Linux-ARM-Kernel Archive on lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH v5 0/2] arm64: dts: rockchip: Add ArmSoM CM5 and CM5-IO
@ 2026-07-19  3:46 Jiaxing Hu
  2026-07-19  3:46 ` [PATCH v5 1/2] dt-bindings: arm: " Jiaxing Hu
  2026-07-19  3:46 ` [PATCH v5 2/2] arm64: dts: " Jiaxing Hu
  0 siblings, 2 replies; 3+ messages in thread
From: Jiaxing Hu @ 2026-07-19  3:46 UTC (permalink / raw)
  To: heiko, robh, krzk+dt, conor+dt
  Cc: devicetree, linux-rockchip, linux-arm-kernel, linux-kernel,
	Jiaxing Hu

Add mainline support for the ArmSoM CM5, an RK3576 compute module, and
its CM5-IO carrier board.

Patch 1 adds the binding, patch 2 the module dtsi and the carrier dts.

The on-module YT8531 PHY has no crystal and takes its 25 MHz reference
from the SoC. Following review of v4, the reference is now handled the
way Andrew Lunn and Maxime Chevallier asked for: the PHY is the consumer
and enables the clock itself. That needs a small motorcomm change sent
to netdev [1]; the clock is optional so this DTS still applies without
it. The earlier dwmac-rk MAC-side patch is dropped.

Tested on a CM5-IO with [1] applied: the YT8531 links at 1000 Mbit/s;
RK806, HYM8563, eMMC, microSD, the USB3 hub and PCIe all probe.
dtbs_check is clean.

Changes in v5:
- The 25 MHz reference is now consumed by the PHY, not the MAC. The PHY
  node gets its clock back (clocks = <&cru REFCLKO25M_GMAC0_OUT>) and is
  named explicitly (ethernet-phy-id4f51.e91b) so it is instantiated from
  the DT without an MDIO ID probe, which it cannot answer before the
  clock runs. &gmac0 drops clk_mac_refout and inherits the SoC clocks.
  This reverses the v4 removal of the PHY-node clock, which was the wrong
  direction.
- Enable &sai6, the I2S CPU DAI the HDMI sound card needs (already in v4;
  re-tested here, HDMI audio card registers).

Changes in v4:
- phy-mode = "rgmii-id", drop the MAC tx_delay/rx_delay; the PHY adds
  both delays (Andrew Lunn). Re-tested at 1000 Mbit/s.
- Pick up Krzysztof's Acked-by on patch 1.

Changes in v3:
- Remove the ES8388 codec from the header and the i2c0 comment; use gpios
  in the vcc_3v3_pcie regulator.

Changes in v2:
- Fix the Type-C DP alt-mode node, move the FUSB302 role-switch endpoint
  into the connector, drop the ES8388 node and dead pinctrls, rename the
  wifi pinctrl node, and describe the PHY reset at the MDIO bus level.

[1] net: phy: motorcomm: enable the reference clock for YT8531

Jiaxing Hu (2):
  dt-bindings: arm: rockchip: Add ArmSoM CM5 and CM5-IO
  arm64: dts: rockchip: Add ArmSoM CM5 and CM5-IO

 .../devicetree/bindings/arm/rockchip.yaml     |   7 +
 arch/arm64/boot/dts/rockchip/Makefile         |   1 +
 .../dts/rockchip/rk3576-armsom-cm5-io.dts     | 402 +++++++++++++
 .../boot/dts/rockchip/rk3576-armsom-cm5.dtsi  | 558 ++++++++++++++++++
 4 files changed, 968 insertions(+)
 create mode 100644 arch/arm64/boot/dts/rockchip/rk3576-armsom-cm5-io.dts
 create mode 100644 arch/arm64/boot/dts/rockchip/rk3576-armsom-cm5.dtsi

-- 
2.43.0



^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2026-07-19  3:46 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-07-19  3:46 [PATCH v5 0/2] arm64: dts: rockchip: Add ArmSoM CM5 and CM5-IO Jiaxing Hu
2026-07-19  3:46 ` [PATCH v5 1/2] dt-bindings: arm: " Jiaxing Hu
2026-07-19  3:46 ` [PATCH v5 2/2] arm64: dts: " Jiaxing Hu

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox