x86: Fix xen serial console by removing conflicting PATA driver
The Xen serial console has been broken since the xen_domu subtarget
was merged in the generic x86 subtarget (commits 1d6879ee and 371b382a).
The reason for the broken serial console seems to be an IRQ conflict
between the serial console driver and the PATA_LEGACY driver:
[ 1.330125] genirq: Flags mismatch irq 8. 00000000 (hvc_console) vs. 00000000 (platform[pata_legacy.4])
[ 1.330134] hvc_open: request_irq failed with rc -16.
[ 1.330148] Warning: unable to open an initial console.
Just drop the PATA_LEGACY driver from the x86/generic and x86_64
subtargets, since this driver is marked experimental and only supports
very old ISA devices anyway. It is still included in the x86/legacy
subtarget where it rightfully belongs.
This was done by simply running `make kernel_menuconfig CONFIG_TARGET=subtarget`
and then saving without changing any option.
Most of the removed options can be explained because they are already
present in the target config or in the generic 4.9 config:
- PAE-related options, enabled by default on x86 by 961c0eac
- LZO-related options, enabled by default since 4.9
As far as I understand the build system, this shouldn't have any
user-visible impact, because the build system already merges the
various kernel configs during build.
treewide: use only board_name function to get name
Do not parse /tmp/sysinfo/board_name, /proc/cpuinfo or the device tree
compatible string directly. Always use the board_name function to get
the board name.
The admswconfig package still reads /proc/cpuinfo directly. The code
looks somehow broken and the whole adm5120 which uses this package
looks unmaintained. Leave it as it is for now.
For targets using the generic board detection and board specific
settings in diag.sh, the board name is still unset at the time the
set_state() provided by diag.sh is called by 10_indicate_preinit.
Change the execution order to ensure the boardname is populated before
required the first time. Do the target specific board detection as
early as possible, directly followed by the generic one to allow a
seamless switch to the generic function for populating /tmp/sysinfo/.
John Marrett [Thu, 13 Jul 2017 12:03:07 +0000 (08:03 -0400)]
ramips: fix GL-inet GL-MT300N-V2 WAN/LAN MAC address
Correct MAC address lookup to appropriate offset based on vendor
source.
Override the WAN MAC to use the same address as LAN. The switch driver
increments the base MAC address for the WAN vlan but the stock firmware
uses the same MAC address for all interfaces.
Based on vendor source commit
https://github.com/domino-team/lede-1701/commit/efb0518
Signed-off-by: John Marrett <johnf@zioncluster.ca>
473-fix-marvell-phy-initialization-issues.patch
-----------------------------------------------
Removed hunk 5 which got upstreamed
403-net-phy-avoid-setting-unsupported-EEE-advertisments.patch
404-net-phy-restart-phy-autonegotiation-after-EEE-advert.patch
--------------------------------------------------------------
Adapted these 2 RFC patches, merging the delta's from an upstream commit
(see below) which made it before these 2.
180-usb-xhci-add-support-for-performing-fake-doorbell.patch
-----------------------------------------------------------
- Moved fake_doorbell bitmask due to new item
Compile tested on: cns3xxx, imx6
Run tested on: cns3xxx, imx6
Alin Nastac [Fri, 16 Jun 2017 12:16:07 +0000 (14:16 +0200)]
netfilter: add iptables-mod-rpfilter package
Unlike /proc/sys/net/ipv4/conf/INTF/rp_filter flag, rule iptables -t raw
-I PREROUTING -m rpfilter --invert -j DROP prevents conntrack table to
become full when a packet flood with randomly selected source IP addresses
is received from the lan side.
base-files: automatically handle paths and symlinks for RAMFS_COPY_BIN
Depending on busybox applet selection, paths of basic utiilties may differ,
and may not work as symlinks to busybox. Simply using whatever binary is
found in PATH and detecting symlinks automatically is more robust and
easier to maintain.
The list of binaries is also slightly cleaned up and duplicates are
removed.
RAMFS_COPY_* are moved to platform.sh toplevel. The unneeded
linksys_preupgrade function is removed, and the nand_do_upgrade call is
moved to platform_do_upgrade.
dnsmasq: restore ability to include/exclude raw device names
Commit 5cd88f4 "dnsmasq: remove use of uci state for getting network ifname"
broke the ability to specify unmanaged network device names for inclusion
and exclusion in the uci configuration.
Restore support for raw device names by falling back to the input value
when "network_get_device" yields no result.
Piotr Dymacz [Wed, 5 Jul 2017 08:44:02 +0000 (10:44 +0200)]
ar71xx: support LED in TP-Link TL-WR902AC RJ45 socket
TP-Link TL-WR902AC has another one, tiny LED inside RJ45 socket which is
attached to GPIO15. Add support for it, use it as a "lan" LED and rename
old one to "internet". Also, fix uci option name for "lan" LED in system
configuration ("wan" -> "lan").
firmware-utils: mktplinkfw2: use static board struct for custom values
It seems simpler to store all custom (command line set) option values in
a struct identical to the predefined ones. It doesn't require:
1) Having so many global variables
2) Copying data from the predefined boards
Serial port is 115200 baud, on the 4 pin header; pins from the bottom to
top are GND, RX, TX.
To upgrade from the vendor firmware, upload factory.img
The DM200 bootloader supports flashing over TFTP; hold the reset button
while powering on the device, and wait for the power light to start
flashing green before releasing. The device is now listening on
192.168.0.1/24, and can be sent a factory.img or a netgear image with a
TFTP put.
Once the image is loaded, it will be written to the flash, and the
device will reboot; this will take a few minutes.
Thanks to Edward O'Callaghan and Baptiste Jonglez, who implemented their
own ports for this device and provided valuable feedback.
Most of the ubnt-erx definition can be reused; the package removals in
DEVICE_PACKAGES have become redundant after d17cb4a68a45 "ramips: purge
default packages on MT7621".
Hans Dedecker [Thu, 6 Jul 2017 17:16:58 +0000 (19:16 +0200)]
odhcpd: update to the latest version
f0d78e7 ndp: optimize check_addr6_updates code 94afe3b ndp: fix syslog tracing for netlink neigbor and address events 18df6cc treewide: rework logic to retrieve IPv6 interface addresses 803b83e router: use enum to specify order and index of iov struct 5dad295 treewide: rework code to get rid of fixed IPv6 address arrays 3e4c8ad config: rework code to get rid of IFNAMSIZ usage ab7813e treewide: use angle-brackets to include libubox header files
Felix Fietkau [Wed, 5 Jul 2017 18:29:39 +0000 (20:29 +0200)]
mt76: update to the latest version
Fixes mt7603 stablity and performance issues
af32615 mt7603: change auto rate control register initialization 01fb9ba mt7603: fix control/status retries count estimation cf4ba12 mt7603: avoid tx rate sampling using no retransmissions 32eab50 mt7603: set wtbl entry vif index c4e3dea mt7603: use the real vif index in txwi header for normal tx. e90a81a mt7603: fix channel width fall back in TXWI
After the update of GNU automake, the flex configure fails with
configure.ac:31: error: version mismatch. This is Automake 1.15.1,
configure.ac:31: but the definition used by this AM_INIT_AUTOMAKE
configure.ac:31: comes from Automake 1.15.
Apply the autoreconf host fixup to properly regenerate configure before
to avoid triggering the version mismatch. Also refresh the patch while
we're at it.
Daniel Golle [Wed, 5 Jul 2017 14:25:23 +0000 (16:25 +0200)]
bzip2: add symlink to binary
Other distributions incl. the OpenWrt ImageBuilder and SDK
expect to find the bzip2 executable in /bin.
Create a symlink at that location for compatibility.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Paul Wassi [Fri, 10 Mar 2017 18:10:29 +0000 (19:10 +0100)]
kirkwood: switch to kernel 4.9
Add patches-4.9, some of them (heavily) rewritten:
- ea4500 is upstream available, keep only LEDE changes in dts
- ea3500 is changed to match the structure of the upstream ea4500 dts
- nsa310s rewritten to include the common dtsi
- nsa325 is dropped, since already upstream
Signed-off-by: Paul Wassi <p.wassi@gmx.at> Signed-off-by: Alberto Bursi <alberto.bursi@outlook.it>
[refresh kernel config, add on100, use the switchdev based mv88e6171
driver for the linksys boards, keep lede specific rootfs/kernel
partition names for linksys boards, reorder patches] Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Wed, 28 Jun 2017 21:36:37 +0000 (23:36 +0200)]
lantiq: use img file extension for DGN3500 factory images
The Netgear UI in basic mode refuses the upgrade file if the the
fileextension is not img. The expert/advanced mode accepts any
fileextension. Use img to make it work in any case.
Piotr Dymacz [Tue, 4 Jul 2017 13:40:43 +0000 (15:40 +0200)]
firmware-utils: mktplinkfw: add option for endianness swap
This adds command line option in "mktplinkfw" tool for endianness swap
in kernel load address and entry point fields. As in "mktplinkfw2" tool,
we will need this for little-endian targets, like "ramips".
Piotr Dymacz [Sun, 2 Jul 2017 16:32:38 +0000 (18:32 +0200)]
ar71xx: add support for TP-Link TL-WR902AC v1
TP-Link TL-WR902AC v1 is a pocket-size, dual-band (AC750), successor of
TL-MR3020 (both devices use very similar enclosure, in same size). New
device is based on Qualcomm QCA9531 v2 + QCA9887. FCC ID: TE7WR902AC.
Specification:
- 650/391/216 MHz (CPU/DDR/AHB)
- 1x 10/100 Mbps Ethernet
- 1x USB 2.0 (GPIO-controlled power)
- 64 MB of RAM (DDR2)
- 8 MB of FLASH
- 2T2R 2.4 GHz (QCA9531)
- 1T1R 5 GHz (QCA9887)
- 5x LED (GPIO-controlled), 2x button, 1x 3-pos switch
- UART pads on PCB (TP1 -> TX, TP2 -> RX, TP3 -> GND, TP4 -> 3V3, jumper
resitors are missing on TX/RX lines)
- 1x micro USB (for power only)
Flash instructions:
Use "factory" image under vendor GUI.
Recovery instructions:
This device contains tftp recovery mode inside U-Boot. You can use it to
flash LEDE (use "factory" image) or vendor firmware.
1. Configure PC with static IP 192.168.0.66/24 and tftp server.
2. Rename "lede-ar71xx-generic-tl-wr902ac-v1-squashfs-factory.bin"
to "wr902acv1_un_tp_recovery.bin" and place it in tftp server dir.
3. Connect PC with LAN port, press the reset button, power up the router
and keep button pressed until WPS LED lights up.
4. Router will download file from server, write it to flash and reboot.
Root access over serial line in vendor firmware: root/sohoadmin.
We can use "tplink-fw" mtd splitter for TP-Link devices which use kernel
with TP-Link header embedded inside "safeloader" image type and thus get
rid of statically defined "kernel" and "rootfs" partitions in cmdline.
LEDE supports few devices using TP-Link firmware format (V2 or V3):
ArcherC20i, ArcherC50, ArcherMR200, TDW8970, TDW8980, TL-WR840N v4,
TL-WR841N v13 and VR200v
Testing mktplinkfw2 tool with official (vendor generated) firmware files
for above devices has shown an error when comparing calculated and
included MD5 sum, e.g.:
> mktplinkfw2 -i Archer_C20iv1_0.9.1_3.2_up_boot\(170221\)_2017-02-21_17.14.03.bin | grep -A 1 MD5Sum1
Header MD5Sum1 : 22 5a cb 92 10 d2 95 7b df 62 9a f8 62 17 37 10 (*ERROR*)
--> expected : ad 19 11 d1 78 98 a7 42 5f 2e 64 da 8a 34 ec cb
This problem has been verified to occur with:
Archer_C20iv1_0.9.1_3.2_up_boot(170221)_2017-02-21_17.14.03.bin
Archer MR200v1_0.9.1_1.1_up_boot_v004a.0 Build 160905 Rel.60037n.bin
TD-W8970v3_0.9.1_2.0_up_boot(160816)_2016-08-16_10.40.57.bin
TD-W8980v1_0.6.0_1.8_up_boot(150514)_2015-05-14_11.16.43.bin
Archer_VR200vv2_0.2.0_0.8.0_up_boot(161202)_2016-12-05_14.39.06.bin
For some images, e.g.:
Archer_C50v3_EU_0.9.1_0.3_up_boot[170417-rel52298].bin
TL-WR840Nv4_EU_0.9.1_4.16_up_boot[170421-rel70692].bin
TL-WR841Nv13_0.9.1_3.16_up_boot(161012).bin
mktplinkfw2 calculates zero MD5 so these has to be fixed separately:
> mktplinkfw2 -i TL-WR841Nv13_0.9.1_3.16_up_boot\(161012\).bin | grep -A 1 MD5Sum1
Header MD5Sum1 : 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 (*ERROR*)
--> expected : 6f 1d 9b 57 5d 42 14 6d bf a2 03 9d 46 7d 55 55
It's most likely that MD5 salt used in mktplinkfw2 has been always wrong
(and it's not a matter of e.g. a vendor change). Update it to fix MD5
calculation.
This has been also verified to calculate MD5 correctly for other (not
yet supported) devices, e.g.:
Archer_C3150v2_0.1.0_0.9.1_up_boot(160812)_2016-08-12_10.52.54.bin
Archer_C3200v1_0.9.1_0.1_up_boot(160704)_2016-07-04_15.48.28.bin
Mirko Parthey [Mon, 26 Jun 2017 16:25:45 +0000 (18:25 +0200)]
brcm47xx: use DMZ LED as status indicator
On the Linksys WRT54GSv1, the Power LED flickers in the "off" state.
Indicate status using the DMZ LED if available, since it has solid "on" and
"off" states.
This change was tested on the WRT54GSv1, but may also affect other brcm47xx
devices.
Mirko Parthey [Mon, 26 Jun 2017 16:25:46 +0000 (18:25 +0200)]
brcm47xx: resolve GPIO conflict for WRT54GSv1
On the Linksys WRT54GSv1, the adm6996 switch driver and the
gpio_button_hotplug module both claim GPIO 6, which is connected to the
Reset button. When the switch driver's request wins, the Reset button
cannot work. This makes it impossible to enter failsafe mode without a
serial console.
Stop requesting the "adm_rc" GPIO in the switch driver, since it is not
used anywhere.
Piotr Dymacz [Tue, 27 Jun 2017 20:49:21 +0000 (22:49 +0200)]
ar71xx: image: specify TPLINK_HWID for TP-Link RE450
TPLINK_HWID hasn't been specified for TP-Link RE450 since the begin.
As we don't want to break sysupgrade (all existing LEDE release images
for this board have TPLINK_HWID set to 0x0), set it explicitly to 0x0.
Piotr Dymacz [Wed, 21 Jun 2017 12:16:15 +0000 (14:16 +0200)]
ramips: add support for TP-Link TL-WR840N v4 and TL-WR841N v13
TP-Link TL-WR840N v4 and TL-WR841N v13 are simple N300 routers with
5-port FE switch and non-detachable antennas. Both are very similar
and are based on MediaTek MT7628NN (aka MT7628N) WiSoC.
The difference between these two models is in number of available
LEDs, buttons and power input switch.
This work is partially based on GitHub PR#974.
Specification:
- MT7628N/N (580 MHz)
- 64 MB of RAM (DDR2)
- 8 MB of FLASH
- 2T2R 2.4 GHz
- 5x 10/100 Mbps Ethernet
- 2x external, non-detachable antennas
- UART (J1) header on PCB (115200 8n1)
- TL-WR840N v4: 5x LED (GPIO-controlled), 1x button
- TL-WR841N v13: 8x LED (GPIO-controlled*), 2x button, power input
switch
* WAN LED in TL-WR841N v13 is a dual-color, dual-leads type which isn't
(fully) supported by gpio-leds driver. This type of LED requires both
GPIOs state change at the same time to select color or turn it off.
For now, we support/use only the green part of the LED.
Factory image notes:
These devices use version 3 of TP-Link header, fortunately without RSA
signature (at least in case of devices sold in Europe). The difference
lays in the requirement for a non-zero value in "Additional Hardware
Version" field. Ideally, it should match the value stored in vendor
firmware header on device ("0x4"/"0x13" for these devices) but it seems
that anything other than "0" is correct.
We are able to prepare factory firwmare file which is accepted and
(almost) correctly flashed from the vendor GUI. As it turned out, it
accepts files without U-Boot image with second header at the beginning
but due to some kind of bug in upgrade routine, flashed image gets
corrupted before it's written to flash.
Tests showed that the GUI upgrade routine copies value of "Additional
Hardware Version" from existing firmware into offset "0x2023c" in
provided file, _before_ storing it in flash. In case of vendor firmware
upgrade files (which all include U-Boot image and two headers), this
offset points to the matching field in kernel+rootfs firmware part
header. Unfortunately, in case of LEDE factory image file which contains
only one header, it points to the offset "0x2023c" in kernel image. This
leads to a corrupted kernel and ends up with a "soft-bricked" device.
The good news is that U-Boot in these devices contains well known tftp
recovery mode, which can be triggered with "reset" button. What's more,
in comparison to some of older MediaTek based TP-Link devices, this
recovery mode doesn't write whole file at offset "0x0" in flash, without
verifying provided file in advance. In case of recovery mode in these
devices, first "0x20000" bytes are always skipped and "0x7a0000" bytes
from rest of the file are stored in flash at offset "0x20000".
Flash instruction:
Until (if at all) TP-Link fixes described problem, the only way to flash
LEDE image in these devices is to use tftp recovery mode in U-Boot:
1. Configure PC with static IP 192.168.0.66/24 and tftp server.
2. Rename "lede-ramips-mt7628-tl-wr84...-squashfs-tftp-recovery.bin"
to "tp_recovery.bin" and place it in tftp server directory.
3. Connect PC with one of LAN ports, press the reset button, power up
the router and keep button pressed for around 6-7 seconds, until
device starts downloading the file.
4. Router will download file from server, write it to flash and reboot.
To access U-Boot CLI, keep pressed "4" key during boot.
Piotr Dymacz [Fri, 23 Jun 2017 21:07:10 +0000 (23:07 +0200)]
build: move mktplinkfw2 related commands to image-commands.mk
There are already two targets (lantiq, ramips) which use mktplinkfw2
tool for creating images. This de-duplicates code, introduces two new
build commands: tplink-v2-header, tplink-v2-image and makes use of
them in place of old, (sub)target specific ones.
Piotr Dymacz [Wed, 21 Jun 2017 12:15:55 +0000 (14:15 +0200)]
firmware-utils: mktplinkfw2: support additional hardware version
As it turned out, some of new MediaTek based TP-Link devices use value
from field at 0x3c offset in version 3 of TP-Link header to specify
"Additional Hardware Version".
Value from this field is validated during regular (GUI) firmware upgrade
on devices like TL-WR840N v4 or TL-WR841N v13. If it's zero (based on
some tests, it seems that firmware will accept anything != 0), errors
like below are printed on console and upgrade fails:
projects / thirdparty / openwrt.git / log
