There is currently a difference how upstream and downstream define
the switch in the dts. Downstream holds the switch as a member
node below a root switchcore parent. Upstream uses the switch as
the parent.
Upstream:
ethernet-switch@1b000000 {
mdio-controller@ca00 { };
ethernet { };
ethernet-ports { };
}
Downstream:
switchcore@1b000000 {
ethernet-switch {
ethernet-ports { };
};
mdio-controller@ca00 { };
ethernet { };
}
Align downstream to upstream and merge the ethernet-switch into
the parent node. For this to work adapt the port lookup in the MDIO
and PCS driver.
Remark! With this commit the boot process will give the spurious
error message "rtl838x_eth 1b000000.ethernet-switch:ethernet eth0:
Failed to create a device link to DSA switch 1b000000.ethernet-switch"
This comes from the fact that the switch is the parent of the ethernet
device. Thus a link back from ethernet device to the switch is no
longer possible. Testing shows that the error is just cosmetic.
Link: https://github.com/openwrt/openwrt/pull/23599
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
A SerDes can host multiple PCS links: QSGMII binds four ports to one
SerDes, USXGMII variants up to eight. Today pcs-handle references the
SerDes as a whole, with no way to express which link inside the SerDes
a port wants. The driver gets away with this because it carries its own
port->link bookkeeping and the link slot is implicit in DSA's port
iteration order -- functional, but the wiring information lives nowhere
in DT.
The upcoming fwnode_pcs migration moves PCS lookup to the generic
fwnode provider API, which disambiguates multiple instances per fwnode
via phandle cells. To make that landable as small, code-only commits,
the DT needs to carry the link index ahead of time.
Bump #pcs-cells from 0 to 1 on every SerDes node in the four SoC DTSIs
and append the link cell to every pcs-handle reference across boards
and the SWITCH_PORT_* macros. Cell values match the existing wiring:
0 for single-link SerDes (10GBase-R, SGMII, fiber, single-link
USXGMII), 0..3 per SerDes for QSGMII and USXGMII-QX, 0..7 for the
RTL9311 octal USXGMII layout.
No code reads the new cell yet -- of_parse_phandle_with_args() in the
PCS driver already cooperates with cells = 0 or 1, and the DSA glue
uses of_parse_phandle() which ignores cells entirely. The change is
runtime-neutral on its own; it exists so the follow-up code patches
can be a few lines each instead of dragging a bridge counter into the
driver to invent slot numbers DT could have provided directly.
Link: https://github.com/openwrt/openwrt/pull/23539
Signed-off-by: Jonas Jelonek <jelonek.jonas@gmail.com>
Previously, sds->num_of_links was incremented from rtpcs_create() as
each DSA port bound its phylink_pcs. The count therefore relied on a
temporal contract (DSA must finish enumerating before pcs_config runs)
and on rtpcs_create() being the single chokepoint for all consumers.
Replace this with a probe-time scan of pcs-handle references in the
live OF tree: for every available consumer node carrying a pcs-handle
property pointing at one of our SerDes subnodes, bump that SerDes'
num_of_links. After the scan, the count is final regardless of when
or whether DSA later calls in.
To allow of_parse_phandle_with_args() to walk the property correctly,
add #pcs-cells = <0> to every serdes@N node in the 838x/839x/930x/931x
.dtsi files. A future cell-bearing form remains possible without
touching the scan.
Over-references (DT pointing more consumers at one SerDes than the
hardware can carry) are clamped at RTPCS_MAX_LINKS_PER_SDS and warned
about, but do not fail probe — the correctly-wired ports on that
SerDes still come up, and only the surplus reference is dropped.
The bounds check and the bare ++ in rtpcs_create() become redundant
under the scan-driven count and are removed.
This decouples num_of_links from DSA call ordering and is a prereq
for migrating to fwnode_pcs providers, where rtpcs_create() goes away
as the centralised counter.
Link: https://github.com/openwrt/openwrt/pull/23484
Signed-off-by: Jonas Jelonek <jelonek.jonas@gmail.com>
The pinmux entry for disabling JTAG includes a comment which points to
which GPIOs are sacrificed for using JTAG. However, this comment so far
was only aware of GPIO6 and GPIO7. From RTL931X application notes and
datasheets we know which GPIOs are actually affected here.
Extend the comment to include GPIOs 3-5 too.
Signed-off-by: Jonas Jelonek <jelonek.jonas@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/22827
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
RTL93xx devices can no longer find the switch node in the DTS.
Commit 4c92254 ("relocate/retype switch node") refactored the
switch node definition to better align with upstream. Sadly
the redefinition for RTL93xx devices failed.
- RTL83xx: use "switch0: ethernet-switch"
- RTL93xx: use "switch0: switch@1b000000"
Follow up commit 8b969f7 ("drop realtek,smi-address property)
changed the dts lookup sequence for mdio initialization. On
RTL93xx devices it cannot find the switchnode via
of_get_child_by_name(dev->of_node->parent, "ethernet-switch")
Fix the switch node type for RTL93xx
Fixes: 8b969f7 ("drop realtek,smi-address property)
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/22557
Signed-off-by: Robert Marko <robimarko@gmail.com>
The switch node is currently located outside of the switchcore@1b000000
tree. This makes it hard to find when referencing from other nodes in
this tree. Make it a subnode of switchcore and "retype" it to
ethernet-switch like upstream does.
This is not perfectly aligned as upstream just mixes the switchcore and
the ethernet-switch node into one. But this will be future work for
downstream.
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/22236
Signed-off-by: Robert Marko <robert.marko@sartura.hr>
The SoC specific configuration structure is currently manually
assigned depending on the family_id. This will be removed in
the future. Make use of device_get_match_data() instead.
While we are here rename the structure prefix.
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/21866
Signed-off-by: Robert Marko <robimarko@gmail.com>
RTL93xx devices have 4 smi busses (0-3). Add them to the dts.
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/21438
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
The ethernet driver uses registers in the switchcore range.
Rearrange the DTS nodes accordingly. This allows to make use
of regmap with syscon_node_to_regmap(np->parent) later.
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/21183
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
The Realtek Otto ethernet driver currently uses a single compatible
for all different models. Split this into the the four well known
subtargets. This allows to get rid of the central mach/soc include
later.
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/21183
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
The RTL931x has next to its SPI flash controller a SPI master interface. It
is connected to
* SPI_CS#[1,0]: AH22 , AK22 (aka: GPIO 12, 11)
* SPI_CLK: AL23 (aka: GPIO 8)
* SPI_MISO: AM23 (aka: GPIO 9)
* SPI_MOSI: AL22 (aka: GPIO 10)
It is not the same as the SPI flash controller which uses pins:
* SPI_CS#[1,0]: B24, A24
* SPI_SCLK: A23
* SPI_SDI/SIO0: B21
* SPO_SDO_SIO1: B21
* SPI_SIO2: A22
* SPI_SIO3: B22
* SPI_RSTN: B23
As shown above, the SPI master controller shares its pin with GPIO 8, 9,
10, 11, 12. In some upcoming devices (like the Plasma Cloud PSX28/ESX28),
they will be used for SFP cage signaling. These pins must therefore be
switched manually to the GPIO mode.
The SPI_CTRL0 register provides all necessary configuration to enforce the
GPIO mode of the pins. And until more requirements (and a correct driver)
for the SPI master controller arise, it is therefore possible to use
pinctrl-single to configure it using the devicetree.
Previously the ethernet driver did configure the SPI master controller for
31.25 MHz. It is unknown for which kind of device this was originally made
and what was actually connected there. But this manual write to the
register conflicts potentially with the write of the pinctrl driver to the
same register. Luckily, we don't need this SPI speed configuration in the
ethernet driver. Still, to allow this device an easy migration, the
`spi0-31mhz` configuration was already prepared.
Signed-off-by: Sven Eckelmann <se@simonwunderlich.de>
Link: https://github.com/openwrt/openwrt/pull/20263
Signed-off-by: Robert Marko <robimarko@gmail.com>
Upstream will get support for the Realtek ECC engine with 6.18.
To make use of this in Openwrt
- backport upstream patches
- change config so that ECC will be built for nand subtargets
- define ECC engine in RTL93xx DTS.
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/19746
Signed-off-by: Robert Marko <robimarko@gmail.com>
Until now the the SerDes configuration is realized with helper functions
scattered around the DSA and PHY driver. Give them a new home as a PCS
driver.
The target design is as follows:
- dsa driver manages switch
- pcs driver manages SerDes on high level (this commit)
- mdio driver manages SerDes on low level
This driver adds the high level SerDes access via PCS. It makes use of
the low level mdio SerDes driver to access the registers.
Remark: This initial version provides exactly all phylink_pcs_ops that
are currently part of the DSA driver. So this can be swapped in one of
the next commits as a drop in replacement. To make use of it something
like this is needed:
...
ports = of_get_child_by_name(node, "ethernet-ports");
if (!ports)
return -EINVAL;
for_each_available_child_of_node(ports, port) {
pcs_node = of_parse_phandle(port, "pcs-handle", 0);
of_property_read_u32(port, "reg", &port_nr)) {
priv->pcs[port_nr] = rtpcs_create(dev, pcs_node, port_nr);
}
...
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/20075
Signed-off-by: Robert Marko <robimarko@gmail.com>
Currently following warnings are given
dts/rtl930x.dtsi:166.4-23: Warning (reg_format):
/switchcore@1b000000/i2c@36c:reg: property has invalid length
(8 bytes) (#address-cells == 2, #size-cells == 1)
Obviously default address-cells size is fixed to 64 bit. Align
with upstream and override address size to 32 bit.
Suggested-by: Jonas Jelonek <jelonek.jonas@gmail.com>
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/20091
Signed-off-by: Robert Marko <robimarko@gmail.com>
Until now the SerDes access is realized with some helper functions
in the mdio bus. These were moved around a lot and had no real home.
End that temporary solution to move them where they belong.
The target design for the different Realtek drivers is as follows:
- dsa driver manages switch
- pcs driver manages SerDes on high level (to be developed)
- mdio driver manages SerDes on low level (this commit)
This driver adds the low level SerDes access via mdio. For debugging
purposes the user can interact with the SerDes in different ways.
First, there is a debug interface in
/sys/kernel/debug/realtek_otto_serdes/serdes.X/registers.
With that a dump of all registers can be shown.
> cat /sys/kernel/debug/realtek_otto_serdes/serdes.4/registers
Back SDS 4: 00 01 02 03 04 05 06 07 08
SDS : 0C03 0F00 7060 7106 074D 0EBF 0F0F 0359 5248
SDS_EXT : 0000 0000 85FA 8C6D 5CCC 0000 20D8 0003 79AA
...
Second, one can read/write registers via the mmd functions of the
mdio command line tool. Important to know: The registers are accessed
on the vendor specific MDIO_MMD_VEND1 device address (=30). Additionally
the SerDes page and register are concatenated into the the mmd register.
Top 8 bits are SerDes page and bottom 8 bits are SerDEs register.
E.g.
- mmd 0x0206 : SerDes page 0x02, SerDes register 0x06
- mmd 0x041f : SerDes page 0x04, SerDes register 0x1f
Read register 0x02 on page 0x03 of SerDes 0
> mdio realtek-serdes-mdio mmd 0:30 raw 0x0302
Write register 0x12 on page 0x02 of SerDes 1
> mdio realtek-serdes-mdio mmd 1:30 raw 0x0212 0x2222
For now this driver is only defined in the devicetree and activated
in the kernel build. There is no current consumer but at least
the debugging interface is available. Cleanup of the currently used
SerDes functions will come later.
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/20062
Signed-off-by: Robert Marko <robimarko@gmail.com>
Until now the mdio bus is a subnode of the ethernet device. This
coupling is different from upstream and wrong. Ethernet and mdio
are different devices. Additionally differentiate between mdio
controller and mdio bus. To make it clear:
- There is one mdio controller
- With up to 4 busses (on RTL93xx)
Prepare new mdio controller and bus nodes with SoC specific compatibles.
These will be used later when refactoring the mdio driver probing.
Remark! For now only define the first bus for the RTL93xx targets.
So the driver still relies on "rtl9300,smi-address = <x y>;". It will
need much more refactoring to get totally aligned with upstream.
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/19986
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Registers must not be accessed in parallel by multiple drivers.
Read-modify-write operations are not atomic, and the result of parallel
access is undefined.
The MAC_L2_GLOBAL_CTRL2 register is essentially a pin configuration
register and is represented by a pinmux node in the devicetree. Operations
on this register by the realtek,rtl838x-eth driver must therefore also be
reflected in the devicetree.
Since the MDIO sets used are board-specific, the pins must be enabled in
the board’s devicetree. This can be achieved using the pinctrl properties
for the realtek,rtl83xx-switch.
&switch0 {
pinctrl-names = "default";
pinctrl-0 = <&pinmux_enable_mdc_mdio_0>,
<&pinmux_enable_mdc_mdio_1>;
....
};
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Link: https://github.com/openwrt/openwrt/pull/19815
Signed-off-by: Robert Marko <robimarko@gmail.com>
The pinmux-related registers on the RTL931X SoC family are spread across
various non-consecutive registers. It might be tempting to modify them
directly in a specific driver (SPI, LED, etc.), but this would cause issues
with parallel, non-locked read-modify-write operations, which are required
to update individual portions of these registers.
Instead, it is better to use the devicetree pinctrl properties to define
the correct configurations for the various operation modes.
One important setting here is the LED Sync bit. This allows the LED
controller to generate an additional positive edge on the `STCP`
("STore Clock Pin", also known as `RCLK`) of the LED shift register after
the actual content has already been shifted in using the normal shift
clock. The LED shift register is then expected to copy the content from the
shift register section into the storage registers, which act as the actual
LED output control. This functionality is available in, and commonly used
with, the SNx4HC595 family of shift registers.
To activate it, simply register it in the default state of the
"realtek,rtl83xx-switch" node:
&switch0 {
pinctrl-names = "default";
pinctrl-0 = <&pinmux_enable_led_sync>;
....
};
It would be nicer when this can be directly added to the led subnode. But
for this to work, `realtek,rtl9300-leds` must first be an actual driver
(known to the driver core).
[1] https://www.ti.com/lit/ds/symlink/sn74hc595.pdf
Suggested-by: Bevan Weiss <bevan.weiss@gmail.com>
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Link: https://github.com/openwrt/openwrt/pull/19815
Signed-off-by: Robert Marko <robimarko@gmail.com>
The MAC_L2_GLOBAL_CTRL2 register is primarily used for pin configuration.
It is necessary to select specific modes for pins or to free them for use
as GPIOs.
Fixes: 9dbc04785c ("realtek: add rtl8231-aux to rtl931x.dtsi")
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Link: https://github.com/openwrt/openwrt/pull/19815
Signed-off-by: Robert Marko <robimarko@gmail.com>
The RTL8231 auxiliary controller is not defined in the rtl931x.dtsi.
Additionally the pinmux is configured at the wrong address. Fix
this.
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/19776
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
The Realtek DTS's use several macros for convenient phy/port definition.
These are repeated for the RTL83xx targets and most are missing for the
RTL93xx targets. In the near future we want to add high port count
switches with 1GBit Ethernet for them too. As a preparation provide a
central include so the definition is only needed once and is available
for all targets.
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/19772
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Adapt the device tree definitions of rtl93xx devices and the base dtsi
for rtl930x and rtl931x to match with what's expected by the recently
backported RTL9300 I2C driver.
Signed-off-by: Jonas Jelonek <jelonek.jonas@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/19736
Signed-off-by: Robert Marko <robimarko@gmail.com>
Include the NAND specs into the DTS. It is unclear which devices
really need it. Keep it disabled for now. As the SoC register area
is defined too small until now, increase the size to an appropriate
value.
If enabled one can see the following log messages (e.g. Linksys
LGS328C or LGS352C).
[ 1.206600] spi-nand spi1.0: Macronix SPI NAND was found.
[ 1.212795] spi-nand spi1.0: 128 MiB, block size: 128 KiB, page size: 2048, OOB size: 64
[ 1.222217] 3 fixed-partitions partitions found on MTD device spi1.0
[ 1.229466] OF: Bad cell count for /soc/spi@1a400/flash@0/partitions
[ 1.236617] OF: Bad cell count for /soc/spi@1a400/flash@0/partitions
[ 1.244164] Creating 3 MTD partitions on "spi1.0":
[ 1.249620] 0x000000000000-0x000004000000 : "ubifs"
[ 1.423593] 0x000004000000-0x000005e00000 : "firmware"
[ 1.738268] mtdsplit_uimage: no uImage found in "firmware"
[ 1.744577] 0x000005e00000-0x000007c00000 : "runtime2"
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/19583
Signed-off-by: Robert Marko <robimarko@gmail.com>
Until now the timer management on the RTL931x devices depends
on the MIPS default timer. Looking at the clock progress on
these devices one can see that it is totally off. It is running
at half the required speed (e.g. if 1 minute passes the date
command shows that according to the timers only 30 seconds have
elapsed). This is a mix from wrong DTS and bad startup code.
This is not only a cosmetic issue but has effects on every
delay operation inside the kernel. Switch RTL931x to the proven
Otto timer.
Tested on LGS352C based on RTL9311.
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/19205
Signed-off-by: Robert Marko <robimarko@gmail.com>
Until now only the RTL930x devices make use of the following notation.
phy8: ethernet-phy@8 {
compatible = "ethernet-phy-ieee802.3-c22";
phy-is-integrated;
reg = <8>;
sds = <3>;
};
This indicates that the link is driven by a serdes directly without
external phy. As the devices have multiple serdes it must be clarified
what serdes is responsible for that port.
Nevertheless all other devices have the same requirements. E.g. RTL838x
usually drives port 24 from serdes 4 and port 26 from serdes 5. All this
currently works because the driver has a lot of hardcoded port/serdes
mapping.
Make the situation better by adding dts helpers that can describe the
topology as needed.
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/18851
Signed-off-by: Robert Marko <robimarko@gmail.com>
The SMP environment is prepared well for the RTL93X. Now describe the
power cluster controller in the DTS. Tested on RTL9311 based Linksys
LGS352C.
Without patch:
root@OpenWrt:~# dmesg | grep CPU
[ 0.140425] CPU1 revision is: 0001a120 (MIPS interAptiv (multi))
[ 0.191952] Synchronize counters for CPU 1: done.
[ 1.232191] CPU2: failed to start
[ 1.237863] No online CPU in core 1 to start CPU3
[ 2.273784] CPU3: failed to start
[ 2.277589] smp: Brought up 1 node, 2 CPUs
root@OpenWrt:~# cat /proc/cpuinfo | grep -E "model|proc"
processor : 0
cpu model : MIPS interAptiv (multi) V2.0
processor : 1
cpu model : MIPS interAptiv (multi) V2.0
With patch:
root@OpenWrt:~# dmesg | grep CPU
[ 0.000000] CPU0 revision is: 0001a120 (MIPS interAptiv (multi))
[ 0.000000] SLUB: HWalign=32, Order=0-3, MinObjects=0, CPUs=4, Nodes=1
[ 0.000000] Failed to get CPU clock: -2
[ 0.000000] CPU frequency from device tree: 1000MHz
[ 0.133360] smp: Bringing up secondary CPUs ...
[ 0.140418] CPU1 revision is: 0001a120 (MIPS interAptiv (multi))
[ 0.191950] Synchronize counters for CPU 1: done.
[ 0.230103] CPU2 revision is: 0001a120 (MIPS interAptiv (multi))
[ 0.289220] Synchronize counters for CPU 2: done.
[ 0.326189] CPU3 revision is: 0001a120 (MIPS interAptiv (multi))
[ 0.378861] Synchronize counters for CPU 3: done.
[ 0.413829] smp: Brought up 1 node, 4 CPUs
processor : 0
cpu model : MIPS interAptiv (multi) V2.0
processor : 1
cpu model : MIPS interAptiv (multi) V2.0
processor : 2
cpu model : MIPS interAptiv (multi) V2.0
processor : 3
cpu model : MIPS interAptiv (multi) V2.0
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/19110
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
The switchcore node is the central location that describes the Realtek switch
register addresses starting at 0x1b000000. It will be used by current and
future regmap enabled device drivers. The upstream MDIO driver already makes
use of it by calling syscon_node_to_regmap(dev->parent->of_node);
In the current DTS base we have 3 issues that should be fixed:
- rtl838x.dtsi has a length of 0x20000 instead of 0x10000
- rtl839x.dtsi has a length of 0x20000 instead of 0x10000
- rtl931x.dtsi has no switchcore node at all
Align these mismatches with the "good" RTL930x template.
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/18642
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
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>
