Just wanted to report a related issue with tmc2130 drivers+marlin in case it helps anyone searching this. Last night I successfully printed a particular gcode file from octoprint. Printer auto shutdown did it's thing while I slept. Woke up, cleared the bed, and went to print the same gcode a 2nd time this morning, and it kept failing. The y axis stopped moving on it's way to the 3rd probe point when 3 point bed leveling at the start of the print. Checking M122 after this shows the y driver "ola" and "olb" flags!
This was very repeatable and frustrating... until I decided to try another gcode file... then it worked fine! Weird?!
My only guess is octoprint or a plugin somehow corrupts operation with just that one gcode file. I went to edit gcode in octoprint and found no unexpected lines or code added in the startup. In fact the second gcode that did work afterwards has the exact same startup code!
So something else hidden inside is causing the bug when printing that file. Later today I will try to delete the bad file from octoprint and see what happens when I re-upload and print.
I am running print time genius, and a bunch of other plugins that may be related. To be sure, I updated raspi, oprint, plugins, and marlin to the latest releases, and it still happens.
If further data will help, just point me to how to extract it. I have a storage oscope etc too. Not sure how obscure or important this is, but again wanted to report this strange behavior "in case it helps".