Asus Tinker board

As I've been chasing issues with my rspi3-7i90 combo( which until an update actually was controlling my HNC lathe) , I came across this guy, Asus tinker board. Seems to have real eithernet and spi that reads full 32 bit words, anyone have an opinion on this ? Would the stock hm2_spi driver play nice with it? Thanks

www.asus.com/us/Single-Board-Computer/Tinker-Board/

I suspect the problem with the driver would need to be figured out in any case since it
would likely be the starting point for a RockChip SPI driver. What update caused the problem?

2.8 pre, i was using an older hm2_rpspi driver and it pulled up numerous new_pin led errors. I have just recompiled the latest version from github an tried running lcnc from terminal and got this :

Note: Using POSIX realtime
hm2: loading Mesa HostMot2 driver version 0.15
hm2_rpspi: SPI0/CE0 Invalid cookie, read: 2ad5e5ff 2a292724 192a2726, expected: 55aacafe 54534f48 32544f4d
hm2_rpspi: SPI0/CE0 MISO input is bit-shifted by one bit. SPI read clock frequency probably too high.
hm2_rpspi: rtapi_app_main: No such device (-19)
./hnc112517.hal:24: waitpid failed /usr/bin/rtapi_app hm2_rpspi
./hnc112517.hal:24: /usr/bin/rtapi_app exited without becoming ready
./hnc112517.hal:24: insmod for hm2_rpspi failed, returned -1
Shutting down and cleaning up LinuxCNC...
hm2_rpspi: not loaded
<commandline>:0: exit value: 255
<commandline>:0: rmmod failed, returned -1
hm2: unloading
<commandline>:0: unloadrt failed


If I use the hm2_rspi driver , it loads then stops with errors, please see attached file I tried changing the clock rates with no sucess, are these drivers broken?

any input is appreciated

Would this thread have anything to do with the clock freq issue ?

github.com/raspberrypi/linux/issues/2094

Running at too low a speed should not hurt anything. I suspect this is a introduced driver bug that may need git-bisect to unravel. Does 2.7 work? Have you tried lowering the read clock speed?

I haven't tried 2.7, I've been seriously thinking about it as 2.8 with all
The joint/axis changes is kinda difficult. I did change the clock
Speeds, no difference. Is this relevant to my issue?

github.com/LinuxCNC/linuxcnc/issues/412

Issue 412 is not really related, there is a bug in un-fixed SPI bitfiles but it only affects
when a register write is performed (the bug was that SPI writes needed a following read or write
to actually finish the write cycle) AFAIK the only known effect was to cause the
sserial interface to stop working after a sserial race bug was fixed

I suspect the best thing to do is roll back to a working driver and relay the information of what version
works and what doesn't to the author