When checking the outcome of the PHY autonegotiation status, at803x
currently returns false in case the SGMII side is not established.
Due to a hardware-bug, ag71xx needs to fixup the SoCs SGMII side, which
it can't as it is not aware of the link-establishment.
This commit allows to ignore the SGMII side autonegotiation status to
allow ag71xx to do the fixup work.
Signed-off-by: David Bauer <mail@david-bauer.net>
The QCA955X is affected by a hardware bug which causes link-loss of the
SGMII link between SoC and PHY. This happens on change of link-state or
speed.
It is not really known what causes this bug. It definitely occurs when
using a AR8033 Gigabit Ethernet PHY.
Qualcomm solves this Bug in a similar fashion. We need to apply the fix
on a per-device base via platform-data as performing the fixup work will
break connectivity in case the SGMII interface is connected to a Switch.
This bug was first proposed to be fixed by Sven Eckelmann in 2016.
https://patchwork.ozlabs.org/patch/604782/
Based-on-patch-by: Sven Eckelmann <sven.eckelmann@open-mesh.com>
Signed-off-by: David Bauer <mail@david-bauer.net>
This commit adds the ability to configure specific functions of the
at803x series ethernet-PHYs, which were previously configured
exclusively with the help of platform-data, via device-tree.
This is needed to fully support existing boards of the ar71xx platform.
Signed-off-by: David Bauer <mail@david-bauer.net>
Backport an upstream fix for a remotely exploitable TCP denial of service
flaw in Linux 4.9+.
The fixes are included in Linux 4.14.59 and later but did not yet end up in
version 4.9.118.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
The QCA9557 dtsi is currently missing pll-handle and pll-regs for both
eth0 and eth1, therefore PLL settings won't be applied. This commit
fixes this behavior.
Signed-off-by: David Bauer <mail@david-bauer.net>
While finalizing support for the U7623 with 512MB, I made an embarresing
error and configured 1GB RAM for the board. I also forgot to move memory
from the dtsi and to the dts. This commit takes care of my mistakes.
While I am confessing my mistakes, I also note that I made a mistake in
the commit message of the initial U7623 commit. It is the .bin-file, and
not the .gz file that shall be sent to the device via tftp.
v1->v2:
* Remove redundant memory node (thanks Jonas Gorski)
Signed-off-by: Kristian Evensen <kristian.evensen@gmail.com>
Including the tl-wdr3600 image build code just to overwrite most of it
doesn't make much sense and only makes it hard to read.
Furthermore, the tl-wdr4300 image will be marked as compatible with the
tl-wdr3600 this way.
Signed-off-by: Mathias Kresin <dev@kresin.me>
The variables are used in image build recipes and need to be marked as
per devices vars to be stored individual per image define. Otherwise
the last defined variable will be used for all boards.
Signed-off-by: Mathias Kresin <dev@kresin.me>
This patch did the following things:
1. Separate ath9k-leds out of gpio leds so that all other leds will work
before ath9k loded (e.g. during preinit/init stage).
2. Rename wps led to qss since that's how TP-Link mark it.
3. Rename LED prefix to tp-link because that dts is shared by many devices.
4. Rename to wr740n-v1 because v1 is the first and v2 just use the fw of v1.
(This will require a forced sysupgrade if you comes from
the previous wr740n v2 image.)
5. Remove SUPPORTED_DEVICES.
(tl-wr740n-v2 doesn't exist anywhere so it's useless.)
6. Add all WR741ND v1 clones found in ar71xx.
Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
Fix all issues found by the devicetree compiler like wrong address/size
cells as well as wrong/missing/superfluous unit addresses.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Use only the jedec,spi-nor compatible string. Everything else either
never worked or is only support to keep compatibility.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Use the same method for setting queue index pointers consistenly
throughout the source file.
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
This router is called Archer C7 and the tl was used to identify
TP-LINK. Since we have added tplink in dts/board name, the tl
prefix is useless now.
Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
Change lan and it's LED to eth0
It's broken since c7c807cb8c
where I changed the dts but forgot to change default configurations.
Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
1. Swap eth0/eth1
Both devices are using AR9331, the builtin switch on AR9331 is
connected to gmac1 and gmac1 is named as eth1 in ath79.
PS: gmac1 is eth0 and gmac0 is eth1 in ar71xx because of the
reversed initialization order.
2. Fix the incorrect compatible string in dts
Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
phy-handle is used to poll link status. They are useless when
we need fixed-link on these interfaces.
Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
Jonas Gorski commented on the previous patch:
|This is actually the wrong fix and papers over an issue in one of our
|local patches.
|
|We intentionally allow regmap to be built as a module, see
|
|/target/linux/generic/hack-4.14/259-regmap_dynamic.patch
|[...]
|[The regulator code] optionally supports regmap thanks to the stubs
|provided if regmap is disabled - which breaks if you compile regmap
|as a module.
In order to mitigate this issue, this patch reverts the previous patch
and replaces the existing IS_ENABLED(CONFIG_REGMAP) with
IS_REACHABLE(CONFIG_REGMAP). This solves this particular issue as the
regulator code will now automatically fallback to the regmap stubs in
case the kmod-regmap module is enabled, but nothing else sets
CONFIG_REGMAP=y.
Note: There's still a potential issue that this patch doesn't solve:
If someone ever wants to make a OpenWrt kernel package for a
regulator module that requires the REGMAP feature for a target that
doesn't set CONFIG_REGMAP=y but has CONFIG_REGULATOR=y, the resulting
kmod-regulator-xyz package will not work on the target.
Luckily, there aren't any in-tree OpenWrt kernel module packages for
regulators at the moment. On the bright side: regmap is a critical
part nowadays and all new and upcoming architectures require it by
default. This will likely only ever be a problem for legacy targets
and devices that cannot afford to enable REGMAP.
Cc: Jonas Gorski <jonas.gorski@gmail.com>
Cc: John Crispin <john@phrozen.org>
Fixes: d00913d121 ("kernel: modules: fix kmod-regmap")
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
To share mdio addr for IntPHY and ExtPHY,
as described in the documentation (MT7620_ProgrammingGuide.pdf).
(refer: http://download.villagetelco.org/hardware/MT7620/MT7620_ProgrammingGuide.pdf)
when port4 setup to work as gmac mode, dts like:
&gsw {
mediatek,port4 = "gmac";
};
we should set SYSCFG1.GE2_MODE==0x0 (RGMII).
but SYSCFG1.GE2_MODE may have been set to 3(RJ-45) by uboot/default
so we need to re-set it to 0x0
before this changes:
gsw: 4FE + 2GE may not work correctly and MDIO addr 4 cannot be used by ExtPHY
after this changes:
gsw: 4FE + 2GE works and MDIO addr 4 can be used by ExtPHY
Signed-off-by: Chen Minqiang <ptpt52@gmail.com>
When PHY's are defined on the MDIO bus in the DTS, gigabit support was
being masked out for no apparent reason, pegging all such ports to 10/100.
If gigabit support must be disabled for some reason, there should be a
"max-speed" property in the DTS.
Reported-by: James McKenzie <openwrt@madingley.org>
Signed-off-by: Daniel Gimpelevich <daniel@gimpelevich.san-francisco.ca.us>
Mediatek has a reference platform that pairs an MT7620A with an MT7530W,
where the latter responds on MDIO address 0x1f while both chips respond on
0x0 to 0x4. The driver special-cases this arrangement to make sure it's
talking to the right chip, but two different ways in two different places.
This patch consolidates the detection without the current requirement of
both tests to be separately satisfied in the DTS.
Signed-off-by: Daniel Gimpelevich <daniel@gimpelevich.san-francisco.ca.us>
This really simplifies debugging, if a package is not found or a feed is
not reachable, a proper stderr is printed. Currently it would only say
`_call_manifest` failed.
Signed-off-by: Paul Spooren <mail@aparcar.org>
Starting with kernel 4.4, the use of partitions as direct subnodes of the
mtd device is discouraged and only supported for backward compatiblity
reasons.
Signed-off-by: Alex Maclean <monkeh@monkeh.net>
Fix space vs. tabs issue and trainling whitespaces. Use C style
comments or drop the comments if they explain what is already to see in
the devicetree parameters.
Signed-off-by: Mathias Kresin <dev@kresin.me>
The hardware NAT node has the same reg/unit as the ethernet node. One
of them need to be a child of the other.
Make the hardware NAT node a child of the ethernet node since the it
"reference" the netdev in its properties.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Add the ranges property to the PCI bridges where missing. Add the unit
address to PCI bridge where missing.
Rework the complete rt3883 pci node. Drop the PCI unit nodes from the
dtsi. They are not used by any dts file and should be rather in the dts
than in the SoC dtsi. Express the PCI-PCI bridge in a clean devicetree
syntax. The ralink,pci-slot isn't used by any driver, drop it. Move the
pci interrupt controller out of the pci node. It doesn't share the same
reg and therefore should be an independent/SoC child node.
Move the pci related rt3883 pinctrl setting to the dtsi instead of
defining the very same for each rt3883 board.
If the device_type property is used for PCI units, the unit is treated
as pci bridge which it isn't. Drop it for PCI units.
Reference pci-bridges or the pci node defined in the dtsi instead of
recreating the whole node hierarchy. It allows to change the referenced
node in the dtsi without the need to touch all dts.
Fix the PCI(e) wireless unit addresses. All our PCI(e) wireless chips
are the first device on the bus. The unit address has to be the bus
address instead of the PCI vendor/device id.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Since commit c1e7738988f5 ("checks: add gpio binding properties check")
dtc treats any *-gpios and *-gpio property as phandle at least during
checks. The only whitelisted property is nr-gpio.
Use ralink,nr-gpio in favour of ralink,num-gpios to get rid of false
positive warnings.
Signed-off-by: Mathias Kresin <dev@kresin.me>
The cpu interrupt controller doesn't have a reg property, hence we
can't use a unit address in the node name.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Without this patch you will get an error "gpio-export probe deferral
not supported" when you try to export i2c expander gpio pins.
gpio-export is probed long before i2c-bus and i2c expander are created
and it doesn't retry it so none pins are exported.
Signed-off-by: René van Dorst <opensource@vdorst.com>
apply the change to all instances of the gpio exports patch
Signed-off-by: Mathias Kresin <dev@kresin.me>
Revert 290c54473e ("ath79: fix TP-Link Archer C7 v2 wlan1 MAC address")
which obviously aims to have a distinct MAC address per interface.
Unfortunally it doesn't match what is used by the stock firmware and we
shouldn'z use MAC Adresses not reserverd for/assigned to a particular
board.
The correct MAC adress increments for this board are:
wlan0 (5GHz) : -1
wlan1 (2.4GHz) : 0
eth1 (LAN) : 0
eth0 (WAN) : 1
Fixes: FS#408
Signed-off-by: Mathias Kresin <dev@kresin.me>
