Hauke Mehrtens [Sun, 23 Feb 2020 17:57:45 +0000 (18:57 +0100)]
kernel: Use new symbol to deactivate MIPS FPU support
With kernel 5.4 the upstream kernel supports deactivating the FPU
support on MIPS. Use this new upstream feature instead of our older
patch which was removed when porting the kernel patches to kernel 5.4.
This way both options are set which should work for older kernel
versions and also new ones.
Hauke Mehrtens [Sun, 23 Feb 2020 15:46:01 +0000 (16:46 +0100)]
kernel: Remove nvmem hack patch from 5.4
The nvmem framework is now used in net/ethernet/eth.c and the nvmem
sysfs is split into a separate Kconfig option. More work would be needed
to adapt this patch for the broader use. The current patch compiles fine
on ath79, but it breaks the x86 target.
nvmem is also compiled into the kernel for most of our targets for
example ath79 anyway, so patching the kernel to remove it is now harder
and not the case on multiple targets anyway. Instead of making this work
on kernel 5.4 just remove this hack patch.
Hauke Mehrtens [Sun, 23 Feb 2020 15:41:58 +0000 (16:41 +0100)]
kernel: Remove chash.ko from kmod-drm-amdgpu
This module was added with kernel 4.15, but is was removed again with
kernel version 5.3. OpenWrt does not support specifying a kernel version
range so just break it with kernel 4.14 and only support recent kernel
versions.
Hauke Mehrtens [Sun, 23 Feb 2020 15:41:16 +0000 (16:41 +0100)]
kernel: Add snd-intel-nhlt.ko to kmod-sound-hda-intel
With kernel 5.4 kmod-sound-hda-intel also needs snd-intel-nhlt.ko, but
this kernel module is only build on x86, make the OpenWrt kmod depend on
TARGET_x86.
Hauke Mehrtens [Sun, 23 Feb 2020 15:28:21 +0000 (16:28 +0100)]
kernel: Make LIB_ARC4 selectable
This makes it possible to select CONFIG_CRYPTO_LIB_ARC4 directly. We
need this to be able to compile this into the kernel and make use of it
from mac80211 backports.
Hauke Mehrtens [Fri, 31 Jan 2020 12:32:03 +0000 (13:32 +0100)]
kernel: Add crypto libraries to modules
In kernel 5.3 and 5.4 some crypto modules were split into two modules,
one implementing the crypto algorithm and the other integrating it
into the Linux crypto framework.
This adds the new xfrm4_mode_beet, xfrm4_mode_transport,
xfrm4_mode_tunnel and their IPv6 versions on kernel 5.4. These modules
were newly added in kernel 5.2.
Some bigger changes were done to this feature and we did not port this patch yet:
* hack-5.4/207-disable-modorder.patch
This depends on BOOTMEM which was removed from the kernel, this needs some bigger changes:
* hack-5.4/930-crashlog.patch
A different version of the FPU disable patch was merged upstream, OpenWrt needs some adaptations.
* pending-5.4/304-mips_disable_fpu.patch
- no crashlog support yet as a required file got deleted upstream
- Removed patch below, which is now seen as a recursive dependency [1]
- Removed patch below due to build error [2]
- fix still required to avoid identical function def [3]
- Fixes included from Blocktrron
- Fixes included from Chunkeey
- Fix included from nbd regarding "dst leak in Flow Offload"
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com> Signed-off-by: David Bauer <mail@david-bauer.net> Signed-off-by: Christian Lamparter <chunkeey@gmail.com> Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com> Signed-off-by: Robert Marko <robimarko@gmail.com> Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
ath79: also reduce spi-max-frequency for buffalo_wzr-hp-ag300h
In accordance to ebc090e420d1 ("ath79: reduce spi-max-frequency to 50 MHz")
this also reduces the spi-max-frequency to 50 MHz for the last remaining
device with higher frequency in ath79. This will save us from having a
single special case that will require adjustment when the spi driver for
this device is changed in the future.
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de> Acked-by: Chuanhong Guo <gch981213@gmail.com>
The introduction of ebf0d8dadeca ("ath79: add new ar934x spi driver")
made the SPI memory unusable on devices with very high spi-max-frequency
(104 MHz).
Here's how the actual clock is calculated: (AHB_CLK/((CLOCK_DIVIDER+1)*2))
where AHB_CLK is a fixed clock (e.g. 200MHz on AR9331) and CLOCK_DIVIDER
is the parameter we can set. Highest clock according to this formula is
AHB_CLK/2 (100MHz, but that didn't work in device tests).
The next possible value is AHB_CLK/4 (50MHz). Speeds between 50 MHz and
100 MHz will be rounded down, so using values higher than 50 MHz does
not provide any benefit.
Consequently, this patch reduces spi-max-frequency for all devices with
values higher than 50 MHz to 50 MHz (effectively, this only affects
devices with 104 MHz before this patch).
Tested on GL.inet GL-AR150:
Boot fails with 104 MHz but is successful with both 50 MHz and 80 MHz
(fast-read), where the latter two yield identical read speeds.
Fixes: ebf0d8dadeca ("ath79: add new ar934x spi driver") Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
The introduction of ebf0d8dade (ath79: add new ar934x spi driver)
made the SPI memory unusable. Reducing the spi-max-frequency to
a smaller value makes it work again.
Tested on two MikroTik RouterBOARD wAP G-5HacT2HnD devices.
Signed-off-by: Roger Pueyo Centelles <roger.pueyo@guifi.net>
Xu Wang [Sat, 8 Feb 2020 23:04:11 +0000 (23:04 +0000)]
base-files: add all buildinfo with INCLUDE_CONFIG
CONFIG_INCLUDE_CONFIG option is helpful for being able to rebuild the
exact same firmware as you see on a live OpenWRT instance, but it's
crucially missing feeds information, so we can't rebuild the exact same
package versions. This commit fixes this by adding the remaining feeds
(and version) buildinfo files to the image.
Petr Štetiar [Thu, 20 Feb 2020 08:03:54 +0000 (09:03 +0100)]
ppp: backport security fixes
8d45443bb5c9 pppd: Ignore received EAP messages when not doing EAP 8d7970b8f3db pppd: Fix bounds check in EAP code 858976b1fc31 radius: Prevent buffer overflow in rc_mksid()
Signed-off-by: Petr Štetiar <ynezz@true.cz> Fixes: CVE-2020-8597 Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Jo-Philipp Wich [Wed, 26 Feb 2020 15:36:16 +0000 (16:36 +0100)]
Revert "ppp: backport security fixes"
This reverts commit 215598fd03899c19a9cd26266221269dd5ec8cee since it
didn't contain a reference to the CVE it addresses. The next commit
will re-add the commit including a CVE reference in its commit message.
ath79: add support for MikroTik RouterBOARD 922UAGS-5HPacD
This patch ports support for the MikroTik RouterBOARD 922UAGS-5HPacD
with a built-in 802.11ac High-Power radio (31dBm), which was already
available in the ar71xx target.
See https://mikrotik.com/product/RB922UAGS-5HPacD for more info.
Working:
- Board/system detection
- SPI and NAND storage
- PCIe
- USB type A host
- Wireless
- Ethernet
- LEDs (user, phy0)
- Reset button
- Sysupgrade to/from ar71xx
Not supported:
- RSSI LEDs
- SFP cage
Installation methods:
- Sysupgrade from ar71xx (it is advisable to use the -n option to
wipe any previous settings), or
- Boot the initramfs image via TFTP and then flash the sysupgrade
image using "sysupgrade -n"
Signed-off-by: Roger Pueyo Centelles <roger.pueyo@guifi.net>
Old MikroTik devices have the RLE-encoded radio calibration data
directly stored in the art (hard_config) partition, without LZO
compression nor any preceding ERD magic bytes. This commit adds
a fallback for these devices.
Tested on the ath79 target with a MikroTik SXT 5nD r2 (SXT Lite5),
only locally --not yet merged upstream--.
Signed-off-by: Roger Pueyo Centelles <roger.pueyo@guifi.net>
The calgary IOMMU was only used on high-end IBM systems in the early
x86_64 age. This is an unlikely OpenWrt target and in fact upstream
are looking to drop the driver entirely with the bonus that we no
longer see:
[ 0.000000] Calgary: detecting Calgary via BIOS EBDA area
[ 0.000000] Calgary: Unable to locate Rio Grande table in EBDA - bailing!
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
Piotr Dymacz [Fri, 31 Jan 2020 14:22:54 +0000 (15:22 +0100)]
base-files: diag: restore default trigger for 'boot' LED
For devices without a dedicated 'diag' LED, we use sometimes one of
other LEDs for indicating at least 'boot', 'failsafe' and 'upgrade'
stages. In some cases, at the same time these LEDs have defined default
triggers in DTS using 'linux,default-trigger' property. Current 'diag'
setup removes the trigger and turns off 'boot' LED after bootup.
One of the examples of such device is TP-Link TL-WR841N v14 (ramips)
which uses 'wlan' LED with defined 'linux,default-trigger' for 'diag':
This patch extends 'diag.sh' and 'leds.sh' scripts to make sure default
trigger defined in DTS is restored for 'diag' LED which isn't used for
indicating 'running' stage.
Acked-by: Adrian Schmutzler <freifunk@adrianschmutzler.de> Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
Piotr Dymacz [Mon, 20 Jan 2020 18:35:01 +0000 (19:35 +0100)]
hostapd: start hostapd/wpa_supplicant for all wiphy devices
c888e17e06 ("hostapd: manage instances via procd instead of pidfile")
added procd support for managing hostapd and wpa_supplicant daemons
but at the same time limited wiphy names to 'phy*'.
This brings back initial behaviour (introduced in 60fb4c92b6 ("hostapd:
add ubus reload") and makes procd manage daemons for any wiphy device
found in '/sys/class/ieee80211'.
CC: Felix Fietkau <nbd@nbd.name> CC: Daniel Golle <daniel@makrotopia.org> Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
Josef Schlehofer [Sat, 22 Feb 2020 22:03:37 +0000 (23:03 +0100)]
mbedtls: use correct SPDX License Identifier and add License file
License "GPL-2.0+" is deprecated License Identifier according to
SPDX License list [1]. The correct one is GPL-2.0-or-later.
While at it, also add the License file.
[1] https://spdx.org/licenses/
Signed-off-by: Josef Schlehofer <pepe.schlehofer@gmail.com>
Hauke Mehrtens [Sat, 9 Nov 2019 16:06:05 +0000 (17:06 +0100)]
mac80211: Allow IBSS mode and different beacon intervals
ath10k-ct supports the combination to select IBSS (ADHOC) mode and
different beacon intervals together. mac80211 does not like this
combination, but Ben says this is ok, so remove this check.
ath79: add missing reset-gpios for NanoStation Loco M (XW)
When porting support from ar71xx to ath79, the reset-gpios option was
missed. Due to a hardware bug, this would eventually leave the devices
with RX-deaf Ethernet PHY.
Signed-off-by: Roger Pueyo Centelles <roger.pueyo@guifi.net>
Sungbo Eo [Sun, 23 Feb 2020 15:14:22 +0000 (00:14 +0900)]
kirkwood: remove kmod-i2c-mv64xxx from DEVICE_PACKAGES
Commit 9a1f441ac81c ("kirkwood: enable SoC drivers in the kernel config")
enabled I2C_MV64XXX in the kernel config, and the subsequent commit 0d5ba94088ef
("orion: enable SoC drivers in the kernel config") removed kmod-i2c-mv64xxx
package entirely. As the feature is now kernel built-in and the package does not
exist anymore, we can safely remove kmod-i2c-mv64xxx from DEVICE_PACKAGES.
Sungbo Eo [Sun, 23 Feb 2020 15:14:22 +0000 (00:14 +0900)]
kirkwood: add kmod-hwmon-core to DEVICE_PACKAGES
kmod-hwmon-lm* will not get into images unless kmod-hwmon-core is added to
DEVICE_PACKAGES as well.
Signed-off-by: Sungbo Eo <mans0n@gorani.run>
[only address kmod-hwmon-core in this commit] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
Sungbo Eo [Sun, 23 Feb 2020 15:12:23 +0000 (00:12 +0900)]
kirkwood: fix device node name of Iomega ix2-200
The current device node name of ix2-200 is "iom_ix2_200", which results
in a SUPPORTED_DEVICES string "iom,ix2,200" that does not match the
compatible in DTS and the board name used in board.d.
Fix this by replacing the second underscore with a dash, following
vendor_model scheme.
Fixes: 27b2f0fc0fc5 ("kirkwood: add support for Iomega Storcenter ix2-200") Signed-off-by: Sungbo Eo <mans0n@gorani.run>
[commit title/message rephrase] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
David Bauer [Sun, 23 Feb 2020 11:50:30 +0000 (12:50 +0100)]
ath79: fix TP-Link TL-WA901ND v2 PHY stuck in reset
Some newer bootloaders for the TP-Link TL-WA901ND put the ethernet PHY
in reset before loading the kernel, thus the LAN interface is not
working in OpenWrt.
Clear the reset to restore ethernet functionality.
ramips: move compatible for Ubiquiti Edgerouter X to DTS file
So far, the compatible for the Ubiquiti Edgerouter X has been
defined in the DTSI file and inherited for the edgerouterx.dts,
but overwritten for the edgerouterx-sfp.dts. In contrast, the
model was stored in the DTS file in both cases.
To resolve this somewhat confusing situation, move the compatible
with the device name for edgerouterx to the DTS file as well.
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
Martin Schiller [Wed, 5 Feb 2020 07:12:54 +0000 (08:12 +0100)]
kernel: fix xt_connmark.h
Commit a1cfe0dcbb24 (kernel: connmark set-dscpmark follow upstreamimg
attempt") broke the usage of xt_connmark.h in user-space (e.g.
strongswan), because the BIT() macro is unknown there.
kirkwood: do not expose status LED to user config by default
So far, the state of status LEDs is set up in 01_leds for many devices
in kirkwood target. As those LEDs are also controlled by diag.sh,
exposing them to the user via uci config by default seems not helpful
and might even have confusing results for the user.
Thus, remove the ucidef_set_led_default setup for power/status LED, but
do not touch the rest where user control is actually a feature.
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
kirkwood: set default state for status LEDs in device tree
This adds the default-state = "on"; statement for the power or
primary status LED in DTS on kirkwood. This will ensure that this
LED will be lit up very early in the boot process (i.e. before
diag.sh is executed) and thus will provide an additional hint to the
user when problems arise during early boot process.
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
Pawel Dembicki [Tue, 4 Feb 2020 16:26:47 +0000 (17:26 +0100)]
kirkwood: use generic diag.sh
This commit removes the target-specific diag.sh script. This way, the
generic one is used for the target, which uses DT-aliases to specify the
LEDs used.
Generic diag.sh allow to use different LEDs to indicate different states.
Non-red status LEDs for indicating boot and a running system.
Where possible, the red or orange LEDs are used to indicate failsafe
mode and a running upgrade.
Compile-tested: all target devices.
Run-tested: CheckPoint L-50
Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
[remove unrelated cosmetic changes, rename some labels, add pogo_e02] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
Daniel Golle [Sun, 2 Feb 2020 16:09:50 +0000 (18:09 +0200)]
ath79: add support for Teltonika RUT955
Specification:
- 550/400/200 MHz (CPU/DDR/AHB)
- 128 MB of RAM (DDR2)
- 16 MB of FLASH (SPI NOR)
- 4x 10/100 Mbps Ethernet, with passive PoE support on LAN1
- 2T2R 2,4 GHz (AR9344)
- built-in 4G/3G module (example: Quectel EC-25EU)
- internal microSD slot (spi-mmc, buggy and disabled for now)
- RS232 on D-Sub9 port (Cypress ACM via USB, /dev/ttyACM0)
- RS422/RS485 (AR934x high speed UART, /dev/ttyATH1)
- analog 0-24V input (MCP3221)
- various digital inputs and outputs incl. a relay
- 11x LED (4 are driven by AR9344, 7 by 74HC595)
- 2x miniSIM slot (can be swapped via GPIO)
- 2x RP-SMA/F (Wi-Fi), 3x SMA/F (2x WWAN, GPS)
- 1x button (reset)
- DC jack for main power input (9-30 V)
- debugging UART available on PCB edge connector
Serial console (/dev/ttyS0) pinout:
- RX: pin1 (square) on top side of the main PCB (AR9344 is on top)
- TX: pin1 (square) on bottom side
Flash instruction:
Vendor firmware is based on OpenWrt CC release. Use the "factory" image
directly in GUI (make sure to uncheck "keep settings") or in U-Boot web
based recovery. To avoid any problems, make sure to first update vendor
firmware to latest version - "factory" image was successfully tested on
device running "RUT9XX_R_00.06.051" firmware and U-Boot "3.0.2".
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Daniel Golle [Wed, 5 Feb 2020 15:00:57 +0000 (17:00 +0200)]
serial: ar933x_uart: add rs485 support
Add support for RS485 tranceiver with transmit/receive switch hooked
to a RTS GPIO pin.
Use the 'rts-gpios' and 'rs485-rts-active-low' properties as described
in devicetree/bindings/serial/rs485.yaml.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Petr Štetiar [Tue, 11 Feb 2020 10:17:41 +0000 (11:17 +0100)]
ramips: gsw_mt7621: disable PORT 5 MAC RX/TX flow control by default
Looking at the current upstream driver implementation, it seems like the
TX/RX flow control is enabled only if the flow control pause option is
resolved from the device/link partner advertisements (or otherwise set).
On the other hand, our current in-tree driver force enables TX/RX
flow control by default, thus possibly leading to TX timeouts if the
other end sends pause frames (which are not properly handled?):
WARNING: CPU: 3 PID: 0 at net/sched/sch_generic.c:320 dev_watchdog+0x1ac/0x324
NETDEV WATCHDOG: eth0 (mtk_soc_eth): transmit queue 0 timed out
Disabling the flow control on PORT 5 MAC seems to fix this issues as the
pause frames are then filtered out. While at it, I'm removing the if
condition completely as suggested, since this code is run only on mt7621
SoC, so there is no need to check for the silicon revisions.
Ref: https://lists.openwrt.org/pipermail/openwrt-devel/2017-November/009882.html
Ref: https://forum.openwrt.org/t/mtk-soc-eth-watchdog-timeout-after-r11573/50000/12 Suggested-by: Felix Fietkau <nbd@nbd.name> Reported-by: Rosen Penev <rosenp@gmail.com> Signed-off-by: Petr Štetiar <ynezz@true.cz>
Felix Fietkau [Thu, 20 Feb 2020 14:06:14 +0000 (15:06 +0100)]
mt76: update to the latest version
f4415afce213 mt76: mt76u: loop over all possible rx queues in mt76u_rx_tasklet 5b9f949cb760 mt76: mt76u: fix a possible memory leak in mt76u_init fd892bc033fb mt76: mt76u: rely only on data buffer for usb control messagges
projects / thirdparty / openwrt.git / log
