05:46 <DC-IRC> [Discord] <igorpec> there have been some rk3399 related fixes, but we would need a dedicated maintainer for PBpro to verify if there is a change

06:01 <DC-IRC> [Discord] <wsy1> I'll try out one of the new builds tonight, thanks!

14:15 <DC-IRC> [Discord] <ahoneybun> @wsy1 I can test some images as well if you want, we can take turns.

17:09 <DC-IRC> [Discord] <jonh7776> What is the origination of the rk-6.1-rkr3 branch in Armbian's linux-rockchip repo? I originally thought it was a fork of the same branch from the Radxa tree like the rk-6.1-rkr1 but that repo has no such branch.I am still having trouble getting 5120x1440 to work. I am suspecting changed made in rockchip_drm_vop2.c although that is just really a guess at this point. My experiment <clipped message>

17:09 <DC-IRC> [Discord] <jonh7776> s have been unable to prove that out yet. It could still be the hdmi-8k patch although I can't see why that would be incompatible between the rkr1 and rkr3 kernels.

17:10 <DC-IRC> [Discord] <jonh7776> What is the origination of the rk-6.1-rkr3 branch in Armbian's linux-rockchip repo? I originally thought it was a fork of the same branch from the Radxa tree like the rk-6.1-rkr1 but that repo has no such branch.

17:10 <DC-IRC> [Discord] <jonh7776> I am still having trouble getting 5120x1440 to work. I am suspecting changed made in rockchip_drm_vop2.c although that is just really a guess at this point. My experiments have been unable to prove that out yet. It could still be the hdmi-8k patch although I can't see why that would be incompatible between the rkr1 and rkr3 kernels.

17:20 <DC-IRC> [Discord] <spooky8086> rk-6.1-rkr3 comes from Rockchip's latest SDK release. I recall seeing a lot of VOP2 changes when I was reviewing the kernel. It's possibe that 8k has broken, or changes are needed to the 8k overlay now.

17:20 <DC-IRC> [Discord] <jonh7776> p.s. The reason 5120x1440 doesn't work on the rkr3 kernel is it fails to read the DisplayID Extension Block (Block 2) which is where 5120x1440 modes are defined. That may be because vp0 is not selected in the VOP2 HW but in my examination of the device tree it does appear that vp0 is properly selected. Still learning about this area so I am most likely missing something.

17:21 <DC-IRC> [Discord] <jonh7776> @amazingfate had the same suggestion on the 8k overlay. I am still looking at that also.

17:22 <DC-IRC> [Discord] <jonh7776> And yes, there are a lot of changes to the vop2 source code. Sifting through what could be relevant or not is made more challenging by that. 😦

18:49 <DC-IRC> [Discord] <jonh7776> The "not readying of the DisplayID block" was a boondoggle. It is ready. I'm back to looking at filtering of modes in the vop2 driver.

18:49 <DC-IRC> [Discord] <jonh7776> I'll keep looking but the 8k overlay looks to be right on the -75 kernel so far as I can tell.