Florian Eckert [Wed, 11 Aug 2021 09:58:45 +0000 (11:58 +0200)]
target/x86: add grub2-bios-setup to DEFAULT_PACKAGES
With the commit 5876d6a62fc0ae5799e7d9c896356f75c99a6f0a the command under
`/usr/sbin/grub-bios-setup` has been moved to its own package named
`grub-bios-setup`.
The script `81_upgrade_bootloader` under `/lib/preinit` is used by all
x86 targets to update the bootloader. The script is using the command
`grub-bios-setup` for this.
I get the following output at the first boot after the upgrade.
`/etc/preinit: line 9: /usr/sbin/grub-bios-setup: not found`.
To fix this, the DEFAULT_PACKAGES dependency is extended by the entry
`grub2-bios-setup` so that the missing command is installed again.
Alex Low [Mon, 19 Sep 2022 10:20:37 +0000 (12:20 +0200)]
build: harden GitHub workflow permissions
Grant pull-requests write permission to the labeler workflow and
read-only to everything else.
Signed-off-by: Alex Low <aleksandrosansan@gmail.com>
[ wrap to 80 columns and fix wrong author as requested by author itself ] Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
realtek: use upstream recommendation for secondary CPU start
Currently we fix interrupts/timers for the secondary CPU by patching
vsmp_init_secondary(). Get a little bit more generic and use the
upstream recommended way instead. Additionally avoid a check around
register_cps_smp_ops() because it does that itself.
See https://lkml.org/lkml/2022/9/12/522
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
The interrupt controller depends on two control registers. GIMR enables
or disables interrupts and IRRx routes these to MIPS CPU interrupts 2-7.
Wiki currently states "A value of '0' (in IRRx) disconnects this input from
the output line, independent of the line's setting in GIMR."
Contrary to normal intuition this statement DOES NOT mean, that interrupts
can be disabled by IRRx alone. The sad truth was discovered by enabling
SMP for an Zyxel XGS1010 on the 930x target. It shows that driver and
interrupts behave as follows:
- Timer 0 interrupt 7 has active routing to CPU0 and no routing to CPU1
- Timer 1 interrupt 8 has no routing to CPU0 and active routing to CPU1
- Unmasking (enabling) interrupts writes 1 bits to all GIMR registers
- Masking (disabling) interrupts writes 0 bits to both GIMR registers
During operation we can encounter a situation like
- GIMR bit for a interrupt/CPU combination is set to enabed (=1)
- IRRx routing bits for a interrupt/CPU combination are set to disabed (=0)
This setting already allows the hardware to fire interrupts to the target
CPU/VPE if the other CPU/VPE is currently busy. Especially for CPU bound
timer interrupts this is lethal. If timer interrupt 7 arrives at CPU1 and
vice versa for interrupt 8 the restart trigger gets lost. The timer dies
and a msleep() operation in the kernel will halt endlessly.
Fix this by tracking the IRRx active routing setting in a new bitfield with
0="routing active" and 1="no routing". Enable interrupts in GIMR only
for a interrupt & CPU if routing is active. Thus we have
- GIMR = 0 / IRRx = 0 -> everything disabled
- GIMR = 1 / IRRx > 0 -> active and normal routing
- GIMR = 0 / IRRx > 0 -> masked (disabled) with normal routing
- GIMR = 1 / IRRx = 0 -> no longer possible
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Sungbo Eo [Sun, 18 Sep 2022 01:13:56 +0000 (10:13 +0900)]
ipq40xx: disable devices with 3MiB kernel size
The image builds for Linksys EA6350 v3, EA8300, and MR8300 currently
fail on buildbots due to the KERNEL_SIZE, as stated in commit 17b7756b5a20 ("ipq40xx: 5.15: add testing kernel version"). Disable
these boards for now.
The Build prefix is used for image build commands, while the Device
prefix should be used for base recipes for devices. Apply the same
naming convention here.
While touching the file, also fix the mixed indentation.
realtek: move Netgear recipe to subtarget Makefile
There seems to be no reason to have the Netgear switches as part of
the main Makefile. Move it to its subtarget-specific Makefile since
it is only applicable there.
Currently supported HPE 1920 devices all have an RTL838x SoC, but there
are larger switches with RTL839x SoCs, although currently not supported.
Move the build recipe to common.mk so the larger devices can also make
use of the recipe, while moving it out of the main Makefile.
The D-Link DGS-1210 device series currently has supported devices with
both RTL838x and RTL839x SoCs. An image build recipe has been defined in
both subtarget makefiles, but these are mostly identical, save for the
SOC variable.
Move the SOC variable from the DGS-1210 build recipes to the applicable
devices, and put the remaining duplicate code in a shared Makefile.
* RTL8393M SoC, 1 MIPS 34Kc core @ 700MHz
* 128MB DRAM
* 32MB NOR Flash
* 48 x 10/100/1000BASE-T ports
- 6 x External PHY with 8 ports (RTL8218D)
* 4 x Gigabit RJ45/SFP Combo ports
- External PHY with 4 SFP ports (RTL8214FC)
* Power LED
* Reset button on front panel
* UART (115200 8N1) via unpopulated standard 0.1" pin header marked J14
The gpio-restart node is not required but it does reset the switch.
TODO: The 4 combo ports attached to the RTL8214FC are not detect
properly. Linux kernel reports 49 and 50 as "External RTL8393 SERDES"
and 51 and 52 as "RTL8218B (external)". Those ports only work if
u-boot initialize it (for example, loading initramfs image using one
of those ports). A patch to PHY detection is needed for full support.
The firmware recovery using U-Boot is broken for all DGS-1210 tested
devices as pressing RESET does not trigger it (only if pressed from a
running stock image)
Installation using OEM upgrade
------------------------------
1. Make sure you are running OEM firmware in image2 slot (logged as admin):
- > config firmware image_id 2 boot_up
- > reboot
2. Install squashfs-factory_image1.bin to image1 using (logged as admin):
- > download firmware_fromTFTP <tftpserver> factory_image1.bin
- > config firmware image_id 1 boot_up
- > reboot
Installation using serial interface
-----------------------------------
1. Press Escape key during `Hit Esc key to stop autoboot` prompt
2. Press CTRL+C keys to get into real U-Boot prompt
3. Init network with `rtk network on` command
4. Load image with `tftpboot 0x8f000000 openwrt-realtek-rtl839x-d-link_dgs-1210-52-initramfs-kernel.bin` command
5. Boot the image with `bootm` command
Once booted the initramfs, install the squashfs-sysupgrade.bin as a
normal OpenWrt system.
Dual-boot with stock firmware using writable u-boot-env
-------------------------------------------------------
From stock to OpenWrt / boot image 1 (CLI as admin):
- > config firmware image_id 1 boot_up
- > reboot
From OpenWrt to stock / boot image 2: (shell as root)
- # fw_setenv bootcmd 'run addargs ; bootm 0xb4e80000'
- # fw_setenv image '/dev/mtdblock7'
- # reboot
Debrick using serial interface
------------------------------
1. Press Escape key during `Hit Esc key to stop autoboot` prompt
2. In a Windows PC, run 'D-Link Network Assistant v2.0.2.4'. It should
detect the switch
3. Flash the firmware.
Back to stock firmware using dual-boot
--------------------------------------
If you have serial interface, you can change u-boot env vars
interrupting the boot process. If not but you are running OpenWrt, you
can dual-boot (as mentioned eariler) and skip to step 4:
1. Press Escape key during `Hit Esc key to stop autoboot` prompt
2. Press CTRL+C keys to get into real U-Boot prompt
3. Boot the image 2:
- set image /dev/mtdblock7; run addargs; bootm 0xb4e80000
4. Once booted, log as admin and change the boot image to 2
- > config firmware image_id 2 boot_up
- > reboot
5. After the boot, flash image1 with the vendor image
Back to stock firmware using DNA
--------------------------------
1. From an OpenWrt:
- # fw_setenv bootstop on
- # reboot
2. In a Windows PC, run 'D-Link Network Assistant v2.0.2.4'. It should
detect the switch
3. Flash the firmware.
It has been developed and tested on device with F3 revision.
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de> Signed-off-by: Luiz Angelo Daros de Luca <luizluca@gmail.com>
The D-Link DGS device tree was reorganized to better reflect the common
DT parts. The common include is named SOC specific (838X) and it seemed
like a good choice to add another common include in the future for the
RTL839X devices. From the current point of view this option is not really
needed.
1. The common part only includes data that matches RTL839X devices too.
2. The Panasonic DT structure avoids including the basic DTSI inside the
common DTSI.
Taking simplicity of the Panasonic include logic and in perparation to
provide DGS-1210-52 support it makes sense to harmonize this.
- rename common include to reflect its content
- move the link to the root DTSI directly to the device specific DTS
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Sungbo Eo [Sat, 17 Sep 2022 18:06:18 +0000 (03:06 +0900)]
ramips: 5.15: add disabled PINCTRL_AW9523
Otherwise kernel 5.15 will fail to build on subtargets except for mt7621
that has enabled the config.
The disabled PINCTRL_AW9523 config disappears after a refresh, it needs
to be added back manually.
Fixes: 675cf7557829 ("ramips: add config-5.15 for mt7620 subtarget") Fixes: 001176994a58 ("ramips: add config-5.15 for mt76x8 subtarget") Fixes: b9d9f33c330e ("ramips: add config-5.15 for rt288x subtarget") Fixes: 0164dc0c2563 ("ramips: add config-5.15 for rt305x subtarget") Fixes: ef59da866982 ("ramips: add config-5.15 for rt3883 subtarget") Signed-off-by: Sungbo Eo <mans0n@gorani.run>
Sungbo Eo [Sat, 17 Sep 2022 16:07:32 +0000 (01:07 +0900)]
ath79: specify factory.bin recipe for ASUS RP-AC51
Currently factory.bin image recipe of ASUS RP-AC51 is not specified
explicitly and is thus set to the leaked one from the device recipe
right above, i.e. ASUS PL-AC56. Fix it to avoid potential breakage.
Fixes: 416d4483e878 ("ath79: add support for ASUS RP-AC51") Signed-off-by: Sungbo Eo <mans0n@gorani.run>
Daniel Golle [Fri, 16 Sep 2022 23:33:30 +0000 (00:33 +0100)]
mac80211: clean and submit a bunch of rt2x00 patches
Clean and submit patches, mostly related to MT7620 to linux-wireless
mailing list:
https://patchwork.kernel.org/project/linux-wireless/list/?series=677770
Replace local patches with now submitted versions.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
The 'fxload' tool contained in the examples provided with libusb is
actually useful and turns out to be the only way to load firmware into
some rather ancient EZ-USB microcontrollers made by Cypress (formerly
Anchor Chips).
The original 'fxload' tool from hotplug-linux has been abandonned long
ago and requires usbfs to be mounted in /proc/bus/usb/ (like it was in
Linux 2.4...).
Hence the best option is to package the modern 'fxload' from the libusb
examples which (unsurprisingly) uses libusb and works on modern
systems.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Felix Fietkau [Fri, 16 Sep 2022 19:11:05 +0000 (21:11 +0200)]
unetd: update to the latest version
21360a1b1ce6 cli: fix typo abfebece0af1 wg-linux: ship a copy of linux/wireguard.h 1cbb1a543cb3 pex: reduce unnecessary ping traffic 0c2f39e52d5d pex: remove pex event debug spam dcf1362c2104 pex: add support for sending/receiving global PEX messages via unix socket df5f70b8858c ubus: notify on network updates e58a56697131 add DHT discovery service be175767bc67 pex: keep active pex hosts after the specified timeout 543e4a3d2ed7 pex: move rx header check to callback function 395659b9c415 pex: move raw ip send code to sendto_rawudp() in utils.c dda15ea8b3b2 pex: add utility function to get the sockets based on type / address family e88f2cd4d3f0 utils: add support for passings address family to network_get_endpoint() 639cdcdf6eda pex: add support for figuring out the external data port via STUN servers 9144339ebe1f pex: improve handling of a longer list of PEX hosts 38212218ecdd unet-cli: add DHT support 0d37ca75434d pex: automatically create host entries from incoming endpoint port notifications 035fcc56ef60 host: keep multiple endpoint candidates, one for each type a089e8ae7504 pex: avoid sending a query to a host more than once every 15 seconds
libwolfssl-cpu-crypto is a variant of libwolfssl with support for
cryptographic CPU instructions on x86_64 and aarch64.
On aarch64, wolfSSL does not perform run-time detection, so the library
will crash when the AES functions are called. A preinst script attempts
to check for support by querying /proc/cpuinfo, if installed in a
running system. When building an image, the script will check the
DISTRIB_TARGET value in /etc/openwrt_release, and will abort
installation if target is bcm27xx.
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
cc4eb79 ubus: support obtaining numeric error code 01c412c ubus: add toplevel constants for ubus status codes 8e240fa ubus: allow object method call handlers to return a numeric status code 5cdddd3 lib: add limit support to split() and replace() 0ba9c3e fs: add optional third permission argument to fs.open() c1f7b3b lib: remove fixed capture group limit in match() and regex replace()
Daniel Golle [Wed, 14 Sep 2022 19:23:17 +0000 (20:23 +0100)]
mediatek: build USB XHCI support as module
Instead of always including the XHCI driver in the kernel on all
MediaTek boards, selectively include the kernel module only on boards
which actually make use of USB functionality.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Olliver Schinagl [Wed, 31 Aug 2022 15:20:52 +0000 (17:20 +0200)]
realtek: rtl838x: Fix ethernet polling timeout on probe
Due to an oversight we accidentally inverted the timeout check. This
patch corrects this.
Fixes: 9cec4a0ea45b ("realtek: Use built-in functionality for timeout loop") Signed-off-by: Olliver Schinagl <oliver@schinagl.nl>
[ wrap poll_timeout line to 80 char ] Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Olliver Schinagl [Wed, 31 Aug 2022 15:20:52 +0000 (17:20 +0200)]
realtek: Use built-in functionality for timeout loop
In commit 81e3017609be ("realtek: clean up rtl838x MDIO busy wait loop")
a hand-crafted loop was created, that nearly exactly replicate the
iopoll's `read_poll_timeout` functionality.
Ansuel Smith [Tue, 6 Jul 2021 01:28:55 +0000 (03:28 +0200)]
scripts: add clean of build dir to dl_cleanup script
Improve dl_cleanup by adding an option to also clean the build directory
related to the downloaded package.
The script will check every directory in build_dir/ and check if any
old package is present there. If outdated package are found, the old
one are cleared leaving only the last one.
Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com>
Kabuli Chana [Sun, 7 Aug 2022 23:05:51 +0000 (17:05 -0600)]
kernel: netsupport: Add FQ-PIE as an optional sched kmod and extract PIE
add Flow Queuing with Proportional Integral controller Enhanced (FQ-PIE) as an
optional kmod in network support and extract sched-pie from kmod-sched to
allow dependency on just kmod-sched-pie (PIE).
Piotr Dymacz [Mon, 20 Jun 2022 11:13:30 +0000 (13:13 +0200)]
CI: include automatic Pull Request Labeler
This adds GitHub CI action which makes use of 'Labeler', allowing
automatic labeling of new PRs, based on the modified files paths.
Below labels are supported and more can be added later:
- 'target/*'
- 'target/imagebuilder'
- 'kernel'
- 'core packages'
- 'build/scripts/tools'
- 'toolchain'
- 'GitHub/CI'
For more information:
https://github.com/marketplace/actions/labeler
Add support for the TP-Link SG2210P switch. This is an RTL8380 based
switch with eight RJ-45 ports with 802.3af PoE, and two SFP ports.
This device shares the same board with the SG2008P and SG2008. To
model this, declare all the capabilities in the sg2xxx dtsi, and
disable unpopulated on the lower end models.
Specifications:
---------------
- SoC: Realtek RTL8380M
- Flash: 32 MiB SPI flash (Vendor varies)
- RAM: 256 MiB (Vendor varies)
- Ethernet: 8x 10/100/1000 Mbps with PoE (all ports)
2x SFP ports
- Buttons: 1x "Reset" button on front panel
- Power: 53.5V DC barrel jack
- UART: 1x serial header, unpopulated
- PoE: 2x TI TPS23861 I2C PoE controller
Works:
------
- (8) RJ-45 ethernet ports
- (2) SFP ports (with caveats)
- Switch functions
- System LED
Not yet enabled:
----------------
- Power-over-Ethernet (driver works, but doesn't enable "auto" mode)
- PoE LEDs
Enabling SFP ports:
-------------------
The SFP port control lines are hardwired, except for tx-disable. These
lines are controller by the RTL8231 in shift register mode. There is
no driver support for this yet.
However, to enable the lasers on SFP1 and SFP2 respectively:
Install via serial console/tftp:
--------------------------------
The footprints R27 (0201) and R28 (0402) are not populated. To enable
serial console, 50 ohm resistors should be soldered -- any value from
0 ohm to 50 ohm will work. R27 can be replaced by a solder bridge.
The u-boot firmware drops to a TP-Link specific "BOOTUTIL" shell at
38400 baud. There is no known way to exit out of this shell, and no
way to do anything useful.
Ideally, one would trick the bootloader into flashing the sysupgrade
image first. However, if the image exceeds 6MiB in size, it will not
work. The sysupgrade image can also be flashed. To install OpenWrt:
Prepare a tftp server with:
1. server address: 192.168.0.146
2. the image as: "uImage.img"
Power on device, and stop boot by pressing any key.
Once the shell is active:
1. Ground out the CLK (pin 16) of the ROM (U7)
2. Select option "3. Start"
3. Bootloader notes that "The kernel has been damaged!"
4. Release CLK as sson as bootloader thinks image is corrupted.
5. Bootloader enters automatic recovery -- details printed on console
6. Watch as the bootloader flashes and boots OpenWrt.
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
[OpenWrt capitalisation in commit message] Signed-off-by: Sander Vanheule <sander@svanheule.net>
realtek: rtl8380-tl-sg2xxx: use a single "firmware" partition
The "firmware" partition was assembled from two contiguous partitions.
This complexity is unnecessary. Instead of using mtd-concat over
"sys" and "usrimg1", simply declare the "firmware" partition to cover
the flash space instead.
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
realtek: tl-sg2xxx: read MAC address from nvmem-cells
The TP-Link RTL83xx based switches have their MAC address programmed
in the "para" partition. While in theory, the format of this partition
is dynamic, in practice, the MAC address appears to be located at a
consistent address. Thus, use nvmem-cells to read this MAC address.
The main MAC is required for deriving the MAC address of the switch
ports. Instead of reading it via mtd_get_mac_binary(), alias the
ethernet0 node as the label-mac-device, and use get_mac_label().
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Although PHY nodes are labeled, the port nodes were not. Labeling of
ports is useful for 'status = "disabled"' ports, which is supported
since commit 9a7f17e11f5d ("realtek: ignore disabled switch ports")
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
The TP-Link TL-SG2008, TL-SG2008P, and TL-SG2210P use the same board.
The main difference is that some footprints are not populated in the
lower-end models. To model this with minimal duplication, move the
devicetree to a common dtsi, leaving out just the board name.
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
[remove port relabelling from commit message, already merged with commit 18a2b29aa1c9 ("realtek: tl-sg2008p: fix labeling of lan ports")] Signed-off-by: Sander Vanheule <sander@svanheule.net>
Thats all happening due to '' was passed as an argument, which got later
expanded to http://.
In the context of a list constructor '' is not nothing, it is an empty
string element. So fix it by using () as it will yield "nothing" and
thus not introduce an empty string element.
toolchain: Include ./include/fortify for external musl toolchain
When building with an external toolcahin with musl also include
./include/fortify by default. This is also done when we build with the
internal toolchain using musl libc.
Without this extra include the fortify source feature is not working
when using an external musl toolchain. All binaries were compiled
without fortify source when an external musl toolchain was used. All
binaries release done by the OpenWrt project use the internal toolcahin
where fortify source is working.
toolchain: Select USE_SSTRIP with external musl toolchain
When we use the internal toolchain USE_SSTRIP will be selected by
default for musl libc and USE_STRIP when glibc is used. Do the same when
an external toolchain is used. USE_GLIBC will also be set for external
toolchain builds based on the EXTERNAL_TOOLCHAIN_LIBC_USE_GLIBC setting.
Daniel Golle [Thu, 28 Jul 2022 11:40:12 +0000 (13:40 +0200)]
mac80211: rt2x00: experimental improvements for MT7620 wifi
Serge Vasilugin reports:
To improve mt7620 built-in wifi performance some changes:
1. Correct BW20/BW40 switching (see comments with mark (1))
2. Correct TX_SW_CFG1 MAC reg from v3 of vendor driver see
https://gitlab.com/dm38/padavan-ng/-/blob/master/trunk/proprietary/rt_wifi/rtpci/3.0.X.X/mt76x2/chips/rt6352.c#L531
3. Set bbp66 for all chains.
4. US_CYC_CNT init based on Programming guide, default value was 33 (pci),
set chipset bus clock with fallback to cpu clock/3.
5. Don't overwrite default values for mt7620.
6. Correct some typos.
7. Add support for external LNA:
a) RF and BBP regs never be corrected for this mode
b) eLNA is driven the same way as ePA with mt7620's pin PA
but vendor driver explicitly pin PA to gpio mode (for forrect calibration?)
so I'm not sure that request for pa_pin in dts-file will be enough
First 5 changes (really 2) improve performance for boards w/o eLNA/ePA.
Changes 7 add support for eLNA
Configuration w/o eLAN/ePA and with eLNA show results
tx/rx (from router point of view) for each stream:
35-40/30-35 Mbps for HT20
65-70/60-65 Mbps for HT40
Yes. Max results for 2T2R client is 140-145/135-140
with peaks 160/150, It correspond to mediatek driver results.
Boards with ePA untested.
Reported-by: Serge Vasilugin <vasilugin@yandex.ru> Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Flash instruction:
Configure PC with static IP 192.168.99.8/24 and start TFTP server.
Rename "openwrt-ramips-rt305x-zyxel_keenetic-4g-b-squashfs-sysupgrade.bin"
to "rt305x_firmware.bin" and place it in TFTP server directory.
Connect PC with one of LAN ports, press the reset button, power up
the router and keep button pressed until power LED start blinking.
Router will download file from TFTP server, write it to flash and reboot.
Signed-off-by: Sergei Burakov <senior.anonymous@mail.ru>
Daniel Golle [Mon, 12 Sep 2022 13:10:26 +0000 (14:10 +0100)]
mediatek: unset CONFIG_CMDLINE_OVERRIDE for all targets but mt7629
The newly introduced config symbol CONFIG_CMDLINE_OVERRIDE is only set
for mt7629 for now which breaks automated build on all other mediatek
subtargets. Make sure the symbol is configured as 'is not set' for all
remaining subtargets.
Fixes: c27279dc26 ("mediatek: add support for ipTIME A6004MX Add basic support for ipTIME A6004MX.") Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Installation:
- Flash recovery image with TFTP recovery
Revert to stock firmware:
- Flash stock firmware with TFTP recovery
TFTP Recovery method:
1. Unplug the router
2. Hold the reset button and plug in
3. Release when the power LED stops flashing and go off
4. Set your computer IP address manually to 192.168.0.x / 255.255.255.0
5. Flash image with TFTP client to 192.168.0.1
Yoonji Park [Sun, 5 Jun 2022 18:04:18 +0000 (18:04 +0000)]
kernel: add support for mtdsplit-fit offset
Support devices that has vendor custom header before FIT image.
Some devices has vendor custom header before FIT image. In this case mtd-
split can not find FIT image and it results in rootfs mount failure.
Please refer iptime,a6004mx device for further examples.
Daniel Golle [Mon, 12 Sep 2022 00:13:59 +0000 (01:13 +0100)]
xdp-tools: don't rely on host bpf headers
xdp-tools build currently breaks on build hosts which do not have
libbpf headers installed because the build system wrongly tries to
use the host's include path.
Properly pass path to libbpf headers to xdp-tools build system to
fix build e.g. on the buildbots.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
MT7915 requires an additional antenna for background radar scanning.
Disable this feature in the following devices that do not have a
separate DFS antenna:
linksys,e8450
ruijie,rg-ew3200gx-pro
xiaomi,redmi-router-ax6s
Background radar detection is not supported on devices that
using MT7905, so disable this feature in the following devices:
asus,rt-ax53u
jcg,q20
tplink,eap615-wall-v1
xiaomi,mi-router-cr6606
xiaomi,mi-router-cr6608
xiaomi,mi-router-cr6609
yuncore,ax820
Devices with MT7915 lacking a DFS antenna also do not support
background DFS:
totolink,x5000r
cudy,x6
Flash instruction:
Upload generated factory firmware on vendor's web interface.
This device is very similar to the TL-WR841N v8, only two LED GPIOs are
different.
Buttons configuration is similar to TL-WR842ND v2 but both buttons are
active low.
Nick French [Sat, 13 Aug 2022 17:09:50 +0000 (12:09 -0500)]
ath79: add support for TP-Link Deco S4
Add support for TP-Link Deco S4 wifi router
The label refers to the device as S4R and the TP-Link firmware
site calls it the Deco S4 v2. (There does not appear to be a v1)
Hardware (and FCC id) are identical to the Deco M4R v2 but the
flash layout is ordered differently and the OEM firmware encrypts
some config parameters (including the label mac address) in flash
In order to set the encrypted mac address, the wlan's caldata
node is removed from the DTS so the mac can be decrypted with
the help of the uencrypt tool and patched into the wlan fw
via hotplug
UART serial access (115200N1) on board via solder pads:
RX = TP1 pad
TX = TP2 pad
GND = C201 (pad nearest board edge)
The device's bootloader and web gui will only accept images that
were signed using TP-Link's RSA key, however a memory safety bug
in the bootloader can be leveraged to install openwrt without
accessing the serial console. See developer forum S4 support page
for link to a "firmware" file that starts a tftp client, or you
may generate one on your own like this:
```
python - > deco_s4_faux_fw_tftp.bin <<EOF
import sys
from struct import pack
b += b"\xff"*(0x2A0-len(b)) + b"setenv serverip 192.168.0.2\x00"
b += b"\xff"*(0x2D0-len(b)) + b"setenv ipaddr 192.168.0.1\x00"
b += b"\xff"*(0x300-len(b)) + b"tftpboot 0x81000000 initramfs-kernel.bin\x00"
b += b"\xff"*(0x400-len(b)) + b"bootm 0x81000000\x00"
b += b"\xff"*(0x8000-len(b))
sys.stdout.buffer.write(b)
EOF
```
Installation:
1. Run tftp server on pc with static ip 192.168.0.2
2. Place openwrt "initramfs-kernel.bin" image in tftp root dir
3. Connect pc to router ethernet port1
4. While holding in reset button on bottom of router, power on router
5. From pc access router webgui at http://192.168.0.1
6. Upload deco_s4_faux_fw_tftp.bin
7. Router will load and execture in-memory openwrt
8. Switch pc back to dhcp or static 192.168.1.x
9. Flash openwrt sysupgrade image via luci/ssh at 192.168.1.1
Revert to stock:
Press and hold reset button while powering device to start the
bootloader's recovery mode, where stock firmware can be uploaded
via web gui at 192.168.0.1
Please note that one additional non-github commits is also needed:
firmware-utils: add tplink-safeloader support for Deco S4
Michael Pratt [Tue, 2 Nov 2021 17:41:41 +0000 (13:41 -0400)]
ath79: add support for Senao Watchguard AP100
FCC ID: U2M-CAP2100AG
WatchGuard AP100 is an indoor wireless access point with
1 Gb ethernet port, dual-band but single-radio wireless,
internal antenna plates, and 802.3at PoE+
this board is a Senao device:
the hardware is equivalent to EnGenius EAP300 v2
the software is modified Senao SDK which is based on openwrt and uboot
including image checksum verification at boot time,
and a failsafe image that boots if checksum fails
Label has no MAC
Only one Vendor MAC address in flash at art 0x0
eth0 ---- *:e5 art 0x0 -2
phy0 ---- *:e5 art 0x0 -2
**Installation:**
Method 1: OEM webpage
use OEM webpage for firmware upgrade to upload factory.bin
Method 2: root shell
It may be necessary to use a Watchguard router to flash the image to the AP
and / or to downgrade the software on the AP to access SSH
For some Watchguard devices, serial console over UART is disabled.
NOTE: DHCP is not enabled by default after flashing
**TFTP recovery:**
reset button has no function at boot time
only possible with modified uboot environment,
(see commit message for Watchguard AP300)
**Return to OEM:**
user should make backup of MTD partitions
and write the backups back to mtd devices
in order to revert to OEM reliably
It may be possible to use sysupgrade
with an OEM image as well...
(not tested)
**OEM upgrade info:**
The OEM upgrade script is at /etc/fwupgrade.sh
OKLI kernel loader is required because the OEM software
expects the kernel to be no greater than 1536k
and the factory.bin upgrade procedure would otherwise
overwrite part of the kernel when writing rootfs.
**Note on eth0 PLL-data:**
The default Ethernet Configuration register values will not work
because of the external AR8035 switch between
the SOC and the ethernet port.
For AR934x series, the PLL registers for eth0
can be see in the DTSI as 0x2c.
Therefore the PLL registers can be read from uboot
for each link speed after attempting tftpboot
or another network action using that link speed
with `md 0x1805002c 1`.
The clock delay required for RGMII can be applied
at the PHY side, using the at803x driver `phy-mode`.
Therefore the PLL registers for GMAC0
do not need the bits for delay on the MAC side.
This is possible due to fixes in at803x driver
since Linux 5.1 and 5.3
**Note on WatchGuard Magic string:**
The OEM upgrade script is a modified version of
the generic Senao sysupgrade script
which is used on EnGenius devices.
On WatchGuard boards produced by Senao,
images are verified using a md5sum checksum of
the upgrade image concatenated with a magic string.
this checksum is then appended to the end of the final image.
This variable does not apply to all the senao devices
so set to null string as default
Tested-by: Steve Wheeler <stephenw10@gmail.com> Signed-off-by: Michael Pratt <mcpratt@pm.me>
Michael Pratt [Sat, 12 Jun 2021 19:08:18 +0000 (15:08 -0400)]
ath79: add support for Senao WatchGuard AP200
FCC ID: U2M-CAP4200AG
WatchGuard AP200 is an indoor wireless access point with
1 Gb ethernet port, dual-band wireless,
internal antenna plates, and 802.3at PoE+
this board is a Senao device:
the hardware is equivalent to EnGenius EAP600
the software is modified Senao SDK which is based on openwrt and uboot
including image checksum verification at boot time,
and a failsafe image that boots if checksum fails
Label has no MAC
Only one Vendor MAC address in flash at art 0x0
eth0 ---- *:be art 0x0 -2
phy1 ---- *:bf art 0x0 -1
phy0 ---- *:be art 0x0 -2
**Installation:**
Method 1: OEM webpage
use OEM webpage for firmware upgrade to upload factory.bin
Method 2: root shell
It may be necessary to use a Watchguard router to flash the image to the AP
and / or to downgrade the software on the AP to access SSH
For some Watchguard devices, serial console over UART is disabled.
NOTE: DHCP is not enabled by default after flashing
**TFTP recovery:**
reset button has no function at boot time
only possible with modified uboot environment,
(see commit message for Watchguard AP300)
**Return to OEM:**
user should make backup of MTD partitions
and write the backups back to mtd devices
in order to revert to OEM reliably
It may be possible to use sysupgrade
with an OEM image as well...
(not tested)
**OEM upgrade info:**
The OEM upgrade script is at /etc/fwupgrade.sh
OKLI kernel loader is required because the OEM software
expects the kernel to be no greater than 1536k
and the factory.bin upgrade procedure would otherwise
overwrite part of the kernel when writing rootfs.
**Note on eth0 PLL-data:**
The default Ethernet Configuration register values will not work
because of the external AR8035 switch between
the SOC and the ethernet port.
For AR934x series, the PLL registers for eth0
can be see in the DTSI as 0x2c.
Therefore the PLL registers can be read from uboot
for each link speed after attempting tftpboot
or another network action using that link speed
with `md 0x1805002c 1`.
The clock delay required for RGMII can be applied
at the PHY side, using the at803x driver `phy-mode`.
Therefore the PLL registers for GMAC0
do not need the bits for delay on the MAC side.
This is possible due to fixes in at803x driver
since Linux 5.1 and 5.3
**Note on WatchGuard Magic string:**
The OEM upgrade script is a modified version of
the generic Senao sysupgrade script
which is used on EnGenius devices.
On WatchGuard boards produced by Senao,
images are verified using a md5sum checksum of
the upgrade image concatenated with a magic string.
this checksum is then appended to the end of the final image.
This variable does not apply to all the senao devices
so set to null string as default
Tested-by: Steve Wheeler <stephenw10@gmail.com> Tested-by: John Delaney <johnd@ankco.net> Signed-off-by: Michael Pratt <mcpratt@pm.me>
Michael Pratt [Thu, 11 Feb 2021 04:28:49 +0000 (23:28 -0500)]
ath79: add support for Senao WatchGuard AP300
FCC ID: Q6G-AP300
WatchGuard AP300 is an indoor wireless access point with
1 Gb ethernet port, dual-band wireless,
internal antenna plates, and 802.3at PoE+
this board is a Senao device:
the hardware is equivalent to EnGenius EAP1750
the software is modified Senao SDK which is based on openwrt and uboot
including image checksum verification at boot time,
and a failsafe image that boots if checksum fails
MAC address labeled as ETH
Only one Vendor MAC address in flash at art 0x0
eth0 ETH *:3c art 0x0
phy1 ---- *:3d ---
phy0 ---- *:3e ---
**Serial console access:**
For this board, its not certain whether UART is possible
it is likely that software is blocking console access
the RX line on the board for UART is shorted to ground by resistor R176
the resistors R175 and R176 are next to the UART RX pin at J10
however console output is garbage even after this fix
**Installation:**
Method 1: OEM webpage
use OEM webpage for firmware upgrade to upload factory.bin
Method 2: root shell access
downgrade XTM firewall to v2.0.0.1
downgrade AP300 firmware: v1.0.1
remove / unpair AP from controller
perform factory reset with reset button
connect ethernet to a computer
login to OEM webpage with default address / pass: wgwap
enable SSHD in OEM webpage settings
access root shell with SSH as user 'root'
modify uboot environment to automatically try TFTP at boot time
(see command below)
rename initramfs-kernel.bin to test.bin
load test.bin over TFTP (see TFTP recovery)
(optionally backup all mtdblocks to have flash backup)
perform a sysupgrade with sysupgrade.bin
NOTE: DHCP is not enabled by default after flashing
**TFTP recovery:**
server ip: 192.168.1.101
reset button seems to do nothing at boot time...
only possible with modified uboot environment,
running this command in the root shell:
then, before boot, the device will attempt TFTP from 192.168.1.101
looking for file 'test.bin'
to return uboot environment to normal:
fw_setenv bootcmd 'bootm 0x9f0a0000'
**Return to OEM:**
user should make backup of MTD partitions
and write the backups back to mtd devices
in order to revert to OEM
(see installation method 2)
It may be possible to use sysupgrade
with an OEM image as well...
(not tested)
**OEM upgrade info:**
The OEM upgrade script is at /etc/fwupgrade.sh
OKLI kernel loader is required because the OEM software
expects the kernel to be no greater than 1536k
and the factory.bin upgrade procedure would otherwise
overwrite part of the kernel when writing rootfs.
**Note on eth0 PLL-data:**
The default Ethernet Configuration register values will not work
because of the external AR8035 switch between
the SOC and the ethernet port.
For QCA955x series, the PLL registers for eth0 and eth1
can be see in the DTSI as 0x28 and 0x48 respectively.
Therefore the PLL registers can be read from uboot
for each link speed after attempting tftpboot
or another network action using that link speed
with `md 0x18050028 1` and `md 0x18050048 1`.
The clock delay required for RGMII can be applied
at the PHY side, using the at803x driver `phy-mode`.
Therefore the PLL registers for GMAC0
do not need the bits for delay on the MAC side.
This is possible due to fixes in at803x driver
since Linux 5.1 and 5.3
**Note on WatchGuard Magic string:**
The OEM upgrade script is a modified version of
the generic Senao sysupgrade script
which is used on EnGenius devices.
On WatchGuard boards produced by Senao,
images are verified using a md5sum checksum of
the upgrade image concatenated with a magic string.
this checksum is then appended to the end of the final image.
This variable does not apply to all the senao devices
so set to null string as default
Tested-by: Alessandro Kornowski <ak@wski.org> Tested-by: John Wagner <john@wagner.us.org> Signed-off-by: Michael Pratt <mcpratt@pm.me>
Michael Pratt [Sat, 12 Jun 2021 19:24:49 +0000 (15:24 -0400)]
ath79: rename an engenius DTSI to generic senao name
Other vendors can use this DTSI, for example, WatchGuard
there are likely several brands that use the same board design
because of outsourcing hardware from Senao.
For example, Watchguard AP300
has the same hardware as Engenius EAP600
so we use ar9344_engenius_exx600.dtsi for that
Daniel Golle [Sun, 11 Sep 2022 19:24:05 +0000 (20:24 +0100)]
mediatek: fix ledbar of UniFi 6 LR when running custom U-Boot
The RGB LED of the UniFi 6 LR v1 doesn't work when using the Openwrt-
built U-Boot. This is because the vendor loader resets the ledbar
controller while our U-Boot doesn't care.
Add reset-gpio so the ledbar driver in Linux will always reset the
ledbar controller.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
projects / thirdparty / openwrt.git / log
