Orange Pi Zero 2 v1.5 not booting on kernel 6.1.4

[elliotmatson]

Thanks for everyone's work on this project! 

 

I'm working on a few Orange Pi Zero 2 v1.5's, and they won't finish booting on a few of the most recent images. (Just testing jammy right now to keep things consistent)

 

Boot Successfully:

• Armbian_22.11.3_Orangepizero2_jammy_edge_6.1.4
• Armbian_22.05.3_Orangepizero2_jammy_edge_5.17.11

Do Not Boot Successfully:

• Armbian_23.02.2_Orangepizero2_jammy_edge_6.1.11
• Armbian_22.11.1_Orangepizero2_jammy_edge_6.0.10
• Armbian_22.08.7_Orangepizero2_jammy_edge_5.19.16

 

Typically, when the boot fails, all I get is this over serial with no LED's:

U-Boot SPL 2022.07-armbian (Feb 17 2023 - 23:33:06 +0000)
DRAM: 2048 MiB
Trying to boot from MMC1

But occasionally (once every 10-20 boots) I get this with the red LED:

U-Boot SPL 2022.07-armbian (Feb 17 2023 - 23:33:06 +0000)
DRAM: 1024 MiB
Trying to boot from MMC1
NOTICE:  BL31: v2.8(debug):a84636ad-dirty
NOTICE:  BL31: Built : 23:33:00, Feb 17 2023
NOTICE:  BL31: Detected Allwinner H616 SoC (1823)
NOTICE:  BL31: Found U-Boot DTB at 0x4a0903e0, model: OrangePi Zero2
INFO:    ARM GICv2 driver initialized
INFO:    Configuring SPC Controller
INFO:    PMIC: Probing AXP305 on RSB
INFO:    PMIC: aldo1 voltage: 3.300V
INFO:    PMIC: aldo2 voltage: 3.300V
INFO:    PMIC: aldo3 voltage: 3.300V
INFO:    PMIC: bldo1 voltage: 1.800V
INFO:    PMIC: bldo2 voltage: 1.800V
INFO:    PMIC: dcdcd voltage: 1.500V
INFO:    PMIC: dcdce voltage: 3.300V
INFO:    BL31: Platform setup done
INFO:    BL31: Initializing runtime services
INFO:    BL31: cortex_a53: CPU workaround for 855873 was applied
INFO:    BL31: cortex_a53: CPU workaround for 1530924 was applied
INFO:    PSCI: Suspend is unavailable
INFO:    BL31: Preparing for EL3 exit to normal world
INFO:    Entry point address = 0x4a000000
INFO:    SPSR = 0x3c9
INFO:    Changed devicetree.


U-Boot 2022.07-armbian (Feb 17 2023 - 23:33:06 +0000) Allwinner Technology

CPU:   Allwinner H616 (SUN50I)
Model: OrangePi Zero2
DRAM:  1 GiB
Core:  52 devices, 18 uclasses, devicetree: separate
WDT:   Not starting watchdog@30090a0
MMC:   mmc@4020000: 0
Loading Environment from FAT... Unable to use mmc 0:1...
In:    serial@5000000
Out:   serial@5000000
Err:   serial@5000000
Net:   phy interface8
eth0: ethernet@5020000
Autoboot in 1 seconds, press <Space> to stop
switch to partitions #0, OK
mmc0 is current device
Scanning mmc 0:1...
Found U-Boot script /boot/boot.scr
2669 bytes read in 2 ms (1.3 MiB/s)
## Executing script at 4fc00000
U-boot loaded from SD
Boot script loaded from mmc
131 bytes read in 2 ms (63.5 KiB/s)
No FDT memory address configured. Please configure
the FDT address via "fdt addr <address>" command.
Aborting!
No FDT memory address configured. Please configure
the FDT address via "fdt addr <address>" command.
Aborting!
No FDT memory address configured. Please configure
the FDT address via "fdt addr <address>" command.
Aborting!
No FDT memory address configured. Please configure
the FDT address via "fdt addr <address>" command.
Aborting!
No FDT memory address configured. Please configure
the FDT address via "fdt addr <address>" command.
Aborting!
No FDT memory address configured. Please configure
the FDT address via "fdt addr <address>" command.
Aborting!
No FDT memory address configured. Please configure
the FDT address via "fdt addr <address>" command.
Aborting!
10693326 bytes read in 444 ms (23 MiB/s)
Failed to load '/boot/uImage'
Wrong Image Format for bootm command
ERROR: can't get kernel image!
SCRIPT FAILED: continuing...
No EFI system partition
BootOrder not defined
EFI boot manager: Cannot load any image
ethernet@5020000 Waiting for PHY auto negotiation to complete......... TIMEOUT !
missing environment variable: pxeuuid
Retrieving file: pxelinux.cfg/01-02-00-2d-bc-f3-17
ethernet@5020000 Waiting for PHY auto negotiation to complete......... TIMEOUT !
Retrieving file: pxelinux.cfg/00000000

 

I've read most of the other posts about this board, and I'm curious if there's some kind of difference in the v1.5 HW. It seems like v1.5 seems to have more issues, and there are significant visual differences (v1.5, v1.2). Unfortunately OrangePi only published schematics for v1.3, so I'm not sure how to tell what exactly would have changed. 

 

 

[ALIGMSTEN]

Hello @elliotmatson Welcome.

 

Would you mind to please elaborate on the hardware differences you can see.

3 hours ago, elliotmatson said:

and there are significant visual differences

 

Unfortunately, you are correct, there do not seem to be schematics readily available for newer revisions. 

I have worked on V1.3 boards and have been used to having to power cycle occasionally, this might likely be tied to the current development, mainline, edge(here), and the fact that all is WIP.

 

I did look into the power cycling more thoroughly, and although there are a few 'net crumbs' to be found, nothing indicative that I have yet seen or tried.

 

At the moment I've had stable power on's with latest minimal build. Some of the work I've done in isolation has also improved boots.

 

Spoiler

  ___  ____  _   _____             ____
 / _ \|  _ \(_) |__  /___ _ __ ___|___ \
| | | | |_) | |   / // _ \ '__/ _ \ __) |
| |_| |  __/| |  / /|  __/ | | (_) / __/
 \___/|_|   |_| /____\___|_|  \___/_____|

Welcome to Armbian 23.05.0-trunk Bullseye with Linux 6.1.11-sunxi64

No end-user support: built from trunk

System load:   2%               Up time:       35 min
Memory usage:  9% of 984M       IP:            169.254.5.216
10.0.0.6
Usage of /:    3% of 29G

[ Menu-driven system configuration (beta): sudo apt update && sudo apt install armbian-config ]
 

 

 

Right now am testing cpu-frequency-scaling, I will add that v1 to main 6.1.11 fairly shortly its ready, I am adding proper current sensing to my test bench, for more elaborate stability testing, hopefully resulting in better v2

 

Next will most likely follow the progress with DRAM. I also prefer to work with debian os so not entirely sure if ubuntu has some difference.

 

While this doesn't answer your questions, these are known problems, and do not have definitive solutions anywhere right now.

 

[elliotmatson]

The main hardware differences I see are the RAM moving and a lot less supporting passive components, which makes me wonder if the v1.5 is a "value-engineered" model.

 

Thanks for the update! I should have clarified, since there are builds that work on my boards I'm not necessarily looking for a solution, just trying to provide some helpful (hopefully) datapoints. I'm happy to help test as needed, I have some development experience, but I haven't done any tinkering with kernels and u-boot in quite a while

[ALIGMSTEN]

Thanks,

 

Help for h616 always welcome, there is a small amount of mainline development going on.

 

Seems a shame if that happens to be the vendors objective the latest release.

14 hours ago, elliotmatson said:

"value-engineered"

 

Might be useful to collect photos of the different board versions for comparison, as there is no reference right now (just a thought!)