FIR Filtering: Basic Assembly Exercise for TI TMS320C54x 2.21 2001/05/30 2005/01/23 18:59:51.183 US/Central Douglas L. Jones dl-jones@uiuc.edu Swaroop Appadwedula appadwed@uiuc.edu Matthew J. Berry mjberry@uiuc.edu Mark A. Haun markhaun@uiuc.edu Jake Janevitz jake@janovetz.com Michael L. Kramer kramer@ifp.uiuc.edu Dima Moussa dmoussa@uiuc.edu Daniel Grobe Sachs sachs@uiuc.edu Brian Wade bwade@uiuc.edu Jason Noah Laska laska@uiuc.edu Mark D. Butala butala@uiuc.edu Matthew J. Berry mjberry@uiuc.edu Douglas L. Jones dl-jones@uiuc.edu Ricardo Anthony Radaelli-Sanchez ricky@alumni.rice.edu Swaroop Appadwedula appadwed@uiuc.edu Daniel Grobe Sachs sachs@uiuc.edu Robert L. Morrison rlmorris@uiuc.edu Jason Noah Laska laska@uiuc.edu assembly DSP fractional arithmetic hexadecimal mac rptz stl two's-compliment You will work through a section of TI TMS320C54x assembly code by hand. The instructions include multiplication of fractional numbers in two's complement representation.

Assembly Exercise Analyze the following lines of code. Refer to Two's Complement and Fractional Arithmetic for 16-bit Processors, Addressing Modes for TI TMS320C54x, and the Mnemonic Instruction Set manual for help. Anything following a ";" is considered a comment. In this case, the comments indicate the contents of the auxiliary registers, the BK register, and the address registers before the execution of the first instruction, stl. The line FIR_len .set 3 defines the name FIR_len as equal to 3. The BK register contains the length of the circular buffer we want to use. The % modifies the increment operator + so that it behaves as a circular buffer. This means that the address registers will be incremented until the (memory-address mod value-in-BK) = 0. When the increment operator + is followed by a 0, it increments by the value specified in register AR0. Note that any number followed by an "h" or preceded with a 0x represents a hexadecimal value. 1000h and 0x1000 both refer to the decimal number 4096. Assume that the data memory is initialized as follows starting at location 1000h.

Data Memory Assignment (before execution)

After familiarizing yourself with the stl, rptz, and mac instructions, step through each line of code and record the values of the accumulator A and auxiliary registers AR2 and AR3 in the spaces provided in . Additionally, record the value of the memory contents after all three instructions have been "executed" in the blank data memory table provided in .

A AR2 AR3 00 0000 8000h 1000h 1004h at start of code after stl instruction after rptz instruction after first mac instruction after second mac instruction after third mac instruction Execution Results

When working through the exercise, take into account that the accumulator A is a 40-bit register, and that the multiplier is in the fractional arithmetic mode. In this mode, integers on the DSP are interpreted as fractions, and the multiplier will treat them accordingly. This is done by shifting the result of the integer multiplier in the ALU left one bit. (All the arithmetic is fractional in these examples.) Multiplies performed by the ALU (via the mac instruction) produce a result that is twice what you would expect if you just multiplied the two integers together. DSP numerical representation and arithmetic are described further in Two's Complement and Fractional Arithmetic for 16-bit Processors.

Data Memory Assignment (after execution)