patch for BPI-R2-Pro fixing io-domains/gmacs is already applied (for 5.19).
had some trouble with usb on 5.18 (5.17 works after adding second gpio regulator). It was caused by grabbing clocks by name. Peter sent a patch using rockchip names too and now usb is working with his series. 5.19 will have usb3 and sata support for r2pro.
Got pcie3 working (with same code as i tried on v00 where is had worked only sometimes) and sent out rfc patchset for wider testing. Sent a v2 (still rfc/rft) with a lane-map suggested in v1
Rtc is visible,but does not respond on driver init.
Hdmi- and analog- Sound seems broken,hp-detection is still not working.
Mt7531 phy driver is working, dsa driver recognizes/initializes switch after making reset optional. traffic works now too after making fixed cpu-port 6 dynamic. Sent put a RFC series to mailinglist for mt7531 driver changes and adding dts node.
I took the latest version of the Armbian image for Station P2 and immediately launched on BPIR2Pro without changes. Immediately works HDMI LAN (first port), HDMI sound, 3.5 analog sound, eMMC, USB, power button.
As the first test, I used the existing ArmbianTV image
that I am developing for Station P2 (with DTB for this model, without special fixes for BPI R2 pro). Everything works with it as I described (only the lower USB port works). In the near future, I plan to add a separate DTB (with the necessary fixes) specifically for BPIr2pro to my build system. You can see all the patches I use in my GIT.
By the way, I checked the launch and operation of LibreELEC on BPIR2pro, everything starts and works (including directly from USB media using an Armbian SD card, with a new u-boot with support for direct launch of different systems directly from USB). ie, you can easily launch different systems by simply connecting a USB flash drive with test systems.
let me know when you have the dedicated dts for r2pro, or could you look what part may be missing in my hdmi branch for hdmi/analog audio? hdmi soundcard is deferred in my kernel and analogue was found, but left/right is not separated. no response to hp-detect when plugging in/out
maybe my board has a hardware-problem (as i had these 2 working with v00 board), maybe i miss some config-option (use the q64 one for hdmi).
does the hp-detect works for you (using headphone-microphone when 4pin plugged in?)
I see this is a BSP branch, I’m using the 5.17 and 5.18 core core. Why waste time on the old core?
I don’t have a 4 pin connector, I use regular speakers with 3 pins. You can easily check your equipment, use the latest version of Armbian with XFCE (download, unpack and burn to an SD card, you don’t need to change anything), connect and check whether your headphones will work or not. There is already a test clip in the composition that can be quickly launched, do not forget to switch the output from HDMI to analog in the mixer settings. My analog sound works well on BPIR2pro.
Ah sorry,right,copied the branchlink from commit.
Yes i mean the 5.18-r2pro-hdmiv10 which is basicly v11 (have not renamed it after doing rever/add difference patch).
For the VOP2 series 11 patches to work, you need to make a small correction.
I’m not sure if this is a bug, but this line is causing the build issue. After it is turned off, the assembly passes without errors and the HDMI output works. By the way, with the patches of the 11 series, the choice of screen resolution in the latest version of Libreelec began to work for me.
I think this is a temporary error and it will be fixed without me
I am using the full set of patches of the entire series along with the latest version of the kernel sources (without intermediate patches). Perhaps the reason for the build error is this. But I know for sure that HDMI works with the latest kernel version and the latest V11 patches on all my rk356x devices.
By the way, I checked DTB for BPIR2pro only with your patches on my 5.18-rc4 core, HDMI works without problems.
hdmi itself works for me too, but not the hdmi-sound which is deferred on hdmi-node i don’t know why. analoge audio soundcard is initialized but left/right channel was mixed in my last test