: https://github.com/openwrt/openwrt/pull/18066
Signed-off-by: Nick Hainke <vincent@systemli.org>
2 weeks agotools: zstd: update to 1.5.7
Shiji Yang [Fri, 21 Feb 2025 13:12:23 +0000 (21:12 +0800)]
tools: zstd: update to 1.5.7

This new version has some performance enhancements.

Changelog:
https://github.com/facebook/zstd/releases/tag/v1.5.7

Upstreamed patch:
001-Provide-variant-pkg-config-file-for-multi-threaded-s.patch[1]

[1] https://github.com/facebook/zstd/commit/f1f1ae369a4cefd3474b3528e8d1847b18750605

Signed-off-by: Shiji Yang <yangshiji66@qq.com>
Link: https://github.com/openwrt/openwrt/pull/18064
Signed-off-by: Robert Marko <robimarko@gmail.com>
2 weeks agoiproute2: update to 6.13
Robert Marko [Fri, 21 Feb 2025 17:20:20 +0000 (18:20 +0100)]
iproute2: update to 6.13

Update iproute2 to 6.13.

Release notes:
https://lwn.net/ml/all/20250120194053.3744d96b@hermes.local/

Allows us to drop multiple upstreamed patches:
package/network/utils/iproute2/patches/013-endian.h.patch
package/network/utils/iproute2/patches/014-basename.patch
package/network/utils/iproute2/patches/015-limits.h.patch
package/network/utils/iproute2/patches/016-limits.h.patch
package/network/utils/iproute2/patches/017-linux-limits.patch
package/network/utils/iproute2/patches/018-linux-limits.patch

Link: https://github.com/openwrt/openwrt/pull/18067
Signed-off-by: Robert Marko <robimarko@gmail.com>
2 weeks agoselinux-policy: update to version v2.1
Dominick Grift [Thu, 16 Jan 2025 09:23:12 +0000 (10:23 +0100)]
selinux-policy: update to version v2.1

Rebased onto dssp5-base. Baseline is:
ss, tc, stubby, irqbalance, usbutils, ethtool, tcpdump, mtr,
bmon, zram-swap, parted, e2fsprogs, gdisk, block-mount,
kmod-fs-ext4, kmod-fs-f2fs, kmod-usb-storage, f2fs-tools-selinux,
kmod-usb-storage-uas, kmod-usb3, wireguard-tools,
openssh-sftp-server, luci-light, resolveip, blockd

Changes since v2.0:

3dcc957 three issues:
434bad8 /new_root related
04d18a5 README armsr combined-efi
506d8c1 /efi is not a thing in armsr combined-efi
c0db1ed efivarfs remove these filecons
4eb35b7 adds efivars noseclabelfs for armsr combined-efi
bbc6a6b adds /dev/ttyS3 to tty serialtermdev
1467206 README local logins
6ae3185 rename eficapsulemiscnodedev
d43ded0 rename to vportserialtermdev
a47d2f8 fixes virtio port serialtermdev
03aec70 blockmount: make it a bit more robust
70f1ed3 hotplugcall not sure what config triggers this
3338764 boarddetect: i was expecting this
d97548a deal with /dev/tty and /dev/vcs
c6ba4a5 adds virtio block device
d03e216 adds virtio vport serialtermdev for qemu guest agent
2dc0291 validatefirmwareimage: allow getattr of *all* dev chr files
fdfb3a7 adds /dev/efi_capsule_loader for armsr combined target
b129fb9 validatefirmwareimage ordering
b5e81b4 validatefirmwareimage clean up
0932dc5 README typo fixes
a1f88f0 README fix
e6c68be README typo fixes
a232c21 hvcloginserialtermdev: macro not used
d7edd95 support /usr/local and update README
c0d2947 validatefirmwareimage comment
7dbc9b3 validatefirmwareimage: allow find to getattr of dev.except char
f647175 platformtmpfile: elaborate a bit in comment
52f32c1 reintroduce misc.cil
320d77f validatefirmwareimage: /lib/upgrade/platform.sh
365fc65 deal with /efi for combined images
183b412 adds hvc logserialtermdev
20cd42a sshdsysagent: limited support for legacy scp -O with firmware images
1ccee8d validatefirmwareimages: some events related to "combined" images
9b47fc3 jshn reads /dev/urandom
f86def7 adds /tmp/log/apk.log
f1247b3 these are relative to the current namespace
32c0cc8 hotplugcall qemu virtio-console-helper
5cc41f5 uclient-fetch for manually downloading sysupgrade related
c8140bd cgi-io creates firmware and backup atomically
edf517b factoryreset deal with firstboot compatibility
f5116b5 pppd: redundant, is implied with shell client type
86be72c updates README
8c08ca1 luci-mod-system: a bit of speculation here
417f4a5 adds /dev/autofs
1ed537c misc.cil: remove
01d014a selinuxsecfile: be more specific
53fca71 rename blockd module
b4c9b15 ttyloginserialtermdev: ordering
016c3c1 sysagent traversal of /root is enough
d0d7c91 hotplugcall: net/00-sysctl
2821746 adds ttyAMA0 and some incomplete rules for board-detect

Run-tested: ilogic-openwrt_one, ipq40xx-generic-linksys_mr8300

Tested-by: Stefan Hellermann <stefan@the2masters.de>
Signed-off-by: Dominick Grift <dominick.grift@defensec.nl>
Link: https://patchwork.ozlabs.org/project/openwrt/patch/20250116092312.1350223-1-dominick.grift@defensec.nl/
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 weeks agoltq-vdsl-vr11-app: exit more quickly if orderly shutdown fails
Jan Hoffmann [Thu, 30 Jan 2025 10:19:30 +0000 (11:19 +0100)]
ltq-vdsl-vr11-app: exit more quickly if orderly shutdown fails

On exit, the app tries to do an orderly shutdown of the DSL connection
before it is stopped forcibly. Since the driver does 3 attempts with a
timeout of 2 seconds each, this might take about 6 seconds in the worst
case.

This is problematic on sysupgrade, because any process that doesn't exit
within 4 seconds is killed. This means that the DSL connection might not
be stopped at all before the actual system upgrade begins.

To avoid this, use the newly added option in the driver to not retry the
L3 request on failure.

Signed-off-by: Jan Hoffmann <jan@3e8.eu>
Link: https://patchwork.ozlabs.org/project/openwrt/patch/20250130102108.1606919-3-jan@3e8.eu/
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 weeks agoltq-vdsl-vr11: add option for L3 request without retry on failure
Jan Hoffmann [Thu, 30 Jan 2025 10:19:29 +0000 (11:19 +0100)]
ltq-vdsl-vr11: add option for L3 request without retry on failure

This allows to attempt an orderly shutdown via L3 request while avoiding
excessive delay in the failure case (up to 6 seconds with the currently
hard-coded 3 attempts).

Signed-off-by: Jan Hoffmann <jan@3e8.eu>
Link: https://patchwork.ozlabs.org/project/openwrt/patch/20250130102108.1606919-2-jan@3e8.eu/
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 weeks agoltq-vdsl-vr9-mei: avoid unnecessary usage of unsafe_memcpy
Jan Hoffmann [Thu, 6 Feb 2025 22:54:25 +0000 (23:54 +0100)]
ltq-vdsl-vr9-mei: avoid unnecessary usage of unsafe_memcpy

Commit 7bc487c12eef ("kernel: ltq-vdsl-vr9-mei: fix warning about
field-spanning write") patched the driver to use unsafe_memcpy in
MEI_IoctlCmdMsgWrite.

However, this is not actually necessary. The assignment of the variable
"pDestPtr" can be modified so that the compiler knows about the correct
size. This way, the check in the fortified memcpy works correctly.

While at it, also adjust all places where similar code is used to copy
from a CMV_STD_MESSAGE_T struct.

Also mark all related structs as packed, because the code (and the
driver in general) seems to rely on that anyway.

Signed-off-by: Jan Hoffmann <jan@3e8.eu>
Link: https://patchwork.ozlabs.org/project/openwrt/patch/20250206225444.2521817-2-jan@3e8.eu/
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>