C6x assembly programming  (Page 12/11)

Introduction

This module contains details on how to program the TI C6000 family of processors in assembly. The C6000 family of processors has many variants. Therefore, it would not be possible to describe how to program all the processors here. However, the basic architecture and instructions are similar from one processor to another. They differ by the number of registers, the size of the registers, peripherals on the device, etc. This module will assume a device that has 32 general-purpose 32-bit registers and eight functional units, like the C6713 processor.

References

• SPRU198: TMS320C6000 Programmer's Guide
• SPRU186: TMS320C6000 Assembly Language Tools User's Guide 
• SPRU733: TMS320C67x/C67x+ DSP CPU and Instruction Set Reference Guide

Overview of c6000 architecture

The C6000 consists of internal memory, peripherals (serial port, external memory interface, etc), and most importantly, the CPU that has the registers and the functional units for execution of instructions. Although you don't need to care about the internal architecture of the CPU for compiling and running programs, it is necessary to understand how the CPU fetches and executes the assembly instructions to write a highly optimized assembly program.

Core dsp operation

In many DSP algorithms, the Sum of Products or Multiply-Accumulate (MAC) operations are very common. A DSP CPU is designed to handle the math-intensive calculations necessary for common DSP algorithms. For efficient implementation of the MAC operation, the C6000 CPU has two multipliers and each of them can perform a 16-bit multiplication in each clock cycle. For example, if we want to compute the dot product of two length-40 vectors a[n] and x[n], we need to compute:

(For example, the FIR filtering algorithm is exactly the same as this dot product operation.) When an a[n] and x[n]are stored in memory, starting from n=1, we need to compute a[n]x[n]and add it to y (y is initially 0) and repeat this up to n=40. In the C6000 assembly, this MAC operation can be written as:

Ignore .M and .L for now. Here, a , x , prod and y are numbers stored in memory and the instruction MPY multiplies two numbers a and x together and stores the result in prod . The ADD instruction adds two numbers y and prod together storing the result back to y .

Instructions

Below is the structure of a line of assembly code.

Labels identify a line of code or a variable and represent a memory address that contains either an instruction or data. The first character of a label must be must be in the first column and must be a letter or an underscore (_) followed by a letter. Labels can include up to 32 alphanumeric characters.

An instruction that executes in parallel with the previous instruction signifies this with parallel bars (||). This field is left blank for an instruction that does not execute in parallel with the previous instruction.