lanefu changed the topic of #armbian-rockchip to: Armbian - Linux for ARM development boards | Rockchip SoC | www.armbian.com | This channel is relayed to the equivalent Discord channel | this channel is logged
montjoie has quit [Ping timeout: 255 seconds]
montjoie has joined #armbian-rockchip
DC-IRC has quit [Remote host closed the connection]
DC-IRC has joined #armbian-rockchip
juri_ has quit [Ping timeout: 255 seconds]
<DC-IRC>
<marty1938> thanks! Yes i think its most probably due to kernel dts missmatch. The E25 conf BOOT_FDT_FILE="rockchip/rk3568-radxa-e25.dtb" is referring to the one in mainline?
<DC-IRC>
<rpardini> Depends, get `uname -a` -- if it's 5.10.160 it's vendor rk kernel
<DC-IRC>
<marty1938> Nah i run the 6.1.46
<DC-IRC>
<marty1938> Nah i run the 6.1.46 so i guess that refering to current
<DC-IRC>
<rpardini> Oh ok -- and what's the version of the vendor overlay's kernel?
<DC-IRC>
<marty1938> They are using an old one 4.19.193-67-rockchip
<DC-IRC>
<rpardini> bingo, that'll never work.
<DC-IRC>
<marty1938> yes i think i was a bit naive, "what could have changed its just a sdio and some uart"
<DC-IRC>
<marty1938> What i need to do is to write a new dts to generate an overlay based on 6.1?
<DC-IRC>
<rpardini> vendor (rockchip) kernel and mainline are most different.
<DC-IRC>
<rpardini> with some effort, you can find the original overlay dts, reason about it, and write a mainline equivalent dtso
<DC-IRC>
<rpardini> I would recommend using 6.6 (armbian's `edge` kernel)
<DC-IRC>
<marty1938> thanks! ok, reason using 6.6 is to make it easier to get it into mainline later?
<DC-IRC>
<rpardini> overlays won't make it into mainline AFAIK.
<DC-IRC>
<rpardini> no, it's because our 6.6 branch is much easier to work with overlays, essentially drop them in the folder and add to Makefile, no patching required.
<DC-IRC>
<Tonymac32> Yeah overlays will never be packaged in mainline because the act of loading an overlay at run time results in a race condition.
<DC-IRC>
<Tonymac32> There was a huge discussion about this in the past in the mailing list