High-Performance Embedded Computing

104 Chapter 2 CPUs

three cycles for the first access and one cycle for subsequent sequential

accesses; and static-column DRAM with four cycles for the first access and one

cycle for subsequent sequential accesses, based on 70 ns access time. They

found that system performance improved when compressed code was run from

slow memories and that system performance slowed down about 10% when exe-

cuted from fast memory.

branches in

compressed code

If a branch is taken in the middle of a large block, we may not use some of

the instructions in the block, wasting the time and energy required to decom-

press those instructions. As illustrated in Figure 2-23, bran ...