Descending further into the dimly-lit (figuratively and literally, since the documentation for this stuff is so hard to find) recesses of the BBB architecture, another minor success registered today.
I was able to put together a Device Tree Binary (.dtbo) file that the Angstrom OS can associate with my shield's EEPROM image. This basically means that if you attach the cape and then boot up the BBB, the BBB will absorb the .dtbo file into the overall system-wide device tree!
[ 3.605835] bone-capemgr bone_capemgr.9: slot #0: dtbo 'WPBBB-Experiment-00A0.dtbo' loaded; converting to live tree
[ 3.606012] bone-capemgr bone_capemgr.9: slot #0: #2 overlays
[ 3.606744] bone-capemgr bone_capemgr.9: slot #0: Applied #2 overlays.
[ 3.606763] bone-capemgr bone_capemgr.9: loader: done slot-0 WPBBB-Experiment:00A0 (prio 0)
I was able to put together a Device Tree Binary (.dtbo) file that the Angstrom OS can associate with my shield's EEPROM image. This basically means that if you attach the cape and then boot up the BBB, the BBB will absorb the .dtbo file into the overall system-wide device tree!
[ 3.605835] bone-capemgr bone_capemgr.9: slot #0: dtbo 'WPBBB-Experiment-00A0.dtbo' loaded; converting to live tree
[ 3.606012] bone-capemgr bone_capemgr.9: slot #0: #2 overlays
[ 3.606744] bone-capemgr bone_capemgr.9: slot #0: Applied #2 overlays.
[ 3.606763] bone-capemgr bone_capemgr.9: loader: done slot-0 WPBBB-Experiment:00A0 (prio 0)
The format of the ".dts" file (a human-readable source code before it is compiled into a .dtbo binary) is somewhat complex and frankly intimidating, so I'll have to study it carefully if I want to make good use of it.
So anyway, my BBB now recognises my cape and at least has the beginnings of being able to configure itself to match the requirements of the cape's IO needs. There is much work still to do though, I must write code for the ATMega I'm planning to add to the cape, and also the custom BBB device driver to allow the BBB to talk to the cape and pass data back and forth.
Fun :)