PKG_MIRROR_HASH is incorrect and the package check is currently failing.
Fixes: 4990ce613b7d ("ipq-wifi: update to 2024-02-17")
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
According to the MT7620A hardware datasheet, GPIO/14 was originally used for RIN of UARTF, but is now used as the WPS LED.
Corrected pinctrl to ensure it works properly in the future.
Signed-off-by: Coia Prant <coiaprant@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/18278
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Add switch LED definitions for TP-Link TL-WR1043ND family, based on data
extracted from ar71xx board file. Update the LED labels to match current
pattern, i.e. drop the "tp-link:" prefix.
Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/12487
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Attaching PHY driver to the switch, while adding LEDs binding causes the
PHY driver to create additional LED instances, handled incorrectly by
the PHY driver, which are non-functional. Use fixed-link to attach the
switch driver, instead of PHY driver, to prevent that.
This has a side effect of not logging switch port up/down events in the kernel
log.
Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/12487
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Add switch LED definitions for TP-Link Archer C7 v1/2/3 family, based on data
extracted from ar71xx board file. Update the LED labels to match current
pattern, i.e. drop the "tp-link:" prefix.
Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/12487
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Attaching PHY driver to the switch, while adding LEDs binding causes the
PHY driver to create additional LED instances, handled incorrectly by
the PHY driver, which are non-functional. Use fixed-link to attach the
switch driver, instead of PHY driver, to prevent that.
This has a side effect of not logging switch port up/down events in the kernel
log.
Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/12487
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Add switch LED definitions for TP-Link TL-WDR4300 family, based on data
extracted from ar71xx board file. Update the LED labels to match current
pattern, i.e. drop the "tp-link:" prefix.
Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/12487
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Attaching PHY driver to the switch, while adding LEDs binding causes the
PHY driver to create additional LED instances, handled incorrectly by
the PHY driver, which are non-functional. Use fixed-link to attach the
switch driver, instead of PHY driver, to prevent that.
This has a side effect of not logging switch port up/down events in the kernel
log.
Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/12487
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
The ar8216 switch driver supports exposing configuration of AR8327 and
AR8337 switch LEDs to the userspace, however it is only configurable
through platform data, causing the devices ported from ar71xx target to
lack the support.
Since there is still a long way to go until we can migrate the target to
qca8k, an interim solution is needed.
Extend ar8327_hw_config_of function to parse a "leds"
subnode, which will populate the missing platform data based on device
tree contents, and restore the existing support for the LEDs.
Standard bindings apply, mapping "reg" property to LED index, with
addition of "qca,led-mode" property, which selects HW (0) or SW (1)
mode, defaulting to HW mode.
Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/12487
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Switch LEDs configured as active-low remain low instead of high upon
initialization, because in ar8327_leds_init, no distinction is made with
regards to LED pattern based on active_low property - only whether HW
mode is active. Select the proper initial pattern based also on
active_low to fix that.
While at that, simplify the equation ruling pattern selection for
setting brightness, avoiding unnecessary binary XOR operation, not
really valid for 'bool' type.
Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/12487
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Devices with chipidea usb controller does not detect usb hub after
phy-names change to "usb", revert it back to "usb-phy"
Fixes: 787cb9d87edb ("ath79: change phy-names to only usb")
Signed-off-by: Simonas Tamošaitis <simsasss@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/18230
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Add missing LEDs and modem control for ASUS 4G-AX56
- wifi2.4G white
- wifi5G white
- wan two-coloured, white and red
- modem four-coloured white, blue, yellow and red
change
label = "xxxx:modem";
to
color = <LED_COLOR_ID_xxxx>;
function = LED_FUNCTION_MOBILE;
- rssi-1 white
- rssi-2 white
- rssi-3 white
and modem reboot and reset
Combined into one commit
Signed-off-by: Henrik Ginstmark <henrik@ginstmark.se>
Link: https://github.com/openwrt/openwrt/pull/17927
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Add support for the Intel E800 series of cards, with switchdev
support enabled for lower CPU usage.
Signed-off-by: Stephen Howell <howels@allthatwemight.be>
Link: https://github.com/openwrt/openwrt/pull/17564
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
This service automatically establishes connections to any hosts that are members
of the same unet network, and allows publish/subscribe exchanges via ubus channels.
Signed-off-by: Felix Fietkau <nbd@nbd.name>
This is useful for keeping specific data on a device across factory reset.
It uses a separate partition (only UBI supported at the moment) to store
the data. The primary use case is storing sensitive data like cryptographic
keys for maintaining a device as part of a network.
Signed-off-by: Felix Fietkau <nbd@nbd.name>
This does not actually create a new private key. Instead, the salt is replaced,
and a xor key is generated which when merged with the key derived from the new
password transforms into the original private key.
Signed-off-by: Felix Fietkau <nbd@nbd.name>
With upstream changes hitting kernel 6.4 the dtb for u7623 ends up with
both mac (gmac) disabled, since this is now the default status in
mt7623.dtsi. Fix this by including mt7623a.dtsi (which already has all
necessary bits) and enabling all revlevant ports. This will also do
a side hustle of assigning proper clocks for power controller and
specifying proper power domain for few devices.
Link: https://lore.kernel.org/all/20230210182505.24597-1-arinc.unal@arinc9.com
Signed-off-by: Tomasz Maciej Nowak <tmn505@gmail.com>
Link: https://patchwork.ozlabs.org/project/openwrt/patch/20250304164507.60511-2-tmn505@terefe.re/
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Apparently U-Boot will discard whole node if requested pin function is
unknown to the driver. This resulted in inability to interact with
U-Boot on the said board, as U-Boot always assumed the recovery key
pressed and issued recovery procedure. Log snippet:
button_gpio gpio-keys: pinctrl_select_state_full: pinctrl_config_one: err=-38
reset button found
button pushed, resetting environment
Recovery procedure also booted recovery image, which didn't affect much
the 23.05.x release, since the root fs argument was valid, so changes
persisted. But as 24.10.x hit with fitblk, the board will boot only
recovery image (initramfs) because of default bootargs will reset on each
boot and U-Boot provided bootargs took precedence.
Fixes: 42eeb22450f2 ("uboot-mediatek: fix factory/reset button")
Signed-off-by: Tomasz Maciej Nowak <tmn505@gmail.com>
Link: https://patchwork.ozlabs.org/project/openwrt/patch/20250304164507.60511-1-tmn505@terefe.re/
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Since 24.10.0, eth0, used for the WAN interface, does not work. From dmesg:
...
[ 1.831126] mt7530-mdio mdio-bus:1f: MT7530 adapts as multi-chip module
[ 1.846204] mtk_soc_eth 1e100000.ethernet eth0: mediatek frame engine at 0xbe100000, irq 19
...
[ 1.933969] mt7530-mdio mdio-bus:1f: MT7530 adapts as multi-chip module
[ 1.967668] mt7530-mdio mdio-bus:1f: configuring for fixed/rgmii link mode
[ 1.975999] mt7530-mdio mdio-bus:1f eth0 (uninitialized): PHY [mt7530-0:00] driver [MediaTek MT7530 PHY] (irq=21)
[ 1.986907] mt7530-mdio mdio-bus:1f: Link is Up - 1Gbps/Full - flow control rx/tx
[ 1.987149] mtk_soc_eth 1e100000.ethernet eth0: error -17 registering interface eth0
[ 2.004157] mt7530-mdio mdio-bus:1f eth1 (uninitialized): PHY [mt7530-0:01] driver [MediaTek MT7530 PHY] (irq=22)
[ 2.017698] mtk_soc_eth 1e100000.ethernet eth0: entered promiscuous mode
[ 2.024849] DSA: tree 0 setup
...
[ 4.249680] mtk_soc_eth 1e100000.ethernet dsa: renamed from eth0
...
Like #15865, it seems that gmac0 does not rename eth0 to dsa until after the
switch ports are initialized, leading to a name collision (error -17 = EEXIST).
This patch follows #17062 by using openwrt,netdev-name to fix the collision.
Signed-off-by: J. S. Seldenthuis <jseldenthuis@lely.com>
Link: https://github.com/openwrt/openwrt/pull/18082
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
This adds a default network configuration for the
Supermicro SuperServer SYS-E302-9D by adding all
onboard network ports to the default `lan` interface.
The network ports `eth0` till `eth3` use the `igb`
driver, whereas `eth4` till `eth7` use `i40e`.
--- Hardware Highlights ---
CPU: Intel Xeon D-2123IT, 2.20GHz, 4 cores
RAM: Up to 256GB in 4 DIMM Slots
Ports: 4x 1GbE, 2x 10GBase-T, 2x 10G SFP+, 1x IPMI 2.0
Video: 1x VGA
USB: 2x 3.0
Signed-off-by: Til Kaiser <mail@tk154.de>
Link: https://github.com/openwrt/openwrt/pull/17990
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Sets the boot flag for the i40e network device driver
to load it at a more early stage of the boot process.
With commit 0a47d518df0d758e8d3b31264cb0428d57c362c3,
I added a boot priority for the mlx4 and mlx5 drivers.
Also, increase those priorities because I think they
are too low since there is currently no "room" for
built-in network device drivers.
That can cause interface order, i.e., name inconsistencies,
when Mellanox ConnectX cards are inserted or removed.
Signed-off-by: Til Kaiser <mail@tk154.de>
Link: https://github.com/openwrt/openwrt/pull/17990
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Using board definition file extracted from stock firmware yields 50%
throughput improvement in RX direction under iperf3 test.
Make the device use temporary files from firmware_qca-wireless.git
temporarily, as well as select the specific variant in the device tree
files. The device uses same board file as the MF286C.
Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/17620
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
ZTE MF286 is an indoor LTE category 12 CPE router with simultaneous
dual-band 802.11ac plus 802.11n Wi-Fi radios and quad-port gigabit
Ethernet switch, FXS and external USB 2.0 port.
Software-wise it's compatible with previous MF286A, save for different
5GHz Wi-Fi board definition file, requiring a separate image.
Hardware highlights:
- CPU: QCA9563 SoC at 775MHz,
- RAM: 128MB DDR2,
- NOR Flash: MX25L1606E 2MB SPI Flash, for U-boot only,
- NAND Flash: W25N01GV 128MB SPI NAND-Flash, for all other data,
- Wi-Fi 5GHz: QCA9886 2x2 MIMO 802.11ac Wave2 radio,
- WI-Fi 2.4GHz: QCA9563 3x3 MIMO 802.11n radio,
- Switch: QCA8337v2 4-port gigabit Ethernet, with single SGMII CPU port,
- WWAN: MDM9250-based category 12 internal LTE modem
in extended mini-PCIE form factor, with 5 internal antennas and
2 external antenna connections, single mini-SIM slot.
- FXS: one external ATA port (handled entirely by modem part) with two
physical connections in parallel,
- USB: Single external USB 2.0 port,
- Switches: power switch, WPS, Wi-Fi and reset buttons,
- LEDs: Wi-Fi, Test (internal). Rest of LEDs (Phone, WWAN, Battery,
Signal state) handled entirely by modem. 4 link status LEDs handled by
the switch on the backside.
- Label MAC device: eth0
Internal modem of MF286C is supported via uqmi.
Console connection: connector X2 is the console port, with the following
pinout, starting from pin 1, which is the topmost pin when the board is
upright:
- VCC (3.3V). Do not use unless you need to source power for the
converer from it.
- TX
- RX
- GND
Default port configuration in U-boot as well as in stock firmware is
115200-8-N-1.
Installation:
Due to different flash layout from stock firmware, sysupgrade from
within stock firmware is impossible, despite it's based on QSDK which
itself is based on OpenWrt.
STEP 0: Stock firmware update:
As installing OpenWrt cuts you off from official firmware updates for
the modem part, it is recommended to update the stock firmware to latest
ath79: support ZTE MF286C
STEP 1: Booting initramfs image:
Method 1: using serial console (RECOMMENDED):
- Have TFTP server running, exposing the OpenWrt initramfs image, and
set your computer's IP address as 192.168.0.22. This is the default
expected by U-boot. You may wish to change that, and alter later
commands accordingly.
- Connect the serial console if you haven't done so already,
- Interrupt boot sequence by pressing any key in U-boot when prompted
- Use the following commands to boot OpenWrt initramfs through TFTP:
setenv serverip 192.168.0.22
setenv ipaddr 192.168.0.1
tftpboot 0x81000000 openwrt-ath79-nand-zte_mf286c-initramfs-kernel.bin
bootm 0x81000000
(Replace server IP and router IP as needed). There is no emergency
TFTP boot sequence triggered by buttons, contrary to MF283+.
- When OpenWrt initramfs finishes booting, proceed to actual
installation.
STEP 2: Backing up original software:
As the stock firmware may be customized by the carrier and is not
officially available in the Internet, IT IS IMPERATIVE to back up the
stock firmware, if you ever plan to returning to stock firmware.
It is highly recommended to perform backup using both methods, to avoid
hassle of reassembling firmware images in future, if a restore is
needed.
Method 1: after booting OpenWrt initramfs image via TFTP:
- Connect your USB-UART adapter
- Dump stock firmware located on stock kernel and ubi partitions:
ssh root@192.168.1.1: cat /dev/mtd9 > mtd3_ubiconcat0.bin
ssh root@192.168.1.1: cat /dev/mtd4 > mtd4_kernel.bin
ssh root@192.168.1.1: cat /dev/mtd9 > mtd5_ubiconcat1.bin
And keep them in a safe place, should a restore be needed in future.
Method 2: using stock firmware:
- Connect an external USB drive formatted with FAT or ext4 to the USB
port.
- The drive will be auto-mounted to /var/usb_disk
- Check the flash layout of the device:
cat /proc/mtd
It should show the following:
mtd0: 000a0000 00010000 "u-boot"
mtd1: 00020000 00010000 "u-boot-env"
mtd2: 00140000 00010000 "reserved1"
mtd3: 000a0000 00020000 "fota-flag"
mtd4: 00080000 00020000 "art"
mtd5: 00080000 00020000 "mac"
mtd6: 000c0000 00020000 "reserved2"
mtd7: 00400000 00020000 "cfg-param"
mtd8: 00400000 00020000 "log"
mtd9: 000a0000 00020000 "oops"
mtd10: 00500000 00020000 "reserved3"
mtd11: 00800000 00020000 "web"
mtd12: 00300000 00020000 "kernel"
mtd13: 01a00000 00020000 "rootfs"
mtd14: 01900000 00020000 "data"
mtd15: 03200000 00020000 "fota"
mtd16: 01d00000 00020000 "firmware"
Differences might indicate that this is NOT a MF286C device but
one of other variants.
- Copy over all MTD partitions, for example by executing the following:
for i in 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15; do cat /dev/mtd$i > \
/var/usb_disk/mtd$i; done
"Firmware" partition can be skipped, it is a concatenation
of "kernel" and "rootfs".
- If the count of MTD partitions is different, this might indicate that
this is not a MF286C device, but one of its other variants.
- (optionally) rename the files according to MTD partition names from
/proc/mtd
- Unmount the filesystem:
umount /var/usb_disk; sync
and then remove the drive.
- Store the files in safe place if you ever plan to return to stock
firmware. This is especially important, because stock firmware for
this device is not available officially, and is usually customized by
the mobile providers.
STEP 3: Actual installation:
- Set your computer IP to 192.168.1.22/24
- scp the sysupgrade image to the device:
scp -O openwrt-ath79-nand-zte_mf286a-squashfs-sysupgrade.bin \
root@192.168.1.1:/tmp/
- ssh into the device and execute sysupgrade:
sysupgrade -n /tmp/openwrt-ath79-nand-zte_mf286a-squashfs-sysupgrade.bin
- Wait for router to reboot to full OpenWrt.
STEP 4: WAN connection establishment
Since the router is equipped with LTE modem as its main WAN interface, it
might be useful to connect to the Internet right away after
installation. To do so, please put the following entries in
/etc/config/network, replacing the specific configuration entries with
one needed for your ISP:
config interface 'wan'
option proto 'qmi'
option device '/dev/cdc-wdm0'
option auth '<auth>' # As required, usually 'none'
option pincode '<pin>' # If required by SIM
option apn '<apn>' # As required by ISP
option pdptype '<pdp>' # Typically 'ipv4', or 'ipv4v6' or 'ipv6'
For example, the following works for most polish ISPs
config interface 'wan'
option proto 'qmi'
option device '/dev/cdc-wdm0'
option auth 'none'
option apn 'internet'
option pdptype 'ipv4'
The required minimum is:
config interface 'wan'
option proto 'qmi'
option device '/dev/cdc-wdm0'
In this case, the modem will use last configured APN from stock
firmware - this should work out of the box, unless your SIM requires
PIN which can't be switched off.
If you have build with LuCI, installing luci-proto-qmi helps with this
task.
Restoring the stock firmware:
- Boot to initramfs as in step 3:
- Completely detach ubi0 partition using ubidetach /dev/ubi0_0
- Copy over the stock kernel image using scp to /tmp
- Erase kernel and restore stock kernel:
(scp mtd4_kernel.bin root@192.168.1.1:/tmp/)
mtd write kernel /tmp/mtd4_kernel.bin
rm /tmp/mtd4_kernel.bin
- Copy over the stock partition backups one-by-one using scp to /tmp, and
restore them individually. Otherwise you might run out of space in
tmpfs:
(scp -O mtd3_ubiconcat0.bin root@192.168.1.1:/tmp/)
mtd write ubiconcat0 /tmp/mtd3_ubiconcat0.bin
rm /tmp/mtd3_ubiconcat0.bin
(scp -O mtd5_ubiconcat1.bin root@192.168.1.1:/tmp/)
mtd write ubiconcat1 /tmp/mtd5_ubiconcat1.bin
rm /tmp/mtd5_ubiconcat1.bin
- If the write was correct, force a device reboot with
reboot -f
Quirks and known issues
- It was observed, that CH340-based USB-UART converters output garbage
during U-boot phase of system boot. At least CP2102 is known to work
properly.
- Kernel partition size is increased to 4MB compared to stock 3MB, to
accomodate future kernel updates - at this moment OpenWrt 5.10 kernel
image is at 2.5MB which is dangerously close to the limit. This has no
effect on booting the system - but keep that in mind when reassembling
an image to restore stock firmware.
- uqmi seems to be unable to change APN manually, so please use the one
you used before in stock firmware first. If you need to change it,
please use protocok '3g' to establish connection once, or use the
following command to change APN (and optionally IP type) manually:
echo -ne 'AT+CGDCONT=1,"IP","<apn>' > /dev/ttyUSB0
- The only usable LED as a "system LED" is the blue debug LED hidden
inside the case. All other LEDs are controlled by modem, on which the
router part has some influence only on Wi-Fi LED.
- GPIO5 used for modem reset is a suicide switch, causing a hardware
reset of whole board, not only the modem. It is attached to
gpio-restart driver, to restart the modem on reboot as well, to ensure
QMI connectivity after reboot, which tends to fail otherwise.
- Modem, as in MF283+, exposes root shell over ADB - while not needed
for OpenWrt operation at all - have fun lurking around.
The same modem module is used as in older MF286.
Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/17620
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Using board definition file extracted from stock firmware yields 50%
throughput improvement in RX direction under iperf3 test.
Make the device use temporary files from firmware_qca-wireless.git
temporarily, as well as select the specific variant in the device tree
files.
Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/17620
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Currently, if the phy driver does not implement the led_brightness_set
function, setting the LED will result in the following error message:
leds mdio-bus:*:green:lan: Setting an LED's brightness failed (-524)
Backport a patch to silence this error message.
Signed-off-by: Chukun Pan <amadeus@jmu.edu.cn>
Link: https://github.com/openwrt/openwrt/pull/18080
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
RTL8261N is used on some Airoha and Realtek devices. Move the driver
from Mediatek to generic so it can be used everywhere.
Signed-off-by: Andrew LaMarche <andrewjlamarche@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/18163
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
This commit adds support for Maginon MC-1200AC.
Hardware specifications:
SoC: MediaTek MT7621
Flash: 16 MB SPI Flash
RAM: 128 MB RAM
Ethernet:
2x 1G RJ45 ports
WLAN:
2.4GHz: MediaTek MT7603E
5GHz: MediaTek MT7613BE
LEDs: Red and blue status lights
Power: 12V DC
UART: 3.3V, 115200 baud, 8N1, like printed on silkscreen (GND,TX,RX,3.3V)
MAC addresses
-------------
+---------+-------------------+
| | MAC example |
+---------+-------------------+
| LAN | 80:3F:5D:xx:xx:72 |
| WAN | 80:3F:5D:xx:xx:73 |
| WLAN 2g | 80:3F:5D:xx:xx:74 |
| WLAN 5g | 80:3F:5D:xx:xx:75 |
+---------+-------------------+
Installation:
The firmware can be flashed via the U-Boot recovery web interface.
To access it, hold the reset button while powering on the device.
U-Boot recovery web interface is then avaiable at 192.168.10.1.
Alternatively, the image can be loaded using the U-Boot serial interface and TFTP.
Signed-off-by: Simon Etzlstorfer <simon@etzi.at>
Link: https://github.com/openwrt/openwrt/pull/17671
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
PKG_MIRROR_HASH is incorrect and fails the download from mirrors.
Fixes: 0aaabffdea2c ("selinux-policy: update to version v2.6")
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
Adds latest 6.6 patches from the Raspberry Pi repository.
These patches were generated from:
https://github.com/raspberrypi/linux/commits/rpi-6.6.y/
With the following command:
git format-patch -N v6.6.83..HEAD
(HEAD -> 08d4e8f52256bd422d8a1f876411603f627d0a82)
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
The CONFIG_NET_SWITCHDEV option is needed by CONFIG_DSA and some other
options. It is boolean, we have to compile it into the kernel it self.
Activate it for all targets in the generic configuration, it is already
activated for most of them. This allows to install DSA drivers as a
module.
On the ramips/mt7620 target the kernel would grown by 4.5kB.
For some small targets which do not support a DSA switch by default the
option is deactivated.
Link: https://github.com/openwrt/openwrt/pull/17668
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
This has several advantages:
* reduction in the size of the kernel and the complete image. Individual
devices only need two of the four binaries. In combination with the second
commit it reduces kernel size by 64.2 kB and image size by 22.8 kB,
* the option to extend this package with firmware for future SoCs,
* combining the kernel and binary blobs with another licence may not be
fully compatible with the licence used by Linux. The current PHY firmware
is built into the kernel. This comit converts it to a package.
Tested on AVM 5490 and BT Home Hub 5A.
Signed-off-by: Aleksander Jan Bajkowski <olek2@wp.pl>
Link: https://github.com/openwrt/openwrt/pull/17669
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
