s-2

Thanks, works like a charm! I guess I was just a little too early to the party for this particular box 😇

Okay, finally managed to get this thing into mask ROM mode 🙂 Shorting the resistor arrays on the top PCB layer near the eMCP helped (the visible difference is, there is just no UART output in maskrom). The device is listed, but can not display info about flash (not sure if that's to be expected), will look further into which loader to use from here on etc. $ rkdeveloptool ld DevNo=1 Vid=0x2207,Pid=0x320b,LocationID=101 Maskrom $ rkdeveloptool rid Reading flash ID failed! $ rkdeveloptool rfi Read Flash Info failed! // edit: okay, booting multitool rom microSD works just the same way as entering maskrom, by shorting the eMCp pins, simple as that. trying to restore backup for a start, I had used 6.1.50 from https://imola.armbian.com/dl/rk322x-box/archive/ (downloaded ~3 weeks ago, but can try 6.1.63 as well). When using the "Enhanced By Google" search on the official armbian website, it would always lead me back to this thread. Are there any other official download sources for the latest (official) images? Indeed the size of multitool.img.gz linked in the first post also changed since then, using the latest version now. // edit: they're identical, size difference was only reported during download

Oh wow, that seemed almost too obivous, so these are indeed a completely different family of devices... sorry for the noise. Okay thanks, I will just try more pins then, maybe also CLK line from external microSD slot etc. 🙂

Hi everyone, I'm quite new to both these rockchip devices as well as armbian (at least from a developer's perspective), but have quite some experience with OpenWrt (adding BSP for several wireless routers e.g. ath79, ramips-mt7621), so am familiar with device tree files and the low level hardware stuff. For some fun over the holidays, I grabbed a few cheap rockchip boxes from AliExpress - starting with the rk322x ones here. Sorry I didn't read through the full 79 pages of this thread yet, so not sure about how much of this is still relevant, and what is in mainline armbian already. Box #1: "MX Q Pro 4K", eMMC + separate RAM chips. L-shaped PCB "MXQ-18DP58_01" with SD card on the thinner side Could not manage to get it to boot multitool or armbian from SD card, the box would halt with a dim blue LED. Looking at the serial console (booting OEM Android), the TX pin seems to be shared with the external SD card, so there is some garbage visible after "FwPartOffset", when it probes for SD card, but none is inserted: When the SD card is inserted, it boots until "FwPartOffset" with lots of garbage following, i.e. it is apparently reading lots of stuff from the SD card. Couldn't get much further with this for now, so proceeded to Box #2: "H20 4K Ultra HD", green board with eMCP, board label "H20-221-V1.71-2-B HXW23071201 20230714". This one perfectly booted multitool from microSD, made a full backup, downloaded `Armbian_23.8.1_Rk322x-box_bookworm_current_6.1.50.img.xz`, thinking "what could probably go wrong", dared to write this image to flash. Trying to restore the backup, but now it's not even booting to multitool from microSD anymore (This box is probably a new version of H20, as also mentioned in this thread https://forum.armbian.com/topic/29644-h20-box-with-rk3228a-bricked/). Backup: https://mega.nz/file/SZ433bJA#RQZTSJrsjyJWOXlFWV6TLaqxMB0-C4JtExWyQbAyqp8 Tried to enter maskrom to re-flash the backup with rkdeveloptool, shorting resistors to GND as seen in some post here..., but no success yet. Can't see anything enumerating on USB. Is the cable quality relevant here? I built a ~10cm USB A<->A cable based on flat cable, with only GND, D- and D+ connected (the 5V from the H20 power plug go directly to the VBUS pins of the pins, don't want to feed this back to my host coputer). So there is probably no VBUS detection required for enumerating the USB, since it is forced to 5V all the time anyways? Is there any special output to be expected on the serial console to verify maskrom was entered? I only see DDR Version V1.10 20190926 In ID:0xFFF 330MHz LPDDR3 Bus Width=32 Col=10 Bank=8 Row=15 CS=2 Die Bus-Width=32 Size=2048MB mach:14 rd addr 0xC0000000 = 0xFFFF ERR What is located at 0xC000000? The output is the same no matter if an sd card is inserted or not. My understanding is, mask ROM mode will be entered when no other bootable source was detected, that's why we short the eMMC clock line to GND? (i.e. also microSD needs to be removed then) So, how do I know I'm in maskrom mode, can someone show a serial log? Maybe I just need to try different pins, or something was damaged while soldering a push-button to the CLK line (are those tiny capacitors really needed? 😇 )