Page 1 of 1

Сustom board based on RV1106G3 not booting past "Starting kernel..." (copied from Luckfox Pro Max)

Posted: 2025-06-24 8:53
by pndgaga
Hi,

We’ve assembled a custom board based on the RV1106G3 SoC. The schematic and layout for the SoC were copied exactly from the Luckfox Pro Max datasheet. We're using the same SD card image (Buildroot system) that successfully runs on the official Luckfox Pro Max dev board.

On our custom board, the bootlog is almost identical to the one from the dev board. However, it gets stuck right after the Starting kernel ... message — the kernel never seems to start.

Here is what I’ve tried so far:

Verified UART switching and pinmux.

Inserted early debug messages at the beginning of the kernel startup code — they do not appear, which suggests that the CPU never enters the kernel properly.

Below is the full boot log. As you can see, everything looks normal up to the kernel loading phase:

Code: Select all

DDR V1.10 ec2fae0c96 wesley.yao 22/11/15-10:58:09
S5P1
f967
rgef1
DDRConf2
DDR3, BW=16 Col=10 Bk=8 CS0 Row=14 CS=1 Die BW=16 Size=256MB
924MHz
DDR bin out

U-Boot SPL board init
U-Boot SPL 2017.09 (Jul 10 2024 - 13:57:10)
unrecognized JEDEC id bytes: 00, 00, 00
unknown raw ID 0 0 0
Trying to boot from MMC2
ENVF: Primary 0x00000000 - 0x00008000
ENVF: OK
Trying fit image at 0x440 sector
## Verified-boot: 0
## Checking uboot 0x00200000 (lzma @0x00400000) ... sha256(a3f28ca904...) + sha256(fc01dd7588...) + OK
## Checking fdt 0x0025bed0 ... sha256(51bba657c3...) + OK
Total: 228.864 ms

Jumping to U-Boot(0x00200000)


U-Boot 2017.09 (Jul 10 2024 - 13:57:10 +0800)

Model: Rockchip RV1106 EVB Board
PreSerial: 2, raw, 0xff4c0000
DRAM:  256 MiB
Sysmem: init
Relocation Offset: 0fd85000
Relocation fdt: 0edfacb0 - 0edfede0
CR: M/C/I
Using default environment

no mmc device at slot 1
ENVF: Primary 0x00000000 - 0x00008000
ENVF: OK
mmc@ffa90000: 0, mmc@ffaa0000: 1 (SD)
Bootdev(atags): mmc 1
MMC1: Legacy, 52Mhz
PartType: ENV
DM: v2
No misc partition
boot mode: None
FIT: no signed, no conf required
DTB: rk-kernel.dtb
HASH(c): OK
Model: Luckfox Pico Max
CLK: (sync kernel. arm: enter 816000 KHz, init 816000 KHz, kernel 0N/A)
  apll 816000 KHz
  dpll 924000 KHz
  gpll 1188000 KHz
  cpll 1000000 KHz
  aclk_peri_root 400000 KHz
  hclK_peri_root 200000 KHz
  pclk_peri_root 100000 KHz
  aclk_bus_root 300000 KHz
  pclk_top_root 100000 KHz
  pclk_pmu_root 100000 KHz
  hclk_pmu_root 200000 KHz
No misc partition
## retrieving sd_update.txt ...
bad MBR sector signature 0x0000
 Invalid partition 1 
Net:   eth0: ethernet@ffa80000
Hit key to stop autoboot('CTRL+C'):  0
## Booting FIT Image at 0xe90d680 with size 0x002ec600
Fdt Ramdisk skip relocation
No misc partition
## Loading kernel from FIT Image at 0e90d680 ...
   Using 'conf' configuration
## Verified-boot: 0
   Trying 'kernel' kernel subimage
     Description:  unavailable
     Type:         Kernel Image
     Compression:  uncompressed
     Data Start:   0x0e916a80
     Data Size:    3027200 Bytes = 2.9 MiB
     Architecture: ARM
     OS:           Linux
     Load Address: 0x00008000
     Entry Point:  0x00008000
     Hash algo:    sha256
     Hash value:   66929aa0a4d3f91d2b47fa6dae6865e75d0a7a606cda288fb10179af657ecf17
   Verifying Hash Integrity ... sha256+ OK
## Loading fdt from FIT Image at 0e90d680 ...
   Using 'conf' configuration
   Trying 'fdt' fdt subimage
     Description:  unavailable
     Type:         Flat Device Tree
     Compression:  uncompressed
     Data Start:   0x0e90de80
     Data Size:    35369 Bytes = 34.5 KiB
     Architecture: ARM
     Load Address: 0x00c00000
     Hash algo:    sha256
     Hash value:   5bbf3ddfda5c9b1853ee5b6c19a8fd387a5d0fdb15aee1f921a80a7e33bab5eb
   Verifying Hash Integrity ... sha256+ OK
   Loading fdt from 0x0e90d680 to 0x00c00000
   Booting using the fdt blob at 0x00c00000
   Loading Kernel Image from 0x0e916a80 to 0x00008000 ... OK
   kernel loaded at 0x00008000, end = 0x002eb100
   Using Device Tree in place at 00c00000, end 00c0ba28
Adding bank: 0x00000000 - 0x10000000 (size: 0x10000000)
Total: 646.464 ms

Starting kernel ...


Do you have any suggestions on what could be causing the kernel not to start? Are there any hardware traps, bootstraps, or PMIC/GPIO differences I might have missed that are critical for kernel boot?

Any help or insights would be greatly appreciated!

Thanks in advance!

Re: Сustom board based on RV1106G3 not booting past "Starting kernel..." (copied from Luckfox Pro Max)

Posted: 2025-06-25 2:23
by Crocodile
Hello, we do not offer technical support for custom boards. I suggest you check the power supply issue. When loading the kernel and allocating memory, the power consumption will increase momentarily.

Re: Сustom board based on RV1106G3 not booting past "Starting kernel..." (copied from Luckfox Pro Max)

Posted: 2025-06-25 6:53
by pndgaga
Crocodile wrote: 2025-06-25 2:23 Hello, we do not offer technical support for custom boards. I suggest you check the power supply issue. When loading the kernel and allocating memory, the power consumption will increase momentarily.

Thank you for your response.

We’ve checked the power rails during kernel load, and the supply voltages remain stable — there’s no noticeable drop or brownout when the kernel is being loaded. So at this point, we believe power delivery is not the issue.

Previously, I’ve also tried modifying the Device Tree to match our board more closely, but unfortunately that didn’t help — the kernel still hangs right after the “Starting kernel...” message.

Given this, I’d like to ask for your advice:
Would you recommend focusing further efforts on debugging the software side (e.g. Device Tree, kernel configuration, bootloader), or should we revisit possible hardware-level differences — such as GPIO strap pins, PMIC initialization sequences, or other early-stage boot dependencies?

Any pointers would be greatly appreciated.

Thanks again!

Re: Сustom board based on RV1106G3 not booting past "Starting kernel..." (copied from Luckfox Pro Max)

Posted: 2025-06-26 1:28
by Crocodile
As mentioned in the previous response, we do not offer technical support for the hardware design. Incorrect troubleshooting approaches can also delay your development progress. So I have no more suggestions to offer.