| Age | Commit message (Collapse) | Author |
|---|
|
Reserve a 2 MiB memory region to record kmsg dumps, console, ftrace and
userspace messages. The implemented memory split allows capturing and
reading corresponding ring buffers:
* dmesg: 6 dumps, 128 KiB each
* console: 128 KiB
* ftrace: 128 KiB for each of 8 CPUs (1 MiB total)
* userspace messages: 128 KiB
Signed-off-by: Sam Protsenko <semen.protsenko@linaro.org>
Link: https://lore.kernel.org/r/20231008033633.21304-1-semen.protsenko@linaro.org
[krzysztof: move the node to alphabetically sorted position]
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
|
|
The E850-96 board has a micro-USB socket and two USB 2.0 host sockets.
The USB role (host or peripheral) is selected automatically depending on
micro-USB cable attachment state:
- micro-USB cable is attached: USB device role
- micro-USB cable is detached: USB host role
USB can't act simultaneously as a device and a host, because Exynos850
SoC has only one USB controller and there are no external USB
controllers on the E850-96 board. So the USB switch chip (specifically
TS3USB221A) connects SoC USB lines either to micro-USB connector or to
USB hub chip (LAN9514), w.r.t. micro-USB cable attachment state.
When USB works in the host role, Ethernet capability becomes available
too, as the LAN9514 chip (providing USB hub) also enables Ethernet PHY
and Ethernet MAC.
Dynamic role switching is done in gpio-usb-b-connector, using current
micro-USB VBUS line level as a trigger:
- VBUS=high: SoC USB lines are wired to micro-USB socket
- VBUS=low: SoC USB lines are wired to USB hub chip
In order to make USB host functional when the board was booted with
micro-USB cable disconnected, role-switch-default-mode = "host" is used.
One can use E850-96 board schematics [1] to figure out how exactly all
related USB hardware connections and lines reflect into corresponding
device tree definitions.
As PMIC regulators are not implemented yet, we rely on USB LDOs being
already enabled in the bootloader. A dummy regulator is provided to
"usbdrd" vdd nodes for now.
[1] https://www.96boards.org/documentation/consumer/e850-96b/hardware-docs/
Signed-off-by: Sam Protsenko <semen.protsenko@linaro.org>
Link: https://lore.kernel.org/r/20230825215445.28309-3-semen.protsenko@linaro.org
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
|
|
Switch source of LED activity to hci0-power from RX, to match bindings
(same effect expected):
exynos850-e850-96.dtb: leds: led-5:linux,default-trigger: 'oneOf' conditional failed, one must be fixed:
'hci0rx' is not one of ['backlight', 'default-on', 'heartbeat', 'disk-activity', 'ide-disk', 'timer', 'pattern']
'hci0rx' does not match '^cpu[0-9]*$'
'hci0rx' does not match '^hci[0-9]+-power$'
'hci0rx' does not match '^mmc[0-9]+$'
'hci0rx' does not match '^phy[0-9]+tx$'
Link: https://lore.kernel.org/r/20230120173116.341270-3-krzysztof.kozlowski@linaro.org
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
|
|
The aliases for typical blocks which are disabled by default in DTSI
(like I2C, UART and MMC) should be defined in the board DTS. The board
should add aliases only for enabled blocks according to its specific
order.
On Exynos850, move aliases of enabled blocks to E850-96 board and remove
unused ones.
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20220221075219.10827-1-krzysztof.kozlowski@canonical.com
|
|
E850-96 is a 96boards development board manufactured by WinLink. It
incorporates Samsung Exynos850 SoC, and is compatible with 96boards
mezzanine boards [1], as it follows 96boards standards.
This patch adds minimal support for E850-96 board. Next features are
enabled in board dts file and verified with minimal BusyBox rootfs:
* User buttons
* LEDs
* Serial console
* Watchdog timers
* RTC
* eMMC
[1] https://www.96boards.org/products/mezzanine/
Signed-off-by: Sam Protsenko <semen.protsenko@linaro.org>
Link: https://lore.kernel.org/r/20220131130849.2667-3-semen.protsenko@linaro.org
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>
|
