Commit Graph

1683 Commits

Author SHA1 Message Date
Rosen Penev
b2f1c6ed52 ath79: qca: remove mac-address-increment
nvmem-layout allows removal

Signed-off-by: Rosen Penev <rosenp@gmail.com>
2023-11-25 17:14:04 -08:00
Rosen Penev
e816591e22 ath79: qca: convert to nvmem-layout
Allows replacing mac-address-increment with mac-base.

Signed-off-by: Rosen Penev <rosenp@gmail.com>
2023-11-25 17:14:04 -08:00
Rosen Penev
eea4dfaa7b ath79: mikrotik: fix dts warnings
property has invalid length

Signed-off-by: Rosen Penev <rosenp@gmail.com>
2023-11-25 17:14:04 -08:00
Daniel Linjama
a39a49e323 ath79: add support for D-Link COVR-P2500 A1
Specifications:
* QCA9563, 16 MiB flash, 128 MiB RAM, 2T2R 802.11n
* QCA9886 2T2R 801.11ac Wave 2
* QCA7550 Homeplug AV2 1300
* AR8337, 3 Gigabit ports (1, 2: LAN; 3: WAN)

To make use of PLC functionality, firmware needs to be
provided via plchost (QCA7550 comes without SPI NOR),
patched with the Network Password and MAC.

Flashing via OEM Web Interface
* Flash 'factory.bin' using web-interface
* Wait until firmware succesfully installed and device booted
* Hold down reset button to reset factory defaults (~10 seconds)

Flashing via Recovery Web Interface:
* Hold down reset button during power-on (~10 seconds)
* Recovery Web UI is at 192.168.0.50, no DHCP.
* Flash 'recovery.bin' with
  scripts/flashing/dlink_recovery_upload.py
  (Recovery Web UI does not work with modern OSes)

Return to stock
* Hold down reset button during power-on (~10 seconds)
* Recovery Web UI is at 192.168.0.50, no DHCP.
* Flash unencrypted stock firmware with
  scripts/flashing/dlink_recovery_upload.py
  (Recovery Web UI does not work with modern OSes)

Co-developed-by: Sebastian Schaper <openwrt@sebastianschaper.net>
Signed-off-by: Sebastian Schaper <openwrt@sebastianschaper.net>
Signed-off-by: Daniel Linjama <daniel@dev.linjama.com>
2023-11-23 00:26:28 +01:00
Rafał Miłecki
c60e9d1839 ath79: use "fixed-layout" for Embedded Wireless devices
Those devices have Ethernet interfaces using base MAC address increased
by 0x40 in the 3rd indexed byte (00:00:00:FF:00:00). To describe that we
were using a custom (downstream) "mac-address-increment-byte" property.

The same result can be achieved by using "mac-base" with a properly
adjusted offset value (0x40 << 16). It may be not pretty but it should
work without custom property or downstream kernel patch to support it.

Cc: Ansuel Smith <ansuelsmth@gmail.com>
Cc: Catrinel Catrinescu <cc@80211.de>
Cc: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Reviewed-by: Rosen Penev <rosenp@gmail.com>
2023-11-15 07:14:32 +01:00
John Audia
573c8c3d78 kernel: bump 5.15 to 5.15.138
Changelog: https://cdn.kernel.org/pub/linux/kernel/v5.x/ChangeLog-5.15.138

Removed upstreamed:
	generic/backport-5.15/819-v6.6-0018-nvmem-imx-correct-nregs-for-i.MX6SLL.patch[1]
	generic/backport-5.15/819-v6.6-0019-nvmem-imx-correct-nregs-for-i.MX6UL.patch[2]
	generic/backport-5.15/819-v6.6-0020-nvmem-imx-correct-nregs-for-i.MX6ULL.patch[3]

All other patches automatically rebased.

1. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v5.15.138&id=6e22bf6f00d593b0a7e0b2f418fde89317424671
2. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v5.15.138&id=6efd498009b987a92f57f3bdae476f0503364fb7
3. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v5.15.138&id=6e22bf6f00d593b0a7e0b2f418fde89317424671

Build system: x86_64
Build-tested: ramips/tplink_archer-a6-v3
Run-tested: ramips/tplink_archer-a6-v3

Signed-off-by: John Audia <therealgraysky@proton.me>
2023-11-10 17:33:59 +01:00
Lech Perczak
0c47bdb902 ath79: support Fortinet FAP-220-B
Fortinet FAP-220-B is a dual-radio, dual-band 802.11n enterprise managed
access point with PoE input and single gigabit Ethernet interface.

Hardware highlights:
Power: 802.3af PoE input on Ethernet port, +12V input on 5.5/2.1mm DC jack.
SoC: Atheros AR7161 (MIPS 24kc at 680MHz)
RAM: 64MB DDR400
Flash: 16MB SPI-NOR
Wi-Fi 1: Atheros AR9220 2T2R 802.11abgn (dual-band)
Wi-Fi 2: Atheros AR9223 2T2R 802.11bgn (single-band)
Ethernet: Atheros AR8021 single gigabit Phy (RGMII)
Console: External RS232 port using Cisco 8P8C connector (9600-8-N-1)
USB: Single USB 2.0 host port
LEDs: Power (single colour, green), Wi-Fi 1, Wi-Fi 2, Ethernet, Mode, Status
(dual-colour, green and yellow)
Buttons: reset button hidden in bottom grill,
  in the top row, 2nd column from the right.
Label MAC address: eth0

FCC ID: TVE-220102

Serial port pinout:
3 - TxD
4 - GND
6 - RxD

Installation: The same methods apply as for already supported FAP-221-B.

For both methods, a backup of flash partitions is recommended, as stock firmware
is not freely available on the internet.

(a) Using factory image:

1. Connect console cable to the console port
2. Connect Ethernet interface to your PC
3. Start preferred terminal at 9600-8-N-1
4. Have a TFTP server running on the PC.
5. Put the "factory" image in TFTP root
6. Power on the device
7. Break boot sequence by pressing "Ctrl+C"
8. Press "G". The console will ask you for device IP, server IP, and filename.
   Enter them appropriately.
   The defaults are:
   Server IP: 192.168.1.1 # Update accordingly
   Device IP: 192.168.1.2 # Update accordingly
   Image file: image.out # Use for example: openwrt-ath79-generic-fortinet_fap-220-b-squashfs-factory.bin
9. The device will load the firmware over TFTP, and verify it. When
   verification passes, press "D" to continue installation. The device
   will reboot on completion.

(b) Using initramfs + sysupgrade
1. Connect console cable to the console port
2. Connect Ethernet interface to your PC
3. Start preferred terminal at 9600-8-N-1
4. Have a TFTP server running on the PC.
5. Put the "initramfs" image in TFTP root
6. Power on the device.
7. Break boot sequence by pressing "Ctrl+C"
8. Enter hidden U-boot shell by pressing "K". The password is literal "1".
9. Load the initramfs over TFTP:

   > setenv serverip 192.168.1.1 # Your PC IP
   > setenv ipaddr 192.168.1.22 # Device IP, both have to share a subnet.
   > tftpboot 81000000 openwrt-ath79-generic-fortinet_fap-220-b-initramfs-kernel.bin
   > bootm 81000000

10. (Optional) Copy over contents of at least "fwconcat0", "loader", and "fwconcat1"
    partitions, to allow restoring factory firmware in future:

    # cat /dev/mtd1 > /tmp/mtd1_fwconcat0.bin
    # cat /dev/mtd2 > /tmp/mtd2_loader.bin
    # cat /dev/mtd3 > /tmp/mtd3_fwconcat1.bin

    and then SCP them over to safety at your PC.

11. When the device boots, copy over the sysupgrade image, and execute
    normal upgrade:

    # sysupgrade openwrt-ath79-generic-fortinet_fap-220-b-squashfs-sysupgrade.bin

Return to stock firmware:
1. Boot initramfs image as per initial installation up to point 9
2. Copy over the previously backed up contents over network
3. Write the backed up contents back:

   # mtd write /tmp/mtd1_fwconcat0.bin fwconcat0
   # mtd write /tmp/mtd2_loader.bin loader
   # mtd write /tmp/mtd3_fwconcat1.bin fwconcat1

4. Erase the reserved partition:

   # mtd erase reserved

5. Reboot the device

Quirks and known issues:
- The power LED blinking pattern is disrupted during boot, probably due
  to very slow serial console, which prints a lot during boot compared
  to stock FW.
- "mac-address-ascii" device tree binding cannot yet be used for address
  stored in U-boot partition, because it expects the colons as delimiters,
  which this address lacks. Addresses found in ART partition are used
  instead.
- Due to using kmod-owl-loader, the device will lack wireless interfaces
  while in initramfs, unless you compile it in.
- The device heats up A LOT on the bottom, even when idle. It even
  contains a warning sticker there.
- Stock firmware uses a fully read-write filesystem for its rootfs.
- Stock firmware loads a lot of USB-serial converter drivers for use
  with built-in host, probably meant for hosting modem devices.
- U-boot build of the device is stripped of all branding, despite that
  evidence of it (obviously) being U-boot can be found in the binary.
- The user can break into hidden U-boot shell using key "K" after
  breaking boot sequence. The password is "1" (without quotes).
- Telnet is available by default, with login "admin", without password.
  The same is true for serial console, both drop straight to the Busybox
  shell.
- The web interface drops to the login page again, after successfull
  login.
- Whole image authentication boils down to comparing a device ID against
  one stored in U-boot.
- And this device is apparently made by a security company.

Big thanks for Michael Pratt for providing support for FAP-221-B, which
shares the entirety of image configuration with this device, this saved
me a ton of work.

Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
2023-11-03 23:06:07 +01:00
Lech Perczak
6c12c88d2e ath79: image: extract common part for Fortinet FAP series
In preparation for FAP-220-B support, extract the common part of image
recipe for FAP-221-B.

Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
2023-11-03 23:06:07 +01:00
Lech Perczak
c6a090dbf6 ath79: dts: fortinet_loader: extract common part
In preparation for FAP-220-B support, rename ar934x_fortinet_loader.dtsi
to arxxxx_fortinet_loader.dtsi, to avoid confusion, as FAP-220-B shares
flash layout with FAP-221-B exactly despite different SoC.

While at that, add a label to U-boot partition to allow for nvmem MAC
binding in future.

Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
2023-11-03 23:06:07 +01:00
John Audia
7285f7744f kernel: bump 6.1 to 6.1.61
Changelog: https://cdn.kernel.org/pub/linux/kernel/v6.x/ChangeLog-6.1.61

Removed upstreamed:
  generic/backport-6.1/814-v6.6-0018-nvmem-imx-correct-nregs-for-i.MX6SLL.patch[1]
  generic/backport-6.1/814-v6.6-0019-nvmem-imx-correct-nregs-for-i.MX6UL.patch[2]
  generic/backport-6.1/814-v6.6-0020-nvmem-imx-correct-nregs-for-i.MX6ULL.patch[3]

All other patches automatically rebased.

1. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v6.1.61&id=b90b8633ef62314f3a5f5675106e6dcdec981b6f
2. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v6.1.61&id=37495846b1efc23c1767b17ddd6645cc0ccb9946
3. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v6.1.61&id=116671d25915b913374ccdb2956b5fdaff939dc9

Build system: x86/64
Build-tested: x86/64/AMD Cezanne
Run-tested: x86/64/AMD Cezanne

Signed-off-by: John Audia <therealgraysky@proton.me>
2023-11-03 13:02:30 +01:00
Rafał Miłecki
c997634c01 kernel: fix mtd/NVMEM regression affecting U-Boot env NVMEM driver
Fixes: b595670070 ("kernel: backport nvmem v6.6 fixes and v6.7 changes")
Fixes: https://github.com/openwrt/openwrt/issues/13831
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
2023-10-31 17:33:28 +01:00
Weiping Yang
c7baca3bb6 ath79: add support for GL.iNet GL-S200
Specifications:
SoC: QCA9531(650MHz)
RAM: DDR2 128M
Flash: SPI NOR 16M + SPI NAND 128M
WiFi: 2.4GHz with 2 antennas(WiFi/Thread)
Ethernet:
    1xLAN(10/100M)
    2xWAN(10/100M)
Button: 1x Reset Button
Switch: 1x Mode switch
LED: 1x Blue LED + 1x White LED + 1x Orange LED
IOT: Thread + ZigBee/Zwave

By uboot web failsafe:
Push the reset button for 5 seconds util the power led flash faster,
then use broswer to access http://192.168.1.1

Afterwards upgrade can use sysupgrade image.

Signed-off-by: Weiping Yang <weiping.yang@gl-inet.com>
2023-10-31 13:53:11 +01:00
Shiji Yang
e32f70e706 ath79: increase the rfkill debounce interval for TP-Link Archer C7 v2
Due to circuit issue or silicon defect, sometimes the WiFi switch button
of the Archer C7 v2 can be accidentally triggered multiple times in one
second. This will cause WiFi to be unexpectedly shut down and trigger
'irq 23: nobody cared'[1] warning. Increasing the key debounce interval
to 1000 ms can fix this issue. This patch also add the missing rfkill
key label.

[1] Warning Log:
```
[87765.218511] irq 23: nobody cared (try booting with the "irqpoll" option)
[87765.225331] CPU: 0 PID: 317 Comm: irq/23-keys Not tainted 5.15.118 #0
...
[87765.486246] handlers:
[87765.488543] [<85257547>] 0x800c29a0 threaded [<5c6328a2>] 0x80ffe0b8 [gpio_button_hotplug@4cf73d00+0x1a00]
[87765.498364] Disabling IRQ #23
```

Fixes: https://github.com/openwrt/openwrt/issues/13010
Fixes: https://github.com/openwrt/openwrt/issues/12167
Fixes: https://github.com/openwrt/openwrt/issues/11191
Fixes: https://github.com/openwrt/openwrt/issues/7835

Tested-by: Hans Hasert
Signed-off-by: Shiji Yang <yangshiji66@qq.com>
2023-10-29 23:07:43 +01:00
John Audia
cd86f6c24d kernel: bump 6.1 to 6.1.60
Changelog: https://cdn.kernel.org/pub/linux/kernel/v6.x/ChangeLog-6.1.60

All patches automatically rebased.

Build system: x86/64
Build-tested: x86/64/AMD Cezanne, filogic/xiaomi_redmi-router-ax6000-ubootmod
Run-tested: x86/64/AMD Cezanne, filogic/xiaomi_redmi-router-ax6000-ubootmod

Signed-off-by: John Audia <therealgraysky@proton.me>
2023-10-26 12:08:52 +02:00
Marek Behún
b23c2ec831
kernel: ath79, bmips: refresh 5.15 patches
One LED patch for ath79 and one LED patch for bmips need to be
refreshed.

Signed-off-by: Marek Behún <kabel@kernel.org>
2023-10-23 15:19:49 +02:00
John Audia
bb8fd41f9a kernel: bump 6.1 to 6.1.59
Changelog: https://cdn.kernel.org/pub/linux/kernel/v6.x/ChangeLog-6.1.59

Manually rebased:
	bcm27xx/patches-6.1/950-0124-Add-support-for-all-the-downstream-rpi-sound-card-dr.patch

Removed upstreamed:
	generic/backport-6.1/781-v6.6-02-net-dsa-qca8k-fix-potential-MDIO-bus-conflict-when-a.patch[1]

All other patches automatically rebased.

1. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v6.1.59&id=881050b25b1dda7b0f14d40d1b09bf38cb3b427c

Build system: x86/64
Build-tested: x86/64/AMD Cezanne
Run-tested: x86/64/AMD Cezanne

Signed-off-by: John Audia <therealgraysky@proton.me>
2023-10-21 09:33:12 +02:00
Christian Lamparter
0ba5c0bdc1 ath79: AP105: use fixed layout cell "mac-base"
This drops a use of the deprecated "mac-address-increment".

Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
2023-10-20 18:13:37 +02:00
John Audia
b357564463 kernel: bump 6.1 to 6.1.57
Changelog: https://cdn.kernel.org/pub/linux/kernel/v6.x/ChangeLog-6.1.57

Manually rebased:
	generic/pending-6.1/702-net-ethernet-mtk_eth_soc-enable-threaded-NAPI.patch

Removed upstreamed:
	qualcommax/patches-6.1/0134-PCI-qcom-Fixing-broken-pcie-enumeration-for-2_3_3-co.patch[1]

All other patches automatically rebased.

1. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v6.1.57&id=2dfb5f324d799f4545e17631415aba6d302a8e2b

Build system: x86/64
Build-tested: x86/64/AMD Cezanne
Run-tested: x86/64/AMD Cezanne

Reviewed-by: Robert Marko <robimarko@gmail.com>
Signed-off-by: John Audia <therealgraysky@proton.me>
2023-10-14 15:51:53 +02:00
Koen Vandeputte
9188c77cbe ath79: wpj563: enable 2nd USB controller
The compex WPJ563 actually has both usb controllers wired:

usb0 --> pci-e slot
usb1 --> pin header

As the board exposes it for generic use, enable this controller too.

fixes: #13650
Signed-off-by: Koen Vandeputte <koen.vandeputte@citymesh.com>
2023-10-13 17:47:48 +02:00
John Audia
096bb8ed82 kernel: bump 5.15 to 5.15.135
Changelog: https://cdn.kernel.org/pub/linux/kernel/v5.x/ChangeLog-5.15.135

All patches automatically rebased.

Build system: x86_64
Build-tested: ramips/tplink_archer-a6-v3
Run-tested: ramips/tplink_archer-a6-v3

Signed-off-by: John Audia <therealgraysky@proton.me>
2023-10-12 08:26:06 +02:00
David Bauer
666c80d33f ath79: add WWAN serial driver for GL.iNET GL-XE300
The driver for the cellular modems serial interface was missing from the
default device packages.

The driver is required to interact with the modem using AT commands.
Other devices with a 4G modem also ship with this package, thus let's
add it to the default packages for the board.

Signed-off-by: David Bauer <mail@david-bauer.net>
2023-10-09 12:47:41 +02:00
Koen Vandeputte
d382756b85
ath79: mikrotik: fix build error with kernel 6.1
drivers/mfd/rb4xx-cpld.c:167:19: error: initialization of 'void (*)(struct spi_device *)' from incompatible pointer type 'int (*)(struct spi_device *)' [-Werror=incompatible-pointer-types]
  167 |         .remove = rb4xx_cpld_remove,
      |                   ^~~~~~~~~~~~~~~~~
drivers/mfd/rb4xx-cpld.c:167:19: note: (near initialization for 'rb4xx_cpld_driver.remove')
cc1: some warnings being treated as errors
make[8]: *** [scripts/Makefile.build:250: drivers/mfd/rb4xx-cpld.o] Error 1
make[7]: *** [scripts/Makefile.build:500: drivers/mfd] Error 2
make[6]: *** [scripts/Makefile.build:500: drivers] Error 2
make[5]: *** [Makefile:2012: .] Error 2

As the allocated function does nothing, simply delete it.

Signed-off-by: Koen Vandeputte <koen.vandeputte@citymesh.com>
2023-10-06 21:53:13 +02:00
John Audia
e3559fb445
kernel: bump 6.1 to 6.1.54
Changelog: https://cdn.kernel.org/pub/linux/kernel/v6.x/ChangeLog-6.1.54

Removed upstreamed:
	generic/backport-6.1/020-v6.3-02-UPSTREAM-mm-multi-gen-LRU-rename-lrugen-lists-to-lru.patch[1]
	ipq806x/patches-6.1/140-v6.5-hwspinlock-qcom-add-missing-regmap-config-for-SFPB-M.patch[2]

All other patches automatically rebased.

1. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v6.1.54&id=a73d04c460521e45f257d28d73df096e41ece324
2. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v6.1.54&id=e93bc372dbc0bde133c854c03502a95617041972

Build system: x86/64
Build-tested: x86/64/AMD Cezanne
Run-tested: x86/64/AMD Cezanne

Signed-off-by: John Audia <therealgraysky@proton.me>
2023-09-23 13:10:28 +02:00
John Audia
2bda536a3d
kernel: bump 6.1 to 6.1.53
Changelog: https://cdn.kernel.org/pub/linux/kernel/v6.x/ChangeLog-6.1.53

Removed upstreamed:
	bcm53xx/patches-6.1/032-v6.6-0005-ARM-dts-BCM53573-Drop-nonexistent-usb-cells.patch[1]
	bcm53xx/patches-6.1/032-v6.6-0006-ARM-dts-BCM53573-Add-cells-sizes-to-PCIe-node.patch[2]
	bcm53xx/patches-6.1/032-v6.6-0007-ARM-dts-BCM53573-Use-updated-spi-gpio-binding-proper.patch[3]
	bcm53xx/patches-6.1/032-v6.6-0011-ARM-dts-BCM53573-Fix-Tenda-AC9-switch-CPU-port.patch[4]

All other patches automatically rebased.

Build system: x86/64
Build-tested: x86/64/AMD Cezanne
Run-tested: x86/64/AMD Cezanne

1. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v6.1.53&id=ee1d740374aa73fb32857685eb05167ad87458cf
2. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v6.1.53&id=ab5154ae26c446136827451e907db45d7b92a76f
3. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v6.1.53&id=f5ff6897094fa161be55786cb9e5d5b1bf7a9049
4. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v6.1.53&id=0ef736fec61422794c4a991d46c4ec212b01d8d1

Signed-off-by: John Audia <therealgraysky@proton.me>
2023-09-23 13:10:28 +02:00
John Audia
ac422c9788
kernel: bump 5.15 to 5.15.132
Changelog: https://cdn.kernel.org/pub/linux/kernel/v5.x/ChangeLog-5.15.132

Removed upstreamed:
	bcm53xx/patches-5.15/037-v6.6-0006-ARM-dts-BCM53573-Add-cells-sizes-to-PCIe-node.patch[1]
	bcm53xx/patches-5.15/037-v6.6-0007-ARM-dts-BCM53573-Use-updated-spi-gpio-binding-proper.patch[2]
	bcm53xx/patches-5.15/037-v6.6-0008-ARM-dts-BCM5301X-Extend-RAM-to-full-256MB-for-Linksy.patch[3]

All other patches automatically rebased.

1. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v5.15.132&id=b35f3ca1877e024887df205ede952863d65dad36
2. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v5.15.132&id=2840d9b9c8750be270fb1153ccd5b983cbb5d592
3. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v5.15.132&id=f086e859ddc252c32f0438edff241859c0f022ce

Build system: x86_64
Build-tested: ramips/tplink_archer-a6-v3
Run-tested: ramips/tplink_archer-a6-v3

Signed-off-by: John Audia <therealgraysky@proton.me>
2023-09-20 14:13:00 +02:00
Matthijs Kooijman
d2ce3a61aa
ath79: fix packetloss on some WLR-7100
On some WLR-7100 routers, significant packet loss was observed. This is
fixed by configuring a delay on the GMAC0 RXD and RXDV lines.

The values used in this commit are copied from the values used by the
stock firmare (based on register dumping).

Out of four test routers, the problem was consistently observed on two.
It is unclear what the relevant difference is exactly (the two working
routers were v1 001 with AR1022 and v1 002 with AR9342, the two broken
routers were both v1 002 with AR1022). All PCB routing also seems
identical, so maybe there is some stray capacitance on some of these
that adds just enough delay or so...

With this change, the packet loss disappears on the broken routers,
without introducing new packet loss on the previously working routers.

Note that the PHY *also* has delays enabled (through
`qca,ar8327-initvals`) on both RX and TX lines, but apparently that is
not enough, or it is not effective (registers have been verified to be
written).

For detailed discussion of this issue and debug history, see
https://forum.openwrt.org/t/sitecom-wlr-7100-development-progress/79641

Signed-off-by: Matthijs Kooijman <matthijs@stdin.nl>
2023-09-17 16:39:10 +02:00
Roger Pueyo Centelles
8486c677b8 ath79: add support for MikroTik RouterBOARD 750 r2 (hEX lite)
This patch adds support for the MikroTik RouterBOARD 750 r2, marketed as
hEX lite, a small indoor router with 5x 10/100 Mbps Ethernet ports, one
with PoE in. The device was already supported by the ar71xx target.

Specifications:
 - SoC: Qualcomm Atheros QCA9533
 - Flash: 16 MB SPI NOR
 - RAM: 64 MB
 - Ethernet: 4x 10/100 Mbps LAN, 1x 10/100 Mbps WAN (PoE in)
 - LEDs: 5x Ethernet port activity (green), 1x user (green)
 - Buttons: 1x reset

 See https://mikrotik.com/product/RB750r2 for more details.

Not working:
 - Serial port (already not working in ar71xx)

Flashing:
 TFTP boot initramfs image and then perform sysupgrade. Only the
 "Internet" port will ask for an initramfs image. Follow common
 MikroTik procedure as in https://openwrt.org/toh/mikrotik/common.

Signed-off-by: Roger Pueyo Centelles <roger.pueyo@guifi.net>
2023-09-16 12:49:26 +02:00
David Bauer
7f54d9ba1a Revert "ath79: use kernel 6.1 as default"
This reverts commit c94383de01.

THis commit was not meant to be pushed to main yet.

Signed-off-by: David Bauer <mail@david-bauer.net>
2023-09-09 08:42:50 +02:00
David Bauer
c94383de01 ath79: use kernel 6.1 as default
Signed-off-by: David Bauer <mail@david-bauer.net>
2023-09-08 21:00:26 +02:00
Shiji Yang
aee2af0f74 ath79: enable variable sector size erasure for generic subtarget
Make use of minor sector size (4k) erasure on supported flash chips
to improve spi read/write performance.

Signed-off-by: Shiji Yang <yangshiji66@qq.com>
2023-09-08 18:06:30 +02:00
Shiji Yang
7ba69a94f9 ath79: backport gpio immutable irq_chip support
This patch converts the driver to immutable irq-chip, which can
silence some gpio warnings.

Signed-off-by: Shiji Yang <yangshiji66@qq.com>
2023-09-08 18:06:30 +02:00
Shiji Yang
54758cf24b ath79: ignore the abused interrupt-map on PCIe node
ath79 PCIe interrupt controller has stopped working correctly. This
is because the DT exposing a non-sensical interrupt-map property,
and their drivers relying on the kernel ignoring this property[1].

This patch fixes the PCIe init error:
ath9k 0000:00:00.0: of_irq_parse_pci: failed with rc=-14

Notice:
This is just a workaround, not a fix. PCIe driver and related dts
node need to be rewritten.

[1] https://lore.kernel.org/all/20211201114102.13446-1-maz@kernel.org/

Signed-off-by: Shiji Yang <yangshiji66@qq.com>
2023-09-08 18:06:30 +02:00
Shiji Yang
7189b45784 ath79: fix ethernet driver build errors on kernel 6.1
Some net APIs have changed on the new kernel. Update them to fix
compile errors.

Signed-off-by: Shiji Yang <yangshiji66@qq.com>
2023-09-08 18:06:30 +02:00
Shiji Yang
5f59d28bc3 ath79: refresh patches and configs to introduce kernel 6.1 support
All kernel configs are refreshed by
'make kernel_oldconfig CONFIG_TARGET=target' and
'make kernel_oldconfig CONFIG_TARGET=subtarget'.

upstreamed patches:
010-v5.17-spi-ar934x-fix-transfer-and-word-delays.patch
011-v5.17-spi-ar934x-fix-transfer-size.patch
020-v5.18-spi-ath79-Implement-the-spi_mem-interface.patch
030-v5.18-ath79-add-support-for-booting-QCN550x.patch

build and run tested on:
ath79/generic/ar7241
ath79/generic/qca9563
ath79/nand/ar9344

Signed-off-by: Shiji Yang <yangshiji66@qq.com>
2023-09-08 18:06:30 +02:00
Shiji Yang
d9a9caf352 ath79: copy patches and kernel config from 5.15 to 6.1
This is preparation for kernel 6.1 support.

Signed-off-by: Shiji Yang <yangshiji66@qq.com>
2023-09-08 18:06:30 +02:00
Shiji Yang
c60dd7bef9 ath79: rename and sort patches by OpenWrt naming rules
The patches in the ath79 target have not been sorted for a long time
and they are very chaotic now. This patch sorts them again according
to the OpenWrt naming rules[1], so that we can better manage them.

[1] https://openwrt.org/docs/guide-developer/toolchain/use-patches-with-buildsystem#naming_patches

Signed-off-by: Shiji Yang <yangshiji66@qq.com>
2023-09-08 18:06:30 +02:00
Shiji Yang
496280ef4e ath79: add missing symbols by refreshing kernel configs
Some symbols are outdated or missing due to daily kernel bumps. It's
better to re-add them. All configs are automatically refreshed by
'make kernel_oldconfig CONFIG_TARGET=taget' and
'make kernel_oldconfig CONFIG_TARGET=subtarget'

Signed-off-by: Shiji Yang <yangshiji66@qq.com>
2023-09-08 18:06:30 +02:00
Shiji Yang
12f53724c6 ath79: fix first reboot issue on Netgear WNDR4300 v2 and WNDR4500 v3
From the Netgear u-boot GPL code[1]. Bootloader always unconditionally
marks block 768, 1020 - 1023 as bad blocks on each boot. This may lead
to conflicts with the OpenWrt nand driver since these blocks may be good
blocks. In this case, U-boot will override the oob of these blocks so
that break the ubi volume. The system will be damaged after first reboot.
To avoid this issue, manually skip these blocks by using "mtd-concat".

[1] https://www.downloads.netgear.com/files/GPL/EX7300v2series-V1.0.0.146_gpl_src.tar.bz2.zip

Fixes: https://github.com/openwrt/openwrt/issues/8878
Tested-by: Yousaf <yousaf465@gmail.com>
Signed-off-by: Shiji Yang <yangshiji66@qq.com>
2023-08-24 00:04:38 +02:00
Shiji Yang
0f9b8aa3f5 ath79: rework Netgear nand devices image recipe
In Netgear u-boot GPL code, nand devices uses this formula to locate the
rootfs offset.

offset = (((128 + KERNEL_SIZE) / BLOCK_SIZE) + 1) * BLOCK_SIZE;

Howerver, WNDR4500 source code incorrectly define the nand block size to
64k. In some cases, it causes u-boot can't get the correct rootfs offset,
which result in boot failure. This patch workaround it by padding kernel
size to (128k * n - 128 - 1). The additional char '\0' is used to ensure
the (128 + KERNEL_SIZE) can't be divided by the BLOCK_SIZE.

Fixes: https://github.com/openwrt/openwrt/issues/13050
Fixes: 3c1512a25d ("ath79: optimize the firmware recipe for Netgear NAND devices")
Tested-by: Yousaf <yousaf465@gmail.com>
Signed-off-by: Shiji Yang <yangshiji66@qq.com>
2023-08-24 00:04:38 +02:00
John Audia
daed3322d3 kernel: bump 5.15 to 5.15.125
1. Add new symbols to generic config
2. Bump kernel
   Changelog: https://lore.kernel.org/stable/2023080818-groin-gradient-a031@gregkh/

   All patches automatically rebased.

Signed-off-by: John Audia <therealgraysky@proton.me>
2023-08-09 22:06:24 +02:00
John Audia
81c1172c36 kernel: bump 5.15 to 5.15.124
Changelog: https://lore.kernel.org/stable/2023080341-curliness-salary-4158@gregkh/

1. Needed to make a change to to package/kernel/linux/modules/netsupport.mk
   due to upstream moving vxlan to its own directory[1].  @john-tho suggested
   using the the 6.1 xvlan FILES to circumvent.
2. All patches automatically rebased.

1. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v5.15.124&id=77396fa9096abdbfbb87d63e73ad44d5621cf103

Signed-off-by: John Audia <therealgraysky@proton.me>
2023-08-09 22:06:23 +02:00
John Audia
8590531048 kernel: bump 5.15 to 5.15.123
Manually rebased:
	bcm4908/patches-5.15/071-v6.1-0001-net-broadcom-bcm4908_enet-handle-EPROBE_DEFER-when-g.patch
	bcm53xx/patches-5.15/180-usb-xhci-add-support-for-performing-fake-doorbell.patch
	ipq40xx/patches-5.15/902-dts-ipq4019-ap-dk04.1.patch[*]

Removed upstreamed:
	backport-5.15/735-v6.5-net-bgmac-postpone-turning-IRQs-off-to-avoid-SoC-han.patch[1]
	backport-5.15/817-v6.5-01-leds-trigger-netdev-Recheck-NETDEV_LED_MODE_LINKUP-o.patch[2]
	pending-5.15/143-jffs2-reduce-stack-usage-in-jffs2_build_xattr_subsys.patch[3]
	pending-5.15/160-workqueue-fix-enum-type-for-gcc-13.patch[4]
	bcm53xx/patches-5.15/036-v6.5-0003-ARM-dts-BCM5301X-Drop-clock-names-from-the-SPI-node.patch[5]
	bcm53xx/patches-5.15/036-v6.5-0015-ARM-dts-BCM5301X-fix-duplex-full-full-duplex.patch[6]

All other patches automatically rebased.

1. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v5.15.123&id=02474292a44205c1eb5a03634ead155a3c9134f4
2. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v5.15.123&id=86b93cbfe104e99fd3d25a49748b99fb88101573
3. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v5.15.123&id=79b9ab357b6f5675007f4c02ff8765cbd8dc06a2
4. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v5.15.123&id=d528faa9e828b9fc46dfb684a2a9fd8c2e860ed8
5. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v5.15.123&id=5899bc4058e89d5110a23797ff94439c53b77c25
6. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v5.15.123&id=95afd2c7c7d26087730dc938709e025a303e5499

Build system: x86/64
Build-tested: ramips/tplink_archer-a6-v3
Run-tested: ramips/tplink_archer-a6-v3

Co-authored-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: John Audia <therealgraysky@proton.me>
[rebased ipq40xx/patches-5.15/902-dts-ipq4019-ap-dk04.1.patch ]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2023-07-30 18:02:47 +02:00
Rafał Miłecki
daaa0c1b25 ath79: replace "mac-address-ascii" with "mac-base"
With upstream accepted "mac-base" binding there is no need for a
downstream "mac-address-ascii" workaround anymore.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
2023-07-28 10:28:05 +02:00
Felix Baumann
9e86a96af5 ath79: move ubnt-xm 64M RAM boards back to generic
return ubnt_rocket-m and ubnt_powerbridge-m back to ath79-generic
They have enough RAM-ressources to not be considered as tiny.

This reverts the commit f4415f7635 partially

Signed-off-by: Felix Baumann <felix.bau@gmx.de>
2023-07-19 08:04:08 +02:00
Wenli Looi
520c9917f8 ath79: add support for ASUS RT-AC59U / ZenWiFi CD6
ASUS RT-AC59U / RT-AC59U v2 are wi-fi routers with a large number of
alternate names, including RT-AC1200GE, RT-AC1300G PLUS, RT-AC1500UHP,
RT-AC57U v2/v3, RT-AC58U v2/v3, and RT-ACRH12.

ASUS ZenWiFi AC Mini(CD6) is a mesh wifi system. The unit labeled CD6R
is the router, and CD6N is the node.

Hardware:

- SoC: QCN5502
- RAM: 128 MiB
- UART: 115200 baud (labeled on boards)
- Wireless:
  - 2.4GHz: QCN5502 on-chip 4x4 802.11b/g/n
    currently unsupported due to missing support for QCN550x in ath9k
  - 5GHz: QCA9888 pcie 5GHz 2x2 802.11a/n/ac
- Flash: SPI NOR
  - RT-AC59U / CD6N: 16 MiB
  - RT-AC59U v2 / CD6R: 32 MiB
- Ethernet: gigabit
  - RT-AC59U / RT-AC59U v2: 4x LAN 1x WAN
  - CD6R: 3x LAN 1x WAN
  - CD6N: 2x LAN
- USB:
  - RT-AC59U / RT-AC59U v2: 1 port USB 2.0
  - CD6R / CD6N: none

WiFi calibration data contains valid MAC addresses.

The initramfs image is uncompressed because I was unable to boot a
compressed initramfs from memory (gzip or lzma). Booting a compressed
image from flash works fine.

Installation:

To install without opening the case:

- Set your computer IP address to 192.168.1.10/24
- Power up with the Reset button pressed
- Release the Reset button after about 5 seconds or until you see the
  power LED blinking slowly
- Upload OpenWRT factory image via TFTP client to 192.168.1.1

Revert to stock firmware using the same TFTP method.

Signed-off-by: Wenli Looi <wlooi@ucalgary.ca>
2023-07-08 20:19:00 +02:00
Christian Svensson
c170fc78ba kernel: remove CRYPTO_BLAKE2S from all >=5.15
This option was removed from upstream kernel back in 2022.
See commits:
 2d16803c562ecc644803d42ba98a8e0aef9c014e (>=6.0)
 3dd33a09f5dc12ccb0902923c4c784eb0f8c7554 (>=5.15.61 backport)

Signed-off-by: Christian Svensson <blue@cmd.nu>
2023-07-08 16:54:01 +02:00
Wenli Looi
f2f33f77c4 ath79: fix broken 02_network script
Script was broken by an extraneous space.

Signed-off-by: Wenli Looi <wlooi@ucalgary.ca>
2023-07-02 01:21:27 +02:00
Joao Henrique Albuquerque
935a63c59d ath79: add support for COMFAST CF-E380AC v2
COMFAST CF-E380AC v2 is a ceiling mount AP with PoE
support, based on Qualcomm/Atheros QCA9558+QCA9880+AR8035.

There are two versions of this model, with different RAM
and U-Boot mtd partition sizes:
- v1: 128 MB of RAM, 128 KB U-Boot image size
- v2: 256 MB of RAM, 256 KB U-Boot image size

Version number is available only inside vendor GUI,
hardware and markings are the same.

Short specification:

- 720/600/200 MHz (CPU/DDR/AHB)
- 1x 10/100/1000 Mbps Ethernet, with PoE support
- 128 or 256 MB of RAM (DDR2)
- 16 MB of FLASH
- 3T3R 2.4 GHz, with external PA (SE2576L), up to 28 dBm
- 3T3R 5 GHz, with external PA (SE5003L1), up to 30 dBm
- 6x internal antennas
- 1x RGB LED, 1x button
- UART (T11), LEDs/GPIO (J7) and USB (T12) headers on PCB
- external watchdog (Pericon Technology PT7A7514)

COMFAST MAC addresses :
Though the OEM firmware has four adresses in the usual locations,
it appears that the assigned addresses are just incremented in a different way:

Interface    address    location
Lan              *:00           0x0
2.4g             *:0A           n/a (0x0 + 10)
5g               *:02           0x6

Unused Addresses found in ART hexdump
address    location
*:01           0x1002
*:03           0x5006

To keep code consistency the MAC address assignments are made based on increments of the one found in 0x0;

Signed-off-by: Joao Henrique Albuquerque <joaohccalbu@gmail.com>
2023-07-01 16:11:27 +02:00
Pavel Pernička
dac0a133cf ath79: DTS improvement for buzzer on RB951G-2HnD
Mikrotik RB951 router has a buzzer on the board, which makes annoying noises
due to the interference caused by PoE input or Wifi transmission
when no GPIO pin state is set.
I added buzzer node to device's DTS in order to set deault level to 1
and to provide easier access for it.

Signed-off-by: Pavel Pernička <pernicka.pa@gmail.com>
2023-07-01 15:51:26 +02:00
John Audia
42cb0f0f26 kernel: bump 5.15 to 5.15.119
Build system: x86_64
Build-tested: x86_64/ACEMAGICIAN T8PLUS, ramips/tplink_archer-a6-v3
Run-tested: x86_64/ACEMAGICIAN T8PLUS, ramips/tplink_archer-a6-v3

Signed-off-by: John Audia <therealgraysky@proton.me>
2023-07-01 11:42:11 +02:00