[PATCH v3 0/3] Add Luckfox Omni3576 Carrier Board support for RK3576

Heiko Stübner heiko at sntech.de
Tue May 6 14:18:18 PDT 2025


Am Dienstag, 6. Mai 2025, 20:23:55 Mitteleuropäische Sommerzeit schrieb Rob Herring (Arm):
> 
> On Tue, 06 May 2025 07:41:12 -0400, John Clark wrote:
> > This series adds device tree support for the Luckfox Omni3576
> > Carrier Board with the Core3576 Module, powered by the Rockchip
> > RK3576 SoC (four Cortex-A72 cores, four Cortex-A53 cores, Mali-G52
> > MC3 GPU). It enables essential functionality for booting Linux and
> > basic connectivity, with plans for future support of peripherals
> > like WiFi, MIPI-DSI, HDMI, and Ethernet.
> > 
> > The series was first posted as v1 at:
> > https://lore.kernel.org/linux-rockchip/20250502205533.51744-1-inindev@gmail.com
> > v2 at:
> > https://lore.kernel.org/linux-rockchip/20250504102447.153551-1-inindev@gmail.com
> > 
> > Tested features (on Linux 6.15-rc4):
> >  - UART: Serial console operational
> >  - SD card: Mounts and reads/writes successfully
> >  - PCIe: NVMe SSD detected, mounted, and fully functional
> >  - USB 2.0: Host ports operational
> >  - RTC: Timekeeping and wake-up tested
> >  - LED: Heartbeat trigger functional
> >  - eMMC: Enabled in device tree, not populated on tested hardware
> > 
> > The series includes three patches:
> >  1. dt-bindings: vendor-prefixes: Add Luckfox vendor prefix
> >  2. dt-bindings: arm: rockchip: Add Luckfox Omni3576 and Core3576 bindings
> >  3. arm64: dts: rockchip: Add Luckfox Omni3576 board support
> > 
> > The device tree is covered by the existing ROCKCHIP ARCHITECTURE
> > entry in MAINTAINERS. I am aware of ongoing RK3576 upstreaming
> > efforts (e.g., by Collabora) and welcome feedback or collaboration
> > to align with mainline driver development.
> > 
> > Changes in v3:
> >  - Addressed Jonas Karlman's feedback on patch 3/3
> >    (Luckfox Omni3576 device tree):
> >    - Added pinctrl for green LED GPIO (gpio1 RK_PD5) for proper
> >      pin setup.
> >    - Reordered regulator node properties for consistent sequence
> >      (e.g., regulator-name, regulator-min-microvolt,
> >      regulator-max-microvolt, etc.).
> >    - Updated regulator nodes (vcc_3v3_pcie, vbus_5v0_typec,
> >      vcc_5v0_host, vcc_5v0_hdmi) to use 'gpios' property instead
> >      of deprecated 'gpio'.
> >    - Removed pmic-power-off pinctrl state and pinctrl-1 from RK806
> >      PMIC node, as they are vendor-specific and undocumented in
> >      bindings.
> >    - Removed pwrkey node from PMIC, as it lacks binding
> >      documentation.
> >    - Added blank line between properties and child nodes in i2c2
> >      node for DT style compliance.
> >    - Removed no-mmc property from sdmmc node to enable MMC
> >      support, aligning with RK3576 SD v3.0 and MMC v4.51
> >      capabilities, allowing TF card slot to support MMC devices
> >      or eMMC via adapter.
> >  - Removed Ethernet support (gmac0/gmac1 nodes) per Andrew Lunn's
> >    feedback, as it relied on the generic PHY driver with incorrect
> >    RGMII delay settings, which would break with the upcoming
> >    MAE0621A driver.
> >  - Addressed Rob Herring's DTB check warnings, other warnings
> >    (e.g., VOP, PCIe, OTP, HDMI PHY) originate from rk3576.dtsi and
> >    are outside this patchset scope.
> >  - Added RNG node to pick up Nicolas Frattaroli's "add RK3576 RNG
> >    node" patch
> > 
> > Changes in v2:
> >  - Enabled HDMI node per feedback from Heiko Stuebner and Nicolas
> >    Frattaroli; untested due to upstream driver issues.
> >  - Enabled Ethernet 1 node per Heiko's device tree philosophy;
> >    untested due to suspected PHY driver or configuration issues
> >    (removed in v3 per Andrew Lunn).
> >  - Clarified eMMC remains enabled but unpopulated on tested
> >    board, per Heiko.
> > 
> > Signed-off-by: John Clark <inindev at gmail.com>
> > ---
> > John Clark (3):
> >   dt-bindings: vendor-prefixes: Add luckfox prefix
> >   dt-bindings: arm: rockchip: Add Luckfox Omni3576 and Core3576 bindings
> >   arm64: dts: rockchip: Add Luckfox Omni3576 Board support
> > 
> >  .../devicetree/bindings/arm/rockchip.yaml     |  10 +
> >  .../devicetree/bindings/vendor-prefixes.yaml  |   2 +
> >  arch/arm64/boot/dts/rockchip/Makefile         |   1 +
> >  .../dts/rockchip/rk3576-luckfox-core3576.dtsi | 683 ++++++++++++++++++
> >  .../dts/rockchip/rk3576-luckfox-omni3576.dts  |  53 ++
> >  5 files changed, 749 insertions(+)
> >  create mode 100644 arch/arm64/boot/dts/rockchip/rk3576-luckfox-core3576.dtsi
> >  create mode 100644 arch/arm64/boot/dts/rockchip/rk3576-luckfox-omni3576.dts
> > 
> > --
> > 2.39.5
> > 
> > 
> > 
> 
> 
> My bot found new DTB warnings on the .dts files added or changed in this
> series.
> 
> Some warnings may be from an existing SoC .dtsi. Or perhaps the warnings
> are fixed by another series. Ultimately, it is up to the platform
> maintainer whether these warnings are acceptable or not. No need to reply
> unless the platform maintainer has comments.
> 
> If you already ran DT checks and didn't see these error(s), then
> make sure dt-schema is up to date:
> 
>   pip3 install dtschema --upgrade
> 
> 
> This patch series was applied (using b4) to base:
>  Base: attempting to guess base-commit...
>  Base: tags/next-20250506 (best guess, 2/3 blobs matched)
> 
> If this is not the correct base, please add 'base-commit' tag
> (or use b4 which does this automatically)
> 
> New warnings running 'make CHECK_DTBS=y for arch/arm64/boot/dts/rockchip/' for 20250506114115.613616-1-inindev at gmail.com:
> 
> arch/arm64/boot/dts/rockchip/rk3576-luckfox-omni3576.dtb: /soc/rng at 2a410000: failed to match any schema with compatible: ['rockchip,rk3576-rng']

the binding (and driver) for the rk3576-rng have been merged into the
cryptodev tree yesterday [0], so that issue should hopefully go away with
the next linux-next.


[0] https://git.kernel.org/pub/scm/linux/kernel/git/herbert/cryptodev-2.6.git/commit/?id=165ef524bbeb71ccd470e70a4e63f813fa71e7cd





More information about the linux-arm-kernel mailing list