gmcastil
commented
11 months ago Looks like at least some of this is because the bram->map_size is the memory depth in accesses, not the size in bytes. There is an entry in the device tree that indicates the bus width, so I should be able to modify the bram_resource data structure to pull that out.
Adding the EEPROM and SRAM emulators and changing widths, latencies, and everything else broke something because purging with an incrementing pattern doesn't yield what it should. I'll look at this after the Thanksgiving weekend.