I am seeing signs of the chip wanting to be debugged by Eclipse on my Mac. It really does. I can feel it.
Two things seem to be plaguing me:
-
90% of the time it fails to download the code from gdb, citing a "block verification error". Would like to get to the bottom of this.
-
Once in a while it successfully downloads, then it breaks at the reset handler. I don't know why (I haven't set any breakpoints), but it shows me where it is and is looking like it might be starting to work.
Then I hit "resume", and I get this:
Read 2 bytes @ address 0x000003D4 (Data = 0x50D0)
Read 2 bytes @ address 0x000003D6 (Data = 0xDCFB)
Read 2 bytes @ address 0x000003D8 (Data = 0x4805)
Read 2 bytes @ address 0x000003DA (Data = 0x4780)
Starting target CPU...
ERROR: Programming failed @ address 0x00000000 (block verification error)
...Target halted (PC = 0xFFFFFFFE)
Reading all registers
Read 4 bytes @ address 0xFFFFFFFE (Data = 0x00007FFF)
Read 4 bytes @ address 0x1FFFFFFC (Data = 0x00000000)
Reading 64 bytes @ address 0x1FFFFFC0
Read 4 bytes @ address 0x00000000 (Data = 0x20004000)
Read 2 bytes @ address 0xFFFFFFFE (Data = 0x0000)
Read 2 bytes @ address 0x00000000 (Data = 0x4000)
Read 2 bytes @ address 0x00000002 (Data = 0x2000)
Read 2 bytes @ address 0x00000004 (Data = 0x03BD)
It starts the target CPU again, and I get another block verification error. What is it trying to download to address 0?
My gdb startup script looks like this:
mon speed 30 mon endian little mon flash download = 1 mon flash device = NRF51822 mon sleep 1000 mon reset 0 mon sleep 1000
The sleeps seem to help, but this is starting to feel like a bad game of whack-a-mole.
Any ideas?
