The CONFIG_CLOCKSOURCE_WATCHDOG_MAX_SKEW_US is missing. Avoid
complaints and add it with its default 125.
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Copy files and config from 5.15 kernel version. Because of the big version jump
leave out the patches for now so we can treat them individually later on.
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
With commit b53202a8c3 ("realtek: switch to use generic MDIO accessor functions")
the phy access logic was enhanced. A quite big kernel patch was introduced that
allowed to make use of hardware-assisted page access like we have in the realtek
target. Basically it works the following way
- The enhanced bus intercepts page write accesses
- Currently selected pages are stored in internal vars.
- Finally only all-in-one-consistent bus/register accesses are issued
- It intercepted page changes and ensured that only a complete bus call
For the details see https://github.com/openwrt/openwrt/pull/16172
Switching over to newer kernels this patch is really hard to maintain. Its heavy
modifcations exist only in the realtek target. So it does not matter if we keep
it as an kernel modification or directly include it in our driver. To make it the
future brighter we drop this patch and take over its logic. Thus the kernel will
stay totally modification-free. What do we do?
1. Up to now the bus->priv structure directly pointed to ethernet->priv. Create an
explicit private structure rtl838x_bus_priv that not only holds the ethernet->priv
pointer but also space for some bus status tracking vars as well.
2. Wherever we use a reference to ethernet->priv directly replace that by an
additional indirection over the new rtl838x_bus_priv structure.
3. Up to now the phy flag PHY_HAS_REALTEK_PAGES identified that we can use the
alternative paged access from the patch. As this will be no longer available
remove it and provide read_page/write_page functions for each possible PHY.
These functions will be pretty standard as for other Realtek PHYs.
4. The existing mdio bus read/write function rely on the classic MII_ADDR_C45
flag - one interface for two access types. This mixup will be removed on the way
to kernel 6.6. In the future there will be two pairs of access functions. One for
classic access one for c45 style access. Rewrite our functions into 3 parts:
- a classic read/write function: ready for kernel 6.6
- a new c45 read/write function: ready for kernel 6.6
- a legacy read/write wrapper: for current 5.15 for the time being
When we switch to 6.6 we only need to remove the legacy wrappers and link the
new functions. Life can be so easy.
5. The classic read/write functions will incorporate the interception logic that
was originally in the patch.
6. The package convenience functions that were embedded in the patch get lost as
well. Rewrite them inside our phy driver.
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
[Minor checkpatch.pl cleanups]
Signed-off-by: Sander Vanheule <sander@svanheule.net>
There is no need to keep a version specific dts directory.
Rename the folder to its standard location.
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
The company Zyxel rebranded some years ago.
Currently the casing is according to the old branding even
for newer devices which already use the new branding.
This commit aligns the casing of Zyxel everywhere.
Signed-off-by: Goetz Goerisch <ggoerisch@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/15652
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
We should setup the registers for trapping LLDP packets to the CPU.
Currently, these packets are forwarded to all ports which is not desired
behaviour.
Signed-off-by: Kevin Jilissen <info@kevinjilissen.nl>
These registers control the handling of Link Layer Discovery Protocol
(LLDP) packets. This seems to be a typo in the naming.
Signed-off-by: Kevin Jilissen <info@kevinjilissen.nl>
According to the Realtek SDK code, the RTL8214FC, RTL8218B and RTL8218FB
all have the same chip ID 0x6276. Let's add a constant for it, as we're
using it in more than one location.
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
Respect the phy-is-integrated property on ethernet-phy nodes.
There are RTL8393M switches where the PHYs at address 48 and 49 are
provided by an external RTL8214FC. Hardcoding them to use the internal
SerDes makes it impossible to use the ports connected to such an
external PHY. Respect the phy-is-integrated property on ethernet-phy
nodes as a first step to support such ports.
The potential impact for this should be limited to RTL8393 based
switches, and looking at the commit messages and device tree files of
the supported switches based on this SoC, the SFP and/or combo ports are
either not working (D-Link DGS-1210-52, Netgear GS750E, TP-Link
SG2452P/T1600G-52PS), use PHYs at a different address (Panasonic
SwitchM48EG PN28480K), or already have the phy-is-integrated property
set on the PHYs at address 48 and 49.
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
Acked-by: Daniel Golle <daniel@makrotopia.org>
The function introduced in commit 7cbfe5654d is named
filter_port_list_reverse, not filter_port_list_reversed.
Fixes the following error on hpe,1920-8g-poe-65w and
hpe,1920-8g-poe-180w.
/bin/board_detect: /etc/board.d/02_network: line 84: filter_port_list_reversed: not found
Fixes: 7cbfe5654d ("realtek: move port filtering out of uci_set_poe()")
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
Acked-by: Sander Vanheule <sander@svanheule.net>
This device is very similar to the GS1900-24E switch (added in b515ad1),
except that the first 12 of 24 ethernet ports are capable of PoE and the
physical jacks are in the right order - unlike for the GS1900-24E, where
even and uneven ports are flipped (up <-> down on panel).
Zyxel version code for this device (-24EP) is: ABTO
Signed-off-by: Mirko Vogt <mirko-openwrt@nanl.de>
The Zyxel GS1900-8 v2 or Rev.B1 is a newer variant of the GS1900-8, but
otherwise similar to the other GS1900 switches.
Differences
------------
* Front Button labeled RESTORE
* NO Power Switch on rear
* Serial Header next to the barrel power connector
* Part Number ends 0102F
Signed-off-by: Goetz Goerisch <ggoerisch@gmail.com>
D-Link DGS-1210-16 hangs when rebooting and has no support for the reset
button.
Fix both by enabling the same GPIOs for reboot and the reset button as
already used for D-Link DGS-1210-20 and D-Link DGS-1210-28.
Signed-off-by: Richard Kunze <kunze@tivano.de>
Recent OEM firmware versions test the version number embedded in the uimage
"name" header field. The exact restricton is unknown, but "7.0.8.4" seems
to be the lowest number accepted on a GS110TPPv1 which already has that
version or higher.
A "9.9.9.9" version is accepted as valid by the GS110TPPv1 OEM firmware,
and considered both unique enough to identify an OpenWrt image and
moderately future proof against OEM version bumps.
This change is also boot tested on a GS108Tv3 with
"BOOT Loader Version 1.0.0.2 (2018-08-31 17:05:26 UTC)"
to verify that it doesn't break boot on older hardware.
Link: https://forum.openwrt.org/t/72510/58
Signed-off-by: Bjørn Mork <bjorn@mork.no>
PoE devices in the realtek target have the possibility to add PSE info
to the board description via 02_network. Make this available for all
targets, by moving the uci_set_poe() function to the globally available
uci-default.sh script.
Signed-off-by: Sander Vanheule <sander@svanheule.net>
uci_set_poe() now performs two duties: filtering the list of device
ports to exclude non-PoE ports, and generating the PoE related device
config.
Extract the port filtering to an external function, which is made a bit
more readable by the use of 'sort -V [-r] | uniq -u' to filter duplicate
entries out of a (reverse) version sorted list.
Signed-off-by: Sander Vanheule <sander@svanheule.net>
The ZyXEL XGS1250-12 has a chassis fan. The fan is positioned perfectly to
provide additional cooling to the Aquantia NBase-T phys. Testing has shown
that the phys can reach temperatures upwards of 72 degrees Celsius quite
easily at about 20 degrees Celsius ambient.
Support the chassis fan to give the phys a bit of extra cooling.
Signed-off-by: Tobias Schramm <tobias@t-sys.eu>
Initial conversion to new LED color/function format
and drop label format where possible. The same label
is composed at runtime.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Drop redundant label with new LED color/function format declared.
This was needed previously when the new format wasn't supported by
leds.sh functions script. Now that is supported this property
can be removed in favor of the new format.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
The rtl93xx SoC supports both 1000Base-X and 10GBase-CR on its SerDes
interfaces. Enable dynamic switching between mac-signaled modes to
support 1000Base-X and 10GBase-CR on the SFP port.
Signed-off-by: Tobias Schramm <tobias@t-sys.eu>
This patch adds support for 1000Base-X and 10GBase-CR directly on the
SerDes lanes of rtl93xx SoCs.
This fixes SFP/SFP+ support on devices like the XSG1250-12.
Signed-off-by: Tobias Schramm <tobias@t-sys.eu>
Commit daefc646e6 ("realtek: fix ZyXEL initramfs image generation")
fixed a shell expansion issue with zyxel-vers usage. Commit 045baca10b
("realtek: deduplicate GS1900 recipes") took care of this for the
rtl838x and rtl839x subtargets, but the single device officially
supported in rtl930x - the XGS1250-12 - was overlooked. This commit
updates the XGS1250-12 build recipe as well.
Signed-off-by: Stijn Segers <foss@volatilesystems.org>
Backport merged upstream patch that adds support for firmware loader
from NVMEM or attached filesystem for Aquantia PHYs.
Refresh all kernel patches affected by this change.
Also update the path for aquantia .ko that got moved to dedicated
directory upstream.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
[rmilecki: port to 5.15]
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Refresh HSGMII patch due to recent PHY backport that cause
compilation warning for case not handled in phy_interface_num_ports.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
The current dts file of dgs-1210-10p doesn't support link states
for the sfp ports (they are always up).
This patch tries to give better support for this and was run tested
on dgs-1210-10p.
It was heavily inspired from Paul Fertser, RaylynnKnight
and the author of dgs-1210-10mp-f.dts
https://forum.openwrt.org/t/dlink-dgs-1210-10p-with-glc-t-co-sfp/170928
Signed-off-by: Michel Thill <jmthill@gmail.com>
The register constants were duplicated in net/dsa/rtl83xx/debugfs.c and asm
mach-rtl838x/mach-rtl83xx.h. This commit removes this duplication.
Signed-off-by: Peter Körner <git@mazdermind.de>
According to https://svanheule.net/realtek/maple/register/led_sw_ctrl and also
drivers/net/dsa/rtl83xx/debugfs.c LED_SW_CTRL on the RTL838X should be 0xa00c
not 0x0128. Please note, that is is 0x0128 on the RTL8390/cypress SOC family.
Signed-off-by: Peter Körner <git@mazdermind.de>
the given code-format did not correctly express the condition and made the code
harder to read then necessary.
Signed-off-by: Peter Körner <git@mazdermind.de>
The GS110TUP v1 is a managed switch similar to the GS110TPP v1, but with
port 10 as SFP instead of RJ-45 and a total budget of 240 watts. Ports
1-4 support 60-watt 802.3bt PoE and ports 5-8 support 30-watt 802.3at.
The flash layout of the two switches are identical, and the U-Boot
configurations are the same except for having a different magic number,
so installation can be done via the same U-Boot method.
The following command will be needed to enable the port LEDs as per
https://forum.openwrt.org/t/72510/51 :
fw_setenv bootcmd "rtk network on; boota"
Additionally, port 9 (1000base-T from a separate QSGMII PHY) does not
function without this. Port 10 was not tested as no SFP module was
available.
Signed-off-by: Jacob Potter <jacob@j4cbo.com>
[rebase on merged flash layout]
Signed-off-by: Sander Vanheule <sander@svanheule.net>
Flash layouts for GS108Tv3, GS110TPPv1, GS308Tv1 and GS310TPv1 are
almost identical, except for the uimage header magic.
Move the flash layout to the common dtsi, and only place the magic value
in the device dts files.
Signed-off-by: Sander Vanheule <sander@svanheule.net>
Before, PVID is reset for all ports and goes out of bounds. Also, PVID
is later changed by dsa configuration by `ip link` and `bridge vlan`
commands, this does not change the CPU port PVID and CPU PVID stays 0.
It does not allow sending packets from OpenWrt to any connected devices
unless default configuration is changed
This change iterates up to and including cpu_port and sets default PVID
to 1. For lan* ports PVID can be configured with `ip link` and `bridge
vlan` commands
Acked-by: Simon Wunderlich <sw@simonwunderlich.de>
Signed-off-by: Harshal Gohel <hg@simonwunderlich.de>
Fix incorrect register value being set for VLAN_PORT_FWD
Before, the 0b1111 would be set for the register which means outgoing
packets would receive an extra tag, corresponding to the PVID of the
port.
On untagged ports, this meant outgoing packets with a single tag.
On tagged ports, this meant outgoing QinQ packets, where the inner tag
was either the PVID of the untagged ingress port, or the already
assigned original (single) tag.
Acked-by: Simon Wunderlich <sw@simonwunderlich.de>
Signed-off-by: Harshal Gohel <hg@simonwunderlich.de>
Without this, luci shows 10M full duplex when there is no link. So
explicitly set half duplex and unknown speed.
Acked-by: Simon Wunderlich <sw@simonwunderlich.de>
Signed-off-by: Harshal Gohel <hg@simonwunderlich.de>
Use led_setX to determine number of LEDs per port. Introduce macros to
calculate register value and shift for particular LED in a particular
set.
Problem with previous implementation is that it uses is10G status to
determine leds per port. However with usxgmii, driver sets 10g, 5g and
2.5g so even though there are only 2 leds per port it selects 4 leds per
port
This implementation relies on configured led_set node.
Acked-by: Simon Wunderlich <sw@simonwunderlich.de>
Signed-off-by: Harshal Gohel <hg@simonwunderlich.de>
