martinayotte

You need to look first what is logged in this journal, because on most Armbian image journal is even turned off, not producing any /var/log/journal ...

I must admit that I've never used this library, and therefore I don't know how to fix it. But instead I've used this one https://github.com/duxingkei33/orangepi_PC_gpio_pyH3 I've tweaked it to support 64bits pointers, like I did previously for H5, and then adjusted the SW_PORTC_IO_BASE define to point to H6 registers at 0x0300B000..

He means that you can fork this github https://github.com/armbian/build , do a "git clone" to your local PC, do the appropriate change to the kernel config, commit it into your fork, and then send a PullRequest to Armbian github.

That would be better ... But still, according to number you provided, 29.8G won't fit into 29.1G ...

http://wiki.banana-pi.org/Getting_Started_with_R64

You can't copy a single partition into a whole device without partition table ...

You did a typo here , it should be "dd if=/dev/zero of=/mnt/512mbfile.bin bs=1024 count=512000" I presume ...

It is only require 2 or 3 secs been shorted while pressing the "reset" push button SW2103 near the eMMC. As I said to @Igor , those small pads, TP50265 and GND nearby are so small (yes, located on the top side), I took a tweezers and press hard to make sure contact was good, and then gently press "reset". EDIT : I've tried to learn what those other pads on the bottom side are doing, I can't really tell, and the youtube clip is in chinese, which I'm not fluent with it (I don't know a single word) ...

Of course @guidol will say YES ! (but I will leave him to say it by himself ...)

What brings you that "U-Boot is broken." ? Many people here are trying to help you, but you're jumping to wrong conclusion ... If you don't want to increase verbosity, which is quite simple task, you don't even know why causing the kernel reboot in loop during this famous 44secs . I bet it is a UUID issue !!! If you don't do what we are asking for, no one will be able to help you !!! Simple is that !!!

As I said to @Igor last Friday, this DDR Version signature is the one found in the factory android image ! So, maybe your shortening wasn't good enough and for sure was still booting from eMMC the old Android which then stop seeing that SDCard is inserted but not compatible ... With the current build, or at least the one I did 2 weeks ago, you should get this DDR signature : DDR Version 1.14 20180803 In soft reset SRX Channel 0: DDR3, 800MHz Bus Width=32 Col=10 Bank=8 Row=15 CS=1 Die Bus-Width=16 Size=1024MB Channel 1: DDR3, 800MHz Bus Width=32 Col=10 Bank=8 Row=15 CS=1 Die Bus-Width=16 Size=1024MB

Ah ! Ok ! Yes, I saw this commit earlier ... So some IRQ issue ...

The clm_blob is not the firmware itself, it is some blob to determine the country, therefore the allowed channels. The firmware is still loaded as we see "Firmware: BCM43362/1 wl0: Oct 23 2017 10:33:17 version 5.90.240 FWID 01-0". So, maybe the problem is elsewhere ... Did both of you did a network scan with "nmtui" ? In my case, with completely different board, an bananapim2ultra with brcmfmac43430-sdio, I've all the same "dmesg" traces, but WiFi is working fine : [ 7.674071] brcmfmac: brcmf_fw_alloc_request: using brcm/brcmfmac43430-sdio for chip BCM43430/1 [ 7.811280] brcmfmac: brcmf_fw_alloc_request: using brcm/brcmfmac43430-sdio for chip BCM43430/1 [ 7.811367] sdio mmc1:0001:1: Direct firmware load for brcm/brcmfmac43430-sdio.clm_blob failed with error -2 [ 7.811375] sdio mmc1:0001:1: Falling back to syfs fallback for: brcm/brcmfmac43430-sdio.clm_blob [ 9.041310] brcmfmac: brcmf_c_process_clm_blob: no clm_blob available (err=-11), device may have limited channels available [ 9.045735] brcmfmac: brcmf_c_preinit_dcmds: Firmware: BCM43430/1 wl0: Mar 30 2016 11:30:56 version 7.45.77.h8.4 FWID 01-ee8a6268

Why are you mount this partition using loop device ? It could be mounted directly using : mount /dev/mmcblk0p14 /mnt

"i2cdetect -y 0" should be almost instantenous, left than half second for the whole 127 addresses ! If it takes about 1 seconds per address, this means you pullups are defective, or they are placed on the wrong pins.

I hope you don't use the same method used for installing eMMC using a Non-Armbian image with Armbian one which is "nand-sata-install". As I said earlier, on my OPiZeroPlus2-H5, I've installed Armbian on eMMC successfully several times since years and last time was 2 weeks ago with 4.20.y. Did you have done what I suggested, first : Then, for the fourth time, did you have done : If really Yes, what was the results ? When rebooting without SDCard inserted, do you see any u-boot activity ? Is it starting booting the kernel ? if Yes, what is the output just before it abort and reboot in loop ? Does it says that it can't find RootFS with proper UUID ? Did you check if UUID are properly set in both /boot/armbianEnv.txt and /etc/fstab of the eMMC (not the one from SDCard) ? None of the above, since it works for all of us, and you seems to be the only one to get such failure ...

Maybe, but you are mixing symptoms and issues and refuse to do what we suggested ... As I said twice, and now a third time, did you do : If the RootFS on eMMC is Ok, and only refuse to boot directly from eMMC, then it is U-Boot that needs to be recovered ...

Yes, since there are no ones onboard ... Don't know ... Could be that you mistakenly crossed SDA and SCL ?

No ! As @Igor said, you should get some USB-TTL Serial dongle to do some serious debugging. Maybe you didn't short TP50265 properly during u-boot phase.

Always ! Already did, and audio no working there too ... Looking quickly at Vasily and Wens patches, and comparing with Armbian, it seems that there are some missing power regulator.

It is and was always worked successfully on my side, and not only on my OPiZeroPlus2-H5, both on all my boards ...

Doesn't seems to work on Pinebook with Armbian 4.20.y. I will probably check with an Ayufan image later ...

This file is installed by "apt-get install python-dev" ...

Right ! I didn't pay attention to which SPI it was connected ...

It is not implemented either in Armbian overlays, so you would need to decompile/recompile the spidev overlay manually ...