Table of Contents
Advertisement
Figure 6–15. Load Execution Block Diagram
Memory controller
In the E4 stage of a load, the data is received at the CPU core boundary. Final-
ly, in the E5 phase, the data is loaded into a register. Because data is not written
to the register until E5, load instructions have four delay slots. Because pointer
results are written to the register in E1, there are no delay slots associated with
the address modification.
In the following code, pointer results are written to the A4 register in the first
execute phase of the pipeline and data is written to the A3 register in the fifth
execute phase.
LDW
.D1
*A4++,A3
Because a store takes three execute phases to write a value to memory and
a load takes three execute phases to read from memory, a load following a
store accesses the value placed in memory by that store in the cycle after the
store is completed. This is why the store is considered to have zero delay slots.
Functional
unit
.D
E1
E5
Register file
Data
E4
Memory
Functional Unit Hazards
E2
Address
E3
TMS320C67x Pipeline
6-43
Advertisement
Table of Contents
