rockchip64-6.19: arm64: dts: rockchip: describe pcie ethernets on FriendlyElec NanoPC-T6 #9238

rpardini · 2026-01-12T23:35:37Z

🌱 By fully describing the pcie hardware for the RTL8125s (rtl8169) present on the board, we get to alias them to ethernet0/1. In return, (mainline) u-boot will by default patch the runtime FDT though it's fdt_fixup_ethernet() implementation, hopefully resulting in fully stable MAC addresses without the need to program the RTL NIC's EEPROM; as an added benefit, u-boot MAC & kernel MAC would match, facilitating network-boot setups.

Summary by CodeRabbit

New Features
- Added support for PCIe ethernet devices on NanoPC-T6 platform.
- Configured dual ethernet interfaces (ethernet0 and ethernet1) mapped to PCIe bridges.
- Updated MAC address configuration for stable network initialization.

_{✏️ Tip: You can customize this high-level summary in your review settings.}

coderabbitai · 2026-01-12T23:35:43Z

📝 Walkthrough

Walkthrough

A device tree patch for the Rockchip RK3588 NanoPC-T6 board that adds PCIe ethernet device definitions with RTL8125-compatible network interfaces and reserved MAC address fields for U-Boot bootloader patching.

Changes

Cohort / File(s)	Summary
PCIe Ethernet Device Tree Definitions `arch/arm64/boot/dts/rockchip/rk3588-nanopc-t6.dtsi`	Added ethernet aliases (ethernet0, ethernet1) and two PCIe subtrees (&pcie2x1l1, &pcie2x1l2) each containing an ethernet node with RTL8125 compatible driver references. Included placeholder MAC addresses [00 00 00 00 00 00] for U-Boot patching. (+82 lines)

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~5 minutes

Possibly related PRs

mainline-kernel: bump 6.18 to rc2 + rockchip64 fixes + T6/CM3588 NPU DT #8786: Modifies the same rk3588-nanopc-t6.dtsi device tree file to add or adjust NanoPC-T6 hardware entries.

Suggested reviewers

Tonymac32
SuperKali
JohnTheCoolingFan
HeyMeco
TheSnowfield

Poem

🐰 Hops through the device tree with glee,
Two ethernet ports, now all can see!
RTL8125s on PCIe so bright,
MAC addresses waiting for U-Boot's light. 🌟

🚥 Pre-merge checks | ✅ 3

✅ Passed checks (3 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately describes the main change: adding PCIe ethernet descriptions for RTL8125 devices on the FriendlyElec NanoPC-T6 board in the device tree, which aligns with the patch content and objectives.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

📜 Recent review details

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Disabled knowledge base sources:

Jira integration is disabled by default for public repositories

You can enable these sources in your CodeRabbit configuration.

📥 Commits

Reviewing files that changed from the base of the PR and between 25f6eec and bf4d994.

📒 Files selected for processing (1)

patch/kernel/archive/rockchip64-6.19/rk3588-1220-arm64-dts-rockchip-describe-pcie-ethernets-on-FriendlyElec-NanoPC-T6.patch

🧰 Additional context used

🧠 Learnings (13)

📓 Common learnings

Learnt from: rpardini
Repo: armbian/build PR: 9233
File: patch/u-boot/v2026.01/1000-rockchip-rk3588-stable-MAC-addresses-system-fixup-on-fdt-gmacX.patch:86-94
Timestamp: 2026-01-12T17:21:32.630Z
Learning: For U-Boot patches that implement custom MAC address injection into device tree nodes (such as for Rockchip RK3588 gmac interfaces), consider using U-Boot's standard fdt_fixup_ethernet() function instead. This function automatically walks /aliases looking for ethernet, ethernet0, ethernet1, etc., resolves each to a device node, and writes the corresponding environment MAC (ethaddr, eth1addr) into the mac-address or local-mac-address property. The standard approach requires: 1) adding ethernet aliases to board device trees (e.g., ethernet0 = &gmac0), 2) calling fdt_fixup_ethernet(blob) in ft_board_setup() or ft_system_setup(), and 3) ensuring ethaddr/eth1addr are set in the environment. This is simpler, well-tested, and easier to upstream than custom implementations.

Learnt from: EvilOlaf
Repo: armbian/build PR: 0
File: :0-0
Timestamp: 2025-12-19T13:56:45.124Z
Learning: When reviewing kernel or u-boot version bump PRs in the Armbian build system, check if patches existed in previous kernel version directories (e.g., sunxi-6.12, sunxi-6.13) before describing them as new features. If a patch and the majority of its contents existed previously with no major functionality changes, focus the review on the actual changes: the version bump itself and patch compatibility adjustments. Don't describe existing patches being ported/maintained across versions as new features or drivers—this is misleading. The patches are existing code being re-aligned to work with the new upstream version.

Learnt from: EvilOlaf
Repo: armbian/build PR: 0
File: :0-0
Timestamp: 2025-12-17T05:09:02.306Z
Learning: In the Armbian build system, kernel patches (e.g., in patch/kernel/archive/sunxi-6.18/patches.armbian/) contain device tree overlays and other code that have existed for years and are maintained/ported across kernel versions. When reviewing PRs that modify these patch files, focus on the actual changes being made (e.g., Makefile fixes, new additions) rather than reviewing the entire existing content within the patch as if it were new code. The patch file contents are existing, stable code unless explicitly modified in the PR diff.

Learnt from: EvilOlaf
Repo: armbian/build PR: 8968
File: patch/u-boot/u-boot-sunxi/arm64-dts-sun50i-h6-orangepi.dtsi-Rollback-r_rsb-to-r_i2c.patch:36-36
Timestamp: 2025-11-20T18:20:11.985Z
Learning: The rewrite-patches tool (REWRITE_PATCHES=yes) in the Armbian build system can inadvertently introduce semantic changes when the u-boot/kernel git base revision differs from expected state. The tool applies patches, commits them, and re-exports them using git format-patch, which can cause the re-exported patch to reflect the base revision's state rather than preserving the original patch intent. This is particularly problematic for device tree changes like interrupt specifications. The tool currently lacks validation mechanisms to detect such semantic drift, and affected patches must be manually corrected after rewriting.

Learnt from: rpardini
Repo: armbian/build PR: 9159
File: patch/u-boot/u-boot-genio/0026-dts-configs-add-Grinn-GenioSBC-510.patch:161-161
Timestamp: 2026-01-03T20:46:29.189Z
Learning: For the Armbian genio family (config/sources/families/genio.conf and patch/u-boot/u-boot-genio/), when reviewing PRs that include vendor U-Boot patches from Collabora, avoid flagging potential issues in board configurations that are out of scope for the PR's primary focus (e.g., don't flag Genio 510/700 board issues when the PR is focused on radxa-nio-12l/Genio 1200). The maintainer prioritizes keeping vendor patches close to upstream for easier re-copying and maintenance, even if secondary board configs have potential mismatches.
<!-- </add_learning>

Learnt from: leggewie
Repo: armbian/build PR: 0
File: :0-0
Timestamp: 2025-08-29T18:44:47.732Z
Learning: When creating GitHub issues for code improvements in the Armbian build project, include the relevant patches/diffs in the issue description or open a PR directly instead of just describing the changes. This makes it more efficient for the assignee to handle the improvements.

Learnt from: amazingfate
Repo: armbian/build PR: 8619
File: config/sources/families/rockchip.conf:222-230
Timestamp: 2025-09-14T06:32:29.806Z
Learning: In the Armbian build system, the write_uboot_platform() function implementations follow different patterns across Rockchip family files. The newer standard (used in rockchip64_common.inc and rk3506) includes 'status=none' parameter in dd commands, while older implementations (rk3288, rk322x) use an older pattern without this parameter. The rk3506 implementation correctly follows the current Rockchip family standard.

Learnt from: tabrisnet
Repo: armbian/build PR: 8661
File: lib/functions/compilation/armbian-kernel.sh:194-199
Timestamp: 2025-09-25T18:37:00.330Z
Learning: In PR armbian/build#8661, line 235 of lib/functions/compilation/armbian-kernel.sh already contains the corrected comment "BPF link support for netfilter hooks" for NETFILTER_BPF_LINK, not the misleading "BPF_SYSCALL" comment that was flagged during review.

Learnt from: tabrisnet
Repo: armbian/build PR: 8661
File: lib/functions/compilation/armbian-kernel.sh:194-199
Timestamp: 2025-09-25T18:37:00.330Z
Learning: In PR armbian/build#8661, line 235 of lib/functions/compilation/armbian-kernel.sh already contains the corrected comment "BPF link support for netfilter hooks" for NETFILTER_BPF_LINK, not the misleading "BPF_SYSCALL" comment that was flagged during review.

Learnt from: EvilOlaf
Repo: armbian/build PR: 9086
File: patch/kernel/archive/uefi-arm64-6.19/net-stmicro-stmmac-Phytium-onboard-ethernet-drivers-and-ACPI-glue-for-6.x.patch.disabled:625-633
Timestamp: 2025-12-21T10:28:48.455Z
Learning: The patch "net-stmicro-stmmac-Phytium-onboard-ethernet-drivers-and-ACPI-glue-for-6.x.patch" in patch/kernel/archive/uefi-arm64-6.19/ is known to break compilation on kernel 6.19 and is therefore disabled (has .disabled extension).

Learnt from: djurny
Repo: armbian/build PR: 8235
File: packages/bsp/mvebu/helios4/helios4-wol.service:0-0
Timestamp: 2025-05-29T01:56:01.604Z
Learning: For ARM-based devices like Helios4 that use DeviceTree, the expected systemd network interface naming scheme is "end[0-9]+" (DeviceTree alias index), making the regex "^(eth|en[do])[0-9]+" appropriate as it covers traditional "eth", PCI on-board "eno", and DeviceTree "end" interfaces without unnecessary "enp" (PCI slot) support.

Learnt from: pyavitz
Repo: armbian/build PR: 9058
File: patch/u-boot/legacy/u-boot-spacemit-k1/003-SpacemiT-K1X-Fixups.patch:17-25
Timestamp: 2025-12-13T11:45:02.422Z
Learning: In the SpacemiT U-Boot patches for Armbian (patch/u-boot/legacy/u-boot-spacemit-k1/), the environment variable `devnum` is set to the device name string (e.g., "mmc", "nvme") rather than a numeric index, and `distro_bootpart` holds the partition number. This implementation aligns with mainline U-Boot conventions for the SpacemiT platform and has been verified to work correctly by the maintainer.

Learnt from: rpardini
Repo: armbian/build PR: 8879
File: config/sources/families/uefi-x86.conf:0-0
Timestamp: 2025-11-06T15:36:04.682Z
Learning: As of PR #8879, the uefi-x86 family in the Armbian build system now includes kernel patches for the first time. The current and edge branches for uefi-x86 are specifically configured for Apple T2-based x86 machines, including T2-specific patches from the linux-t2 project and custom kernel configuration options for Apple hardware drivers.

Learnt from: SuperKali
Repo: armbian/build PR: 8936
File: patch/kernel/archive/rockchip64-6.18/dt/rk3576-nanopi-r76s.dts:800-802
Timestamp: 2025-11-15T14:33:30.654Z
Learning: In RK3576 device tree files (like rk3576-nanopi-r76s.dts), the serial0 alias is already defined in the upstream kernel's rk3576.dtsi base file, so board-specific DTS files do not need to redeclare it when enabling &uart0 for serial console.

Learnt from: pyavitz
Repo: armbian/build PR: 9058
File: patch/u-boot/legacy/u-boot-spacemit-k1/003-SpacemiT-K1X-Fixups.patch:28-67
Timestamp: 2025-12-13T11:39:08.046Z
Learning: In the Armbian build system for SpacemiT U-Boot patches (patch/u-boot/legacy/u-boot-spacemit-k1/), alignment with mainline U-Boot behavior is prioritized. For example, in boot mode handling, leaving devnum unchanged in the default case (when devtype is cleared) follows mainline conventions rather than explicitly clearing it to handle edge cases.

Learnt from: amazingfate
Repo: armbian/build PR: 8595
File: lib/functions/compilation/patch/drivers_network.sh:0-0
Timestamp: 2025-09-09T07:30:22.419Z
Learning: RTW_WARN_LMT sed workaround in driver_rtl8852bs() function in lib/functions/compilation/patch/drivers_network.sh is necessary for kernel 6.1 compatibility, even when the upstream wifi-rtl8852bs driver has the macro defined for newer kernels. The workaround ensures cross-kernel compatibility from 6.1 to 6.16+.

📚 Learning: 2026-01-12T17:21:32.630Z

Learnt from: rpardini
Repo: armbian/build PR: 9233
File: patch/u-boot/v2026.01/1000-rockchip-rk3588-stable-MAC-addresses-system-fixup-on-fdt-gmacX.patch:86-94
Timestamp: 2026-01-12T17:21:32.630Z
Learning: For U-Boot patches that implement custom MAC address injection into device tree nodes (such as for Rockchip RK3588 gmac interfaces), consider using U-Boot's standard fdt_fixup_ethernet() function instead. This function automatically walks /aliases looking for ethernet, ethernet0, ethernet1, etc., resolves each to a device node, and writes the corresponding environment MAC (ethaddr, eth1addr) into the mac-address or local-mac-address property. The standard approach requires: 1) adding ethernet aliases to board device trees (e.g., ethernet0 = &gmac0), 2) calling fdt_fixup_ethernet(blob) in ft_board_setup() or ft_system_setup(), and 3) ensuring ethaddr/eth1addr are set in the environment. This is simpler, well-tested, and easier to upstream than custom implementations.

Applied to files:

patch/kernel/archive/rockchip64-6.19/rk3588-1220-arm64-dts-rockchip-describe-pcie-ethernets-on-FriendlyElec-NanoPC-T6.patch

📚 Learning: 2025-12-21T10:28:48.455Z

Learnt from: EvilOlaf
Repo: armbian/build PR: 9086
File: patch/kernel/archive/uefi-arm64-6.19/net-stmicro-stmmac-Phytium-onboard-ethernet-drivers-and-ACPI-glue-for-6.x.patch.disabled:625-633
Timestamp: 2025-12-21T10:28:48.455Z
Learning: The patch "net-stmicro-stmmac-Phytium-onboard-ethernet-drivers-and-ACPI-glue-for-6.x.patch" in patch/kernel/archive/uefi-arm64-6.19/ is known to break compilation on kernel 6.19 and is therefore disabled (has .disabled extension).

Applied to files:

patch/kernel/archive/rockchip64-6.19/rk3588-1220-arm64-dts-rockchip-describe-pcie-ethernets-on-FriendlyElec-NanoPC-T6.patch

📚 Learning: 2025-12-17T05:09:02.306Z

Learnt from: EvilOlaf
Repo: armbian/build PR: 0
File: :0-0
Timestamp: 2025-12-17T05:09:02.306Z
Learning: In the Armbian build system, kernel patches (e.g., in patch/kernel/archive/sunxi-6.18/patches.armbian/) contain device tree overlays and other code that have existed for years and are maintained/ported across kernel versions. When reviewing PRs that modify these patch files, focus on the actual changes being made (e.g., Makefile fixes, new additions) rather than reviewing the entire existing content within the patch as if it were new code. The patch file contents are existing, stable code unless explicitly modified in the PR diff.

Applied to files:

patch/kernel/archive/rockchip64-6.19/rk3588-1220-arm64-dts-rockchip-describe-pcie-ethernets-on-FriendlyElec-NanoPC-T6.patch

📚 Learning: 2025-05-29T01:56:01.604Z

Learnt from: djurny
Repo: armbian/build PR: 8235
File: packages/bsp/mvebu/helios4/helios4-wol.service:0-0
Timestamp: 2025-05-29T01:56:01.604Z
Learning: For ARM-based devices like Helios4 that use DeviceTree, the expected systemd network interface naming scheme is "end[0-9]+" (DeviceTree alias index), making the regex "^(eth|en[do])[0-9]+" appropriate as it covers traditional "eth", PCI on-board "eno", and DeviceTree "end" interfaces without unnecessary "enp" (PCI slot) support.

Applied to files:

patch/kernel/archive/rockchip64-6.19/rk3588-1220-arm64-dts-rockchip-describe-pcie-ethernets-on-FriendlyElec-NanoPC-T6.patch

📚 Learning: 2025-11-20T18:20:11.985Z

Learnt from: EvilOlaf
Repo: armbian/build PR: 8968
File: patch/u-boot/u-boot-sunxi/arm64-dts-sun50i-h6-orangepi.dtsi-Rollback-r_rsb-to-r_i2c.patch:36-36
Timestamp: 2025-11-20T18:20:11.985Z
Learning: The rewrite-patches tool (REWRITE_PATCHES=yes) in the Armbian build system can inadvertently introduce semantic changes when the u-boot/kernel git base revision differs from expected state. The tool applies patches, commits them, and re-exports them using git format-patch, which can cause the re-exported patch to reflect the base revision's state rather than preserving the original patch intent. This is particularly problematic for device tree changes like interrupt specifications. The tool currently lacks validation mechanisms to detect such semantic drift, and affected patches must be manually corrected after rewriting.

Applied to files:

patch/kernel/archive/rockchip64-6.19/rk3588-1220-arm64-dts-rockchip-describe-pcie-ethernets-on-FriendlyElec-NanoPC-T6.patch

📚 Learning: 2025-12-19T13:56:45.124Z

Learnt from: EvilOlaf
Repo: armbian/build PR: 0
File: :0-0
Timestamp: 2025-12-19T13:56:45.124Z
Learning: When reviewing kernel or u-boot version bump PRs in the Armbian build system, check if patches existed in previous kernel version directories (e.g., sunxi-6.12, sunxi-6.13) before describing them as new features. If a patch and the majority of its contents existed previously with no major functionality changes, focus the review on the actual changes: the version bump itself and patch compatibility adjustments. Don't describe existing patches being ported/maintained across versions as new features or drivers—this is misleading. The patches are existing code being re-aligned to work with the new upstream version.

Applied to files:

patch/kernel/archive/rockchip64-6.19/rk3588-1220-arm64-dts-rockchip-describe-pcie-ethernets-on-FriendlyElec-NanoPC-T6.patch

📚 Learning: 2025-11-15T14:33:30.654Z

Learnt from: SuperKali
Repo: armbian/build PR: 8936
File: patch/kernel/archive/rockchip64-6.18/dt/rk3576-nanopi-r76s.dts:800-802
Timestamp: 2025-11-15T14:33:30.654Z
Learning: In RK3576 device tree files (like rk3576-nanopi-r76s.dts), the serial0 alias is already defined in the upstream kernel's rk3576.dtsi base file, so board-specific DTS files do not need to redeclare it when enabling &uart0 for serial console.

Applied to files:

patch/kernel/archive/rockchip64-6.19/rk3588-1220-arm64-dts-rockchip-describe-pcie-ethernets-on-FriendlyElec-NanoPC-T6.patch

📚 Learning: 2025-12-13T11:45:02.422Z

Learnt from: pyavitz
Repo: armbian/build PR: 9058
File: patch/u-boot/legacy/u-boot-spacemit-k1/003-SpacemiT-K1X-Fixups.patch:17-25
Timestamp: 2025-12-13T11:45:02.422Z
Learning: In the SpacemiT U-Boot patches for Armbian (patch/u-boot/legacy/u-boot-spacemit-k1/), the environment variable `devnum` is set to the device name string (e.g., "mmc", "nvme") rather than a numeric index, and `distro_bootpart` holds the partition number. This implementation aligns with mainline U-Boot conventions for the SpacemiT platform and has been verified to work correctly by the maintainer.

Applied to files:

patch/kernel/archive/rockchip64-6.19/rk3588-1220-arm64-dts-rockchip-describe-pcie-ethernets-on-FriendlyElec-NanoPC-T6.patch

📚 Learning: 2025-12-17T15:46:22.709Z

Learnt from: EvilOlaf
Repo: armbian/build PR: 9096
File: patch/kernel/archive/sunxi-6.18/patches.armbian/arm64-dts-sun50i-h616-add-overlays.patch:4-5
Timestamp: 2025-12-17T15:46:22.709Z
Learning: In Armbian kernel patches, when patches are renamed or reorganized, the old Subject line may be intentionally preserved in the patch header (as a second Subject line) for archival/historical purposes to document the previous patch name. This is an Armbian-specific practice for tracking patch evolution and should not be flagged as an error when the maintainer has intentionally included it for documentation.

Applied to files:

patch/kernel/archive/rockchip64-6.19/rk3588-1220-arm64-dts-rockchip-describe-pcie-ethernets-on-FriendlyElec-NanoPC-T6.patch

📚 Learning: 2026-01-03T07:27:45.955Z

Learnt from: iav
Repo: armbian/build PR: 9158
File: patch/kernel/archive/rockchip64-6.18/add-board-helios64.patch:598-600
Timestamp: 2026-01-03T07:27:45.955Z
Learning: In Rockchip vendor kernel branches (e.g., rockchip64-6.18), the FUSB302 device tree binding accepts usb-role-switch as a phandle reference to a USB role switch provider (e.g., typec_extcon_bridge), not just as a boolean flag as in mainline kernel bindings. The vendor implementation allows usb-role-switch = <&typec_extcon_bridge>; to work correctly.

Applied to files:

patch/kernel/archive/rockchip64-6.19/rk3588-1220-arm64-dts-rockchip-describe-pcie-ethernets-on-FriendlyElec-NanoPC-T6.patch

📚 Learning: 2025-09-14T06:32:29.806Z

Learnt from: amazingfate
Repo: armbian/build PR: 8619
File: config/sources/families/rockchip.conf:222-230
Timestamp: 2025-09-14T06:32:29.806Z
Learning: In the Armbian build system, the write_uboot_platform() function implementations follow different patterns across Rockchip family files. The newer standard (used in rockchip64_common.inc and rk3506) includes 'status=none' parameter in dd commands, while older implementations (rk3288, rk322x) use an older pattern without this parameter. The rk3506 implementation correctly follows the current Rockchip family standard.

Applied to files:

patch/kernel/archive/rockchip64-6.19/rk3588-1220-arm64-dts-rockchip-describe-pcie-ethernets-on-FriendlyElec-NanoPC-T6.patch

📚 Learning: 2026-01-10T04:28:58.758Z

Learnt from: kasimling
Repo: armbian/build PR: 9177
File: patch/kernel/archive/rockchip64-6.18/rk3588-1212-arm64-dts-Automatic-fan-speed-and-USB-3.0-Type-A-por.patch:37-41
Timestamp: 2026-01-10T04:28:58.758Z
Learning: When reviewing Linux kernel device-tree bindings involving thermal cooling devices, THERMAL_NO_LIMIT can be used for both min_state (lower) and max_state (upper). If used as lower/min_state, the framework converts it to 0; if used as upper/max_state, it converts to the cooling device's max_state. Verify patch entries like: cooling-device = <&fan THERMAL_NO_LIMIT 1> are valid and correctly reflect the intended bounds. Ensure consistent interpretation of NO_LIMIT in both min and max contexts and that the resulting values align with the device's actual limits.

Applied to files:

patch/kernel/archive/rockchip64-6.19/rk3588-1220-arm64-dts-rockchip-describe-pcie-ethernets-on-FriendlyElec-NanoPC-T6.patch

🔇 Additional comments (3)

patch/kernel/archive/rockchip64-6.19/rk3588-1220-arm64-dts-rockchip-describe-pcie-ethernets-on-FriendlyElec-NanoPC-T6.patch (3)

27-28: LGTM!

The ethernet aliases correctly reference the labeled nodes defined below, enabling U-Boot's fdt_fixup_ethernet() to discover and patch these nodes with MAC addresses from the environment (ethaddr, eth1addr). Based on learnings, this is the preferred upstream-friendly approach.

37-52: Well-structured PCIe ethernet binding.

The PCIe bridge and endpoint hierarchy follows standard PCI device tree bindings:

The reg values correctly encode bus/device/function addresses in the phys.hi cell

The compatible = "pci10ec,8125" uses the proper PCI vendor:device ID format (Realtek 10ec, RTL8125)

The mac-address placeholder allows U-Boot's fdt_fixup_ethernet() to inject the runtime MAC

The boot logs in PR comments confirm this structure works correctly with the r8169 driver.

61-76: LGTM!

The second PCIe ethernet node correctly mirrors the first, with appropriately distinct bus address ranges (0x40-0x4f vs 0x20-0x2f) for the separate PCIe slot (pcie2x1l2). The consistent structure ensures both NICs are handled identically by U-Boot and the kernel.

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

rpardini · 2026-01-12T23:37:36Z

## Flattened Device Tree blob at 12000000
   Booting using the fdt blob at 0x12000000
Working FDT set to 12000000
   Loading Ramdisk to eac3f000, end ec7ffbba ... OK
   Loading Device Tree to 00000000ecb14000, end 00000000ecbabfff ... OK
Working FDT set to ecb14000
[image_setup_libfdt] called
[image_setup_libfdt] calling fdt_fixup_ethernet
[fdt_fixup_ethernet] called
<... snip ...>
[fdt_fixup_ethernet] alias #31: name='ethernet0', path='/pcie@fe180000/pcie@0,0/ethernet@0,0'
[fdt_fixup_ethernet] env var for alias 'ethernet0' is 'ethaddr', value='8e:b4:90:97:32:30'
[fdt_fixup_ethernet] Patching node '/pcie@fe180000/pcie@0,0/ethernet@0,0' (offset 39588) with MAC 8e:b4:90:97:32:30
[fdt_fixup_ethernet] alias #32: name='ethernet1', path='/pcie@fe190000/pcie@0,0/ethernet@0,0'
[fdt_fixup_ethernet] env var for alias 'ethernet1' is 'eth1addr', value='8e:b4:90:97:32:31'
[fdt_fixup_ethernet] Patching node '/pcie@fe190000/pcie@0,0/ethernet@0,0' (offset 41100) with MAC 8e:b4:90:97:32:31
[fdt_fixup_ethernet] SUMMARY: aliases found=33, attempted=2, skipped=0, patched=2

Starting kernel ...

[    0.000000] Booting Linux on physical CPU 0x0000000000 [0x412fd050]
[    0.000000] Linux version 6.19.0-rc4-edge-rockchip64 (build@armbian) (aarch64-linux-gnu-gcc (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0, GNU ld (GNU Binutils for Ubuntu) 2.42) #5 SMP PREEMPT Sun Jan  4 22:41:55 UTC 2026
<...snip...>
[    2.633897] r8169 0002:21:00.0: enabling device (0000 -> 0003)
[    2.687792] r8169 0002:21:00.0 eth0: RTL8125B, 8e:b4:90:97:32:30, XID 641, IRQ 118
[    2.688466] r8169 0002:21:00.0 eth0: jumbo features [frames: 16362 bytes, tx checksumming: ko]
[    2.689450] r8169 0004:41:00.0: enabling device (0000 -> 0003)
[    2.713689] r8169 0004:41:00.0 eth1: RTL8125B, 8e:b4:90:97:32:31, XID 641, IRQ 119
[    2.714364] r8169 0004:41:00.0 eth1: jumbo features [frames: 16362 bytes, tx checksumming: ko]

🎉

…endlyElec NanoPC-T6 - By fully describing the pcie hardware for the RTL8125s (rtl8169) present on the board, we get to alias them to ethernet0/1. In return, (mainline) u-boot will by default patch the runtime FDT though it's fdt_fixup_ethernet() implementation, hopefully resulting in fully stable MAC addresses without the need to program the RTL NIC's EEPROM; as an added benefit, u-boot MAC & kernel MAC would match, facilitating network-boot setups.

github-actions · 2026-01-13T00:33:43Z

✅ This PR has been reviewed and approved — all set for merge!

github-actions bot added size/medium PR with more then 50 and less then 250 lines 02 Milestone: First quarter release Needs review Seeking for review Hardware Hardware related like kernel, U-Boot, ... Patches Patches related to kernel, U-Boot, ... labels Jan 12, 2026

rpardini force-pushed the pr/rockchip64-619-arm64-dts-rockchip-describe-pcie-ethernets-on-FriendlyElec-NanoPC-T6 branch from dea58b2 to bf4d994 Compare January 12, 2026 23:45

rpardini marked this pull request as ready for review January 12, 2026 23:46

rpardini requested review from ColorfulRhino, HeyMeco, JohnTheCoolingFan, SuperKali, TheSnowfield, Tonymac32, ahoneybun, amazingfate, brentr, catalinii, chainsx, efectn, igorpecovnik, joekhoobyar, krachlatte, lanefu, paolosabatino, prahal, pyavitz and schwar3kat as code owners January 12, 2026 23:46

igorpecovnik approved these changes Jan 13, 2026

View reviewed changes

github-actions bot added the Ready to merge Reviewed, tested and ready for merge label Jan 13, 2026

github-actions bot removed the Needs review Seeking for review label Jan 13, 2026

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Uh oh!

rockchip64-6.19: arm64: dts: rockchip: describe pcie ethernets on FriendlyElec NanoPC-T6 #9238

rockchip64-6.19: arm64: dts: rockchip: describe pcie ethernets on FriendlyElec NanoPC-T6 #9238

rpardini commented Jan 12, 2026 •

edited by coderabbitai bot

Loading

Uh oh!

coderabbitai bot commented Jan 12, 2026 •

edited

Loading

Walkthrough

Changes

Estimated code review effort

Possibly related PRs

Suggested reviewers

Poem

Uh oh!

rpardini commented Jan 12, 2026

Uh oh!

github-actions bot commented Jan 13, 2026

Uh oh!

Reviewers

Assignees

Labels

Milestone

Development

Uh oh!

2 participants

Uh oh!

rockchip64-6.19: arm64: dts: rockchip: describe pcie ethernets on FriendlyElec NanoPC-T6 #9238

Are you sure you want to change the base?

rockchip64-6.19: arm64: dts: rockchip: describe pcie ethernets on FriendlyElec NanoPC-T6 #9238

Conversation

rpardini commented Jan 12, 2026 • edited by coderabbitai bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary by CodeRabbit

Uh oh!

coderabbitai bot commented Jan 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Walkthrough

Changes

Estimated code review effort

Possibly related PRs

Suggested reviewers

Poem

Uh oh!

rpardini commented Jan 12, 2026

Uh oh!

github-actions bot commented Jan 13, 2026

Uh oh!

Reviewers

Assignees

Labels

Milestone

Development

Uh oh!

2 participants

rpardini commented Jan 12, 2026 •

edited by coderabbitai bot

Loading

coderabbitai bot commented Jan 12, 2026 •

edited

Loading