Naive matrix multiply

This program implements the textbook (naive) matrix multiplication algorithm. The algorithm mirrors the textbook definition of matrix multiplication in linear algebra. The algorithm has known performance issues (data cache misses and data translation lookaside buffer misses). These issues slow down memory access.

The program calls the user-space ARM11 performance counter functions in rpi_pmu.c. It also calls utility functions in test_common.c. Thus, these source modules must be compiled and linked along with naive.c.