The Image processing demo splits L2 in half and I was just copying that. It maps some program components to L2, and I believe some of that mapping is required. It doesn't use DDR except for the NDK heap.
I got one core working a long while ago (with NDK comm to host) but have not had any luck getting NDK and IPC to work together. Things went awry when I tried to use NDK and IPC together.