New Radxa Zero 3E Halts during boot, garbled text on UART?

I have used UART via USB on many SBCs for well on 12 years or more. So familiar with using UART for console access. Have tested by USB to UART cable/breakout board with Pi Zero 2W for example, works fine.

But with this new Zero 3E, even with baud rate 921600, 8,1,0,0, nothing but garbled text. Connected HDMI cable to monitor, the Zero 3E boots, reports rootfs clean, then NOTHING. No output, LED on board goes out. Guess the board is halting or crashing?

I tried the bullseye image, flashed via balenaEtcher. Apparently the only supported or official image.

357252575-ead73fb5-ec43-4289-8aa1-7ee230bceab1

Unfortunately I don’t have a way to capture the HDMI output.

Any suggestions? Right now looks or feel like I got a lemon?

Tried multiple power sources. Tried all 4 images bullseys CLI, bullseye XFCE, Ubuntu CLI, Ubuntu XFCE (jammy), none boot. Behavior is consistently, power up, nothing on HDMI monitor, PuTTy console session just garbled text.

Scanning the forum, it seems there is some discussion, about the official image(s) are broken? Well, unless anyone has some suggestions, this SBC is going back to the seller. And the other Radxa board, a Zero 3W, I planned to order, is not happening. Disappointment, thus far, really liked the specification of these boards.

I prefer using these images: https://joshua-riek.github.io/ubuntu-rockchip-download/boards/radxa-zero3.html

Rockchip SBCs uses 1.5M baud in default (1.500.000 baud 8N1)

Interesting… the web page, I found clearly stated 921600 which seemed odd, not a number I would just use without some direction from the Radxa site. Did the web page changed in the last couple of days? I guess I should have screen printed it.

Just to confirm the support OS images are at the following URL right?

Still have an issue, the image below, will not boot, tried 3 different SD cards…
radxa-zero3_debian_bullseye_cli_b3.img.xz

Starting kernel ...

[   11.561571] fiq_debugger fiq_debugger.0: IRQ fiq not found
[   11.561640] fiq_debugger fiq_debugger.0: IRQ wakeup not found
[   11.561675] fiq_debugger_probe: could not install nmi irq handler
[   13.043042] mpp-iep2 fdef0000.iep: allocate roi buffer failed
[   13.063561] arm-scmi firmware:scmi: Failed. SCMI protocol 22 not active.
[   13.086158] dwhdmi-rockchip fe0a0000.hdmi: IRQ index 1 not found
[   13.089762] no ATF memory for init
[   13.110915] no ATF memory for init
[   13.123005] rk_gmac-dwmac fe010000.ethernet: no regulator found
[   13.522014] rk817-battery rk817-battery: Failed to find matching dt id
[   13.522449] rk817-charger rk817-charger: Failed to find matching dt id
[   13.533309] OF: graph: no port node found in /i2c@fe5c0000/fusb302@22
[   13.566055] arm-scmi firmware:scmi: Failed. SCMI protocol 17 not active.
[   13.573222] rksfc_base v1.1 2016-01-08
[   13.619454] mpp_rkvenc fdf40000.rkvenc: Failed to get leakage
[   13.620045] mpp_rkvenc fdf40000.rkvenc: failed to find power_model node
[   13.620066] mpp_rkvenc fdf40000.rkvenc: failed to initialize power model
[   13.620082] mpp_rkvenc fdf40000.rkvenc: failed to get dynamic-coefficient
[   13.621865] mpp_rkvdec2 fdf80200.rkvdec: shared_niu_a is not found!
[   13.621894] rkvdec2_init:1022: No niu aclk reset resource define
[   13.621910] mpp_rkvdec2 fdf80200.rkvdec: shared_niu_h is not found!
[   13.621923] rkvdec2_init:1025: No niu hclk reset resource define
[   13.622838] mpp_rkvdec2 fdf80200.rkvdec: failed to find power_model node
[   13.622860] mpp_rkvdec2 fdf80200.rkvdec: failed to initialize power model
[   13.622874] mpp_rkvdec2 fdf80200.rkvdec: failed to get dynamic-coefficient
[   13.626206] rockchip-dmc dmc: failed to get vop pn to msch rl
[   13.626519] rockchip-dmc dmc: could not find power_model node
[   14.197748] mmc1: delay init for 600 ms to enable UHS mode

Board LED turns off, this appears to be a halt or freeze. Trying another image from the same source above.

This image, per below, did actually complete boot!
radxa-zero3_debian_bullseye_xfce_b3.img.xz

I thought I saw a reference some where in this forum that the bullseye CLI image is broken? Unfortunately, I don’t need the GUI for my use case, so junk in the trunk for now.

For those that end up searching for the default id and password for the image above, ‘radxa’ for both id and password.

However, my SBC drops into the debug mode? While I try to do anything that uses the Ethernet interface? On the Zero 3E I am testing.

# 
Welcome to irq debugger mode
Enter ? to get command help
debug>

Also came across this…

# apt --fix-broken install
Reading package lists... Done
Building dependency tree... Done
Reading state information... Done
Correcting dependencies... Done
The following package was automatically installed and is no longer required:
  yq
Use 'apt autoremove' to remove it.
The following additional packages will be installed:
  radxa-system-config-kernel-cmdline-ttyfiq0
The following NEW packages will be installed:
  radxa-system-config-kernel-cmdline-ttyfiq0
0 upgraded, 1 newly installed, 0 to remove and 5 not upgraded.
170 not fully installed or removed.
Need to get 0 B/6,344 B of archives.
After this operation, 19.5 kB of additional disk space will be used.
Do you want to continue? [Y/n]

I was trying to stay within Debian proper… but I am getting all kinds of errors just trying to run ‘apt update’ on the existing bullseye images. Thanks for you link information, looks like I will have to use these.

Always nice when ‘apt update’ and ‘apt upgrade’ are successful… and current!
ubuntu-24.04-preinstalled-server-arm64-radxa-zero3.img.xz

1 Like

@grafick, so you don’t use ‘resetup’ at all then?

@Jibun-no-Kage Rsetup? Nope. Not for my use cases anyway. I just want to install jellyfin, docker, nginx, things like that. Rsetup might be needed when my zero3w shows up though, to switch the antenna but we’ll see.

Yeah… and once in a while when you need to update the bootloader was also my thinking. Interesting, I have Zero 3E, just got it to evaluate it. And I have a 3W coming in a week or so as well.