05:08 <naoki> usage of model name ("Model" word) is inconsistent in our docs... we need to fix both docs and sources ;)

05:08 <naoki> ^Radxa

07:58 <arnd> mriesch, mmind00: I see a regression in today's linux-next: arch/arm64/boot/dts/rockchip/rk3568-wolfvision-pf5-display.dtsi:113.6-121.3: Warning (graph_port): /fragment@4/__overlay__: graph port node name should be 'port'

07:59 <mriesch> arnd: oops... let me have a look

08:01 <mmind00> mriesch: ah, the vp2 node at the end

08:03 <mmind00> though I guess in the original rk3568-wolfvision-pf5.dts the check didn't complain about that

08:04 <mmind00> i.e. it's "vp2: port@2" in the rk356x.dtsi and the dtso then references the "&vp2"

08:08 <mriesch> arnd: mmind00: ok i can reproduce it but tbh i have no idea how to fix this. any pointers?

08:09 <mmind00> mriesch: I think that might be more something for the maintainer of the dt-checker how it handles overlays? ... Because that issue seems very much overlay-specific

09:10 <mmind00> mriesch: though in the short term, I guess something like https://paste.debian.net/1314435/ also makes the dtbscheck happy

09:11 <mmind00> because that really seems to be a check "oh an endpoint, the parent needs to be named port"

10:18 <mriesch> mmind00: thanks for the hint.

10:19 <mriesch> arnd: mmind00: i'm busy with putting out other fires, but i'll put a fix on my todo list. if i am feeling really lucky, i'll try to trigger a discussion whether the check makes sense as is