I found the issue, but I'm not sure how to solve it yet. It's extremely weird. U-Boot is not built correctly, that much is obvious. However, reason why is curious. If I put .git folder into source folder, then it builds correctly. It seems that some tool require git metadata to decide something. I have to find which tool causes that.
This is second weird build bug that I'm researching in last week or so...
Ok, so there is some issue with mainline U-Boot. R40 is so rare that nobody raised issue before. I guess some bisecting is in order...
leoneii Issue is solved by kernel update in nightly v12.2 images.
Fix: https://lore.kernel.org/linux-sound/87…[email protected]/
No. Display driver is being worked on, so it won't be supported for some time.
Hm... Not sure what could be wrong, but most likely U-Boot related. Since I don't have time to look into this anytime soon, can you make following experiment. Flash 12.2 image, extract first 1 MB from 12.0 image, remove first 8 KB of that 1 MB dump and flash resulting binary to 12.2 image at offset 8 KB. I hope you understand 
otherwise I can provide set of dd commands to do that.
Actually there is a way to support this chip with mainline driver. My most recent attempt (4 years old) is here:
Wifi worked but suspend was broken, which makes it ineligible to include it to LE. Any efforts on this front would be appreciated 
Sure, but it's implemented if I understood the OP right:
You did not. VP9 HW decoding is supported only by H6. I guess compatibility matrix would be clearer.
We really need this project to continue. There is a bunch of h616/h618 devices (tv boxes) out there that are still waiting for official support.
Yeah, that won't happen anytime soon. Display driver accumulated so much technical debt that it needs to be considerably reworked in order to support H616 and newer SoCs properly. While obviously there are PoC versions out there, mostly based on my early work, none of that works good enough for Kodi needs (Kodi needs features most other software doesn't need). Videos wouldn't be rendered properly in plenty of cases, so this forum would be spammed with a lot of "it doesn't work" messages.
I started working on this redesign of display driver, but it's slow process, since I work on it only in my spare time. However, good thing is that redesign will also solve years old issues for existing devices like occasional glitches, especially during start or end of playback.
No, only mainline U-Boot is supported (currently 2025.1.) If your board is already supported, great, all you need to do is to set correct config name in uboot_helper script. If not, you have to create patch for U-Boot, which introduces correct defconfig for your board and then put it in uboot_helper script.
you have to provide DRAM settings, which you can get in Android DT file from vendor image.
This means that you didn't select correct U-Boot config. U-Boot has to be configured for each board separately, due to different settings, like DRAM (as you already experienced). Board configuration is done here: https://github.com/LibreELEC/Libr…ts/uboot_helper
They use same silicon, they are just named differently, because (i think) they differ in performance. Yes, you can use same code.
If you want to play, here is my attempt: https://github.com/jernejsk/LibreELEC.tv/commits/h616/ but note that display driver is not there yet (problematic scaling of YUV formats). There is also no audio support. Using newer kernel would allow analog audio and if you use extra patches, also hdmi audio.
H616 and H313 are basically same soc, so no. While I worked on H616 support, it doesn't work well due to display driver issues. Currently best supported Allwinner soc is H6.
H313 SoC is not supported by LE, so no, it won't work.
I only have ordinary Orange Pi Win board and everything seems to work fine with LE 12.0.1.
If you have serial console, please add systemd.debug-shell=ttyS0 ssh at the end of APPEND line and remove quiet in extlinux/extlinux.conf. This should print bunch of lines which should help to narrow problem down.
ilchenko Can you reproduce the issue? Patch https://github.com/Pulse-Eight/li…11cf637e3.patch from https://github.com/Pulse-Eight/libcec/pull/616 would probably help.
