TMS320VC5502 Library Reference

TMS320C55x DSP Library
Programmer’s Reference
SPRU422J − May 2000

Revised − May 2013
Preface
Read This First
About This Manual
The Texas Instruments TMS320C55x™ DSPLIB is an optimized DSP Function

Library for C programmers on TMS320C55x devices. It includes over 50
C-callable assembly-optimized general-purpose signal processing routines.
These routines are typically used in computationally intensive real-time
applications where optimal execution speed is critical. By using these routines
you can achieve execution speeds considerable faster than equivalent code
written in standard ANSI C language. In addition, by providing ready-to-use
DSP functions, TI DSPLIB can shorten significantly your DSP application
development time.
Related Documentation
- The MathWorks, Inc. Matlab Signal Processing Toolbox User’s Guide. Na-
tick, MA: The MathWorks, Inc., 1996. .
- Lehmer, D.H. “Mathematical Methods in large-scale computing units.”

Proc. 2nd Sympos. on Large-Scale Digital Calculating Machinery, Cam-
bridge, MA, 1949. Cambridge, MA: Harvard University Press, 1951.
- Oppenheim, Alan V. and Ronald W Schafer. Discrete-Time Signal Proces-

sing. Englewood Cliffs, NJ: Prentice Hall, 1989.
- Digital Signal Processing with the TMS320 Family (SPR012)
- TMS320C55x DSP CPU Reference Guide (SPRU371)
- TMS320C55x Optimizing C Compiler User’s Guide (SPRU281)
Trademarks
TMS320, TMS320C55x, and C55x are trademarks of Texas Instruments.
Matlab is a trademark of Mathworks, Inc.
Read This First iii

iv
Contents
Contents
1 Contents
Introduction to the TMS320C55x DSP Library
1.1 DSP Routines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
1.2 Features and Benefits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
1.3 DSPLIB: Quality Freeware That You Can Build On and Contribute To . . . . . . . . . . . . . . 1-2
2 Contents
Describes how to install the DSPLIB
2.1 DSPLIB Content . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2
2.2 How to Install DSPLIB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3
2.2.1 De-Archive DSPLIB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3
2.2.2 Relocate Library File . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3
2.3 How to Rebuild DSPLIB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-4
2.3.1 For Full Rebuild of 55xdsp.lib . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-4
2.3.2 For Partial Rebuild of 55xdsp.lib (modification of a specific DSPLIB function,
for example fir.asm) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-4
3 Contents
Describes how to use the DSPLIB
3.1 DSPLIB Arguments and Data Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2
3.1.1 DSPLIB Arguments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2
3.1.2 DSPLIB Data Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2
3.2 Calling a DSPLIB Function from C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3
3.3 Calling a DSPLIB Function from Assembly Language Source Code . . . . . . . . . . . . . . . 3-3
3.4 Where to Find Sample Code . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3
3.5 How DSPLIB is Tested − Allowable Error . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-4
3.6 How DSPLIB Deals with Overflow and Scaling Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-4
3.7 Where DSPLIB Goes From Here . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-6
4 Contents
Provides descriptions for the TMS320C55x DSPLIB functions
4.1 Arguments and Conventions Used . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-2
4.2 DSPLIB Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-3
5 DSPLIB Benchmarks and Performance Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1

Describes benchmarks and performance issues for the DSPLIB functions
5.1 What DSPLIB Benchmarks are Provided . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-2
v
Contents
5.2 Performance Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-2
6 Software Updates and Customer Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1

Details the software updates and customer support issues for the TMS320C55x DSPLIB
6.1 DSPLIB Software Updates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-2
6.2 DSPLIB Customer Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-2
7 Overview of Fractional Q Formats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-1

Describes the fractional Q formats used by the DSPLIB functions
A.1 Q3.12 Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-2
A.2 Q.15 Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-2
A.3 Q.31 Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-2
8 Calculating the Reciprocal of a Q15 Number . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-1

Provides the calculations used to find the inverse of a fractional Q15 number
vi
Figures
Figures
4−1 dbuffer Array in Memory at Time j . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-24

4−2 x Array in Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-25
4−3 r Array in Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-25
4−4 x Array in Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-32
4−5 r Array in Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-32
4−6 h Array in Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-32
4−7 x Array in Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-34
4−8 r Array in Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-34
4−9 h Array in Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-34
4−10 x Array in Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-36
4−11 r Array in Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-36
4−12 h Array in Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-36
4−13 x Buffer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-43
4−14 dbuffer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-44
4−15 h Buffers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-44
4−17 x Array in Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-49
4−18 r Array in Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-49
4−20 x Array in Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-52
4−21 r Array in Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-52
4−23 x Array in Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-61
4−24 r Array in Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-62
4−26 x Array in Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-66
4−27 r Array in Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-66
Contents vii
Tables
Tables
4−1 Function Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-2

4−2 Summary Table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-3
A−1 Q3.12 Bit Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-2
A−2 Q.15 Bit Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-2
A−3 Q.31 Low Memory Location Bit Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-2
A−4 Q.31 High Memory Location Bit Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-2
viii
Contents ix
Chapter 1
Introduction
The Texas Instruments TMS320C55x DSP Library (DSPLIB) is an optimized

DSP Function Library for C programmers on TMS320C55x devices. It includes
over 50 C-callable assembly-optimized general-purpose signal processing
routines. These routines are typically used in computationally intensive real-
time applications where optimal execution speed is critical. By using these rou-
tines you can achieve execution speeds considerable faster than equivalent
code written in standard ANSI C language. In addition, by providing ready-to-
use DSP functions, TI DSPLIB can shorten significantly your DSP application
development time.
Topic Page
1.1 DSP Routines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2

1.2 Features and Benefits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
1.3 DSPLIB: Quality Freeware That You Can Build On
and Contribute To . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
1-1
DSP Routines
1.1 DSP Routines

The TI DSPLIB includes commonly used DSP routines. Source code is
provided to allow you to modify the functions to match your specific needs.
The routines included within the library are organized into eight different
functional categories:
- Fast-Fourier Transforms (FFT)
- Filtering and convolution
- Adaptive filtering
- Correlation
- Math
- Trigonometric
- Miscellaneous
- Matrix
1.2 Features and Benefits

- Hand-coded assembly optimized routines
- C-callable routines fully compatible with the TI C55x compiler
- Fractional Q15-format operand supported
- Complete set of examples on usage provided
- Benchmarks (time and code) provided
- Tested against Matlab™ scripts
1.3 DSPLIB: Quality Freeware That You Can Build On and Contribute To
DSPLIB is a free-of-charge product. You can use, modify, and distribute TI
C55x DSPLIB for usage on TI C55x DSPs with no royalty payments. See
section 3.7, Where DSPLIB Goes From Here, for details.
1-2
Chapter 2
Installing DSPLIB
This chapter describes how to install the DSPLIB.
Topic Page
2.1 DSPLIB Content . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2

2.2 How to Install DSPLIB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3
2.3 How to Rebuild DSPLIB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-4
2-1
DSPLIB Content
2.1 DSPLIB Content

The TI DSPLIB software consists of 4 parts:
1) a header file for C programmers under the ”include” sub−directory:
dsplib.h
2) One object library under the ”lib” sub−directory:

55xdsp.lib
3) One source library to allow function customization by the end user under
the ”55x_src” sub−directory
55xdsp.src
4) Example programs and linker command files used under the “55x_test”
Examples sub-directory .
2-2
How to Install DSPLIB
2.2 How to Install DSPLIB
Note:
Read the README.TXT file for specific details of release.
2.2.1 De-Archive DSPLIB

DSPLIB is distributed in the form of an executable self-extracting ZIP file
(c55_dsplib.exe). The zip file automatically restores the DSPLIB individual
components in the same directory you execute the self extracting file. Follow-
ing is an example on how to install DSPLIB, just type:
c55_dsplib.exe −d
The DSPLIB directory structure and content you will find is:
c55_dsplib(dir)
55xdsp.lib : use for standards short-call mode
blt55x.bat : re-generate 55xdsp.lib based on 55xdsp.src
examples(dir) : contains one subdirectory for each routine included in the
library where you can find complete test cases
include(dir)
dsplib.h : include file with data types and function prototypes
tms320.h : include file with type definitions to increase TMS320 porta-
bility
misc.h : include file with useful miscellaneous definitions
doc(dir)
55x_src (dir) : contains assembly source files for functions
2.2.2 Relocate Library File

Copy the C55x DSPLIB object library file, 55xdsp.lib, to your C5500 runtime
support library folder.
For example, if your TI C5500 tools are located in c:\ti\c5500\cgtools\bin and

c runtime support libraries (rts55.lib etc.) in c:\ti\c5500\cgtools\lib, copy
55xdsplib.lib to this folder. This allows the C55x compiler/linker to find
55xdsp.lib.
Installing DSPLIB 2-3

How to Rebuild DSPLIB
2.3 How to Rebuild DSPLIB
2.3.1 For Full Rebuild of 55xdsp.lib

To rebuild 55xdsp.lib, execute the blt55x.bat. This will overwrite any existing
55xdsp.lib.
2.3.2 For Partial Rebuild of 55xdsp.lib

(modification of a specific DSPLIB function, for example fir.asm)
1) Extract the source for the selected function from the source archive:
ar55 x 55xdsp.src fir.asm
2) Re-assemble your new fir.asm assembly source file:

asm55 –g fir.asm
3) Replace the object , fir.obj, in the dsplib.lib object library with the newly
formed object:
ar55 r 55xdsp.lib fir.obj
2-4
Chapter 3
Using DSPLIB
This chapter describes how to use the DSPLIB.
Topic Page
3.1 DSPLIB Arguments and Data Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2

3.2 Calling a DSPLIB Function from C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3
3.3 Calling a DSPLIB Function from Assembly Language
Source Code . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3
3.4 Where to Find Sample Code . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3
3.5 How DSPLIB is Tested — Allowable Error . . . . . . . . . . . . . . . . . . . . . . . 3-4
3.6 How DSPLIB Deals with Overflow and Scaling Issues . . . . . . . . . . . . 3-4
3.7 Where DSPLIB Goes From Here . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-6
3-1
DSPLIB Arguments and Data Types
3.1 DSPLIB Arguments and Data Types
3.1.1 DSPLIB Arguments

DSPLIB functions typically operate over vector operands for greater efficiency.
Though these routines can be used to process short arrays or scalars (unless
a minimum size requirement is noted) , the execution times will be longer in
those cases.
- Vector stride is always equal 1: vector operands are composed of vector

elements held in consecutive memory locations (vector stride equal to 1).
- Complex elements are assumed to be stored in a Re-Im format.
- In-place computation is allowed (unless specifically noted): Source

operand can be equal to destination operand to conserve memory.
3.1.2 DSPLIB Data Types

DSPLIB handles the following fractional data types:
- Q.15 (DATA) : A Q.15 operand is represented by a short data type (16 bit)
that is predefined as DATA, in the dsplib.h header file.
- Q.31 (LDATA) : A Q.31 operand is represented by a long data type (32 bit)
that is predefined as LDATA, in the dsplib.h header file.
- Q.3.12 : Contains 3 integer bits and 12 fractional bits.
Unless specifically noted, DSPLIB operates on Q15-fractional data type

elements. Appendix A presents an overview of Fractional Q formats
3-2
Calling a DSPLIB Function from C
3.2 Calling a DSPLIB Function from C

In addition to installing the DSPLIB software, to include a DSPLIB function in
your code you have to:
- Include the dsplib.h include file
- Link your code with the DSPLIB object code library, 55xdsp.lib or
55xdspx.lib.
- Use a correct linker command file describing the memory configuration

available in your C55x board.
A project file has been included for each function in the examples folder. You
can reference function_t.c files for calling a DSPLIB function from C.
The examples presented in this document have been tested using the Texas
Instruments C55x Simulator. Customization may be required to use it with a
different simulator or development board.
Refer to the TMS320C55x Optimizing C Compiler User’s Guide (SPRU281).
3.3 Calling a DSPLIB Function from Assembly Language Source Code

The TMS320C55x DSPLIB functions were written to be used from C. Calling
the functions from assembly language source code is possible as long as the
calling-function conforms with the Texas Instruments C55x C compiler calling
conventions. Refer to the TMS320C55x Optimizing C Compiler User’s Guide,
if a more in-depth explanation is required.
Realize that the TI DSPLIB is not an optimal solution for assembly-only

programmers. Even though DSPLIB functions can be invoked from an
assembly program, the result may not be optimal due to unnecessary C-calling
overhead.
3.4 Where to Find Sample Code

You can find examples on how to use every single function in DSPLIB, in the
examples subdirectory. This subdirectory contains one subdirectory for each
function. For example, the examples/araw subdirectory contains the following
files:
- araw_t.c: main driver for testing the DSPLIB acorr (raw) function.
- test.h: contains input data(a) and expected output data(yraw) for the acorr
(raw) function as. This test.h file is generated by using Matlab scripts.
Using DSPLIB 3-3

How DSPLIB is Tested − Allowable Error
- test.c: contains function used to compare the output of araw function with
the expected output data.
- ftest.c: contains function used to compare two arrays of float data types.
- ltest.c: contains function used to compare two arrays of long data types.
- ld3.cmd: an example of a linker command you can use for this function.
3.5 How DSPLIB is Tested − Allowable Error

Version 1.0 of DSPLIB is tested against Matlab scripts. Expected data output
has been generated from Matlab that uses double-precision (64-bit) floating-
point operations (default precision in Matlab). Test utilities have been added
to our test main drivers to automate this checking process. Note that a maxi-
mum absolute error value (MAXERROR) is passed to the test function, to set
the trigger point to flag a functional error.
We consider this testing methodology a good first pass approximation. Further

characterization of the quantization error ranges for each function (under ran-
dom input) as well as testing against a set of fixed-point C models is planned
for future releases. We welcome any suggestions you may have on this
respect.
3.6 How DSPLIB Deals with Overflow and Scaling Issues

One of the inherent difficulties of programming for fixed-point processors is
determining how to deal with overflow issues. Overflow occurs as a result of
addition and subtraction operations when the dynamic range of the resulting
data is larger than what the intermediate and final data types can contain.
The methodology used to deal with overflow should depend on the specifics
of your signal, the type of operation in your functions, and the DSP architecture
used. In general, overflow handling methodologies can be classified in five
categories: saturation, input scaling, fixed scaling, dynamic scaling, and
system design considerations.
It’s important to note that a TMS320C55x architectural feature that makes

overflow easier to deal with is the presence of guard bits in all four accumula-
tors. The 40-bit accumulators provide eight guard bits that allow up to 256 con-
secutive multiply-and-accumulate (MAC) operations before an accumulator
overrun – a very useful feature when implementing, for example, FIR filters.
3-4
How DSPLIB Deals with Overflow and Scaling Issues
There are 4 specific ways DSPLIB deals with overflow, as reflected in each
function description:
- Scaling implemented for overflow prevention: In this type of function,

DSPLIB scales the intermediate results to prevent overflow. Overflow
should not occur as a result. Precision is affected but not significantly. This
is the case of the FFT functions, in which scaling is used after each FFT
stage.
- No scaling implemented for overflow prevention: In this type of func-

tion, DSPLIB does not scale to prevent overflow due to the potentially
strong effect in data output precision or in the number of cycles required.
This is the case, for example, of the MAC-based operations like filtering,
correlation, or convolutions. The best solution on those cases is to design
your system , for example your filter coefficients with a gain less than 1 to
prevent overflow. In this case, overflow could happen unless you input
scale or you design for no overflow.
- Saturation implemented for overflow handling: In this type of function,

DSPLIB has enabled the TMS320C55x 32-bit saturation mode (SATD
bit = 1). This is the case of certain basic math functions that require the
saturation mode to be enabled.
- Not applicable: In this type of function, due to the nature of the function
operations, there is no overflow.
- DSPLIB reporting of overflow conditions (overflow flag): Due to the

sometimes unpredictible overflow risk, most DSPLIB functions have been
written to return an overflow flag (oflag) as an indication of a potentially
dangerous 32-bit overflow. However, because of the guard-bits, the C55x
is capable of handling intermediate 32-bit overflows and still produce the
correct final result. Therefore, the oflag parameter should be taken in the
context of a warning but not a definitive error.
As a final note, DSPLIB is provided also in source format to allow customiza-

tion of DSPLIB functions to your specific system needs.
Using DSPLIB 3-5

Where DSPLIB Goes From Here
3.7 Where DSPLIB Goes From Here

We anticipate DSPLIB to improve in future releases in the following areas:
- Increased number of functions: We anticipate the number of functions

in DSPLIB will increase. We welcome user-contributed code. If during the
process of developing your application you develop a DSP routine that
seems like a good fit to DSPLIB, let us know. We will review and test your
routine and possibly include it in the next DSPLIB software release. Your
contribution will be acknowledged and recognized by TI in the Acknowl-
edgments section. Use this opportunity to make your name known by your
DSP industry peers. Simply email your contribution To Whom It May Con-
cern: dsph@ti.com and we will contact you.
- Increased Code portability: DSPLIB looks to enhance code portability

across different TMS320-based platforms. It is our goal to provide similar
DSP libraries for other TMS320™ devices, working in conjunction with
C55x compiler intrinsics to make C-developing easier for fixed-point
devices. However, it’s anticipated that a 100% portable library across
TMS320 devices may not be possible due to normal device architectural
differences. TI will continue monitoring DSP industry standardization acti-
vities in terms of DSP function libraries.
3-6
Chapter 4
Function Descriptions
This chapter provides descriptions for the TMS330C55x DSPLIB functions.
Topic Page
4.1 Arguments and Conventions Used . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-2

4.2 DSPLIB Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-3
4-1
4.1 Arguments and Conventions Used
The following convention has been followed when describing the arguments
for each individual function:
Table 4−1. Function Descriptions
Argument Description
x,y argument reflecting input data vector
r argument reflecting output data vector
nx,ny,nr arguments reflecting the size of vectors x,y, and r respectively. In

functions where nx = nr = nr, only nx has been used.
h Argument reflecting filter coefficient vector (filter routines only)
nh Argument reflecting the size of vector h
DATA data type definition equating a short, a 16-bit value representing a

Q15 number. Usage of DATA instead of short is recommended to
increase future portability across devices.
LDATA data type definition equating a long, a 32-bit value representing a

Q31 number. Usage of LDATA instead of long is recommended to
increase future portability across devices.
ushort Unsigned short (16 bit). You can use this data type directly,
because it has been defined in dsplib.h
4-2
4.2 DSPLIB Functions
The routines included within the library are organized into 8 different functional
categories:
- FFT
- Filtering and convolution
- Adaptive filtering
- Correlation
- Math
- Trigonometric
- Miscellaneous
- Matrix
Table 4−2 lists the functions by these 8 functional catagories.
Table 4−2. Summary Table
(a) FFT
Functions Description
void cfft (DATA *x, ushort nx, type) Radix-2 complex forward FFT − MACRO
void cfft32 (LDATA *x, ushort nx, type); 32-bit forward complex FFT
void cifft (DATA *x, ushort nx, type) Radix-2 complex inverse FFT − MACRO
void cifft32 (LDATA *x, ushort nx, type); 32-bit inverse complex FFT
void cbrev (DATA *x, DATA *r, ushort n) Complex bit-reverse function
void cbrev32 (LDATA *a, LDATA *r, ushort) 32-bit complex bit reverse
void rfft (DATA *x, ushort nx, type) Radix-2 real forward FFT − MACRO
void rifft (DATA *x, ushort nx, type) Radix-2 real inverse FFT − MACRO
void rfft32 (LDATA *x, ushort nx, type) Forward 32-bit Real FFT (in-place)
void rifft32 (LDATA *x, ushort nx, type) Inverse 32-bit Real FFT (in-place)
(b) Filtering and Convolution
ushort fir (DATA *x, DATA *h, DATA *r, DATA *dbuffer, FIR direct form
ushort nx, ushort nh)
ushort fir2 (DATA *x, DATA *h, DATA *r, DATA *dbuffer, FIR direct form (Optimized to use DUAL−MAC)
ushort firs (DATA *x, DATA *h, DATA *r, DATA *dbuffer, Symmetric FIR direct form (generic routine)
ushort nx, ushort nh2)
Function Descriptions 4-3

Table 4−2. Summary Table (Continued)
ushort cfir (DATA *x, DATA *h, DATA *r, DATA *dbuffer, Complex FIR direct form
ushort convol (DATA *x, DATA *h, DATA *r, ushort nr, Convolution
ushort nh)
ushort convol1 (DATA *x, DATA *h, DATA *r, ushort nr, Convolution (Optimized to use DUAL−MAC)
ushort nh)
(b) Filtering and Convolution (Continued)
ushort convol2 (DATA *x, DATA *h, DATA *r, ushort nr, Convolution (Optimized to use DUAL−MAC)
ushort nh)
ushort iircas4 (DATA *x, DATA *h, DATA *r, DATA *dbuffer, IIR cascade direct form II. 4 coefficients per
ushort nbiq, ushort nx) biquad.
ushort iircas5 (DATA *x, DATA *h, DATA *r, DATA *dbuffer, IIR cascade direct form II. 5 coefficients per
ushort nbiq, ushort nx) biquad
ushort iircas51 (DATA *x, DATA *h, DATA *r, DATA *dbuffer, IIR cascade direct form I. 5 coefficients per
ushort nbiq, ushort nx) biquad
ushort iirlat (DATA *x, DATA *h, DATA *r, DATA *pbuffer, Lattice inverse IIR filter
int nx, int nh)
ushort firlat (DATA *x, DATA *h, DATA *r, DATA *pbuffer, Lattice forward FIR filter
int nx, int nh)
ushort firdec (DATA *x, DATA *h, DATA *r, DATA *dbuffer, Decimating FIR filter
ushort nh, ushort nx, ushort D)
ushort firinterp (DATA *x, DATA *h, DATA *r, DATA *dbuffer, Interpolating FIR filter
ushort nh, ushort nx, ushort I)
ushort hilb16 (DATA *x, DATA *h, DATA *r, DATA *dbuffer, FIR Hilbert Transformer
ushort iir32 (DATA *x, LDATA *h, DATA *r, LDATA *dbuffer, Double-precision IIR filter
ushort nbiq, ushort nr)
(c) Adaptive filtering
ushort dlms (DATA *x, DATA *h, DATA *r, DATA *des, LMS FIR (delayed version)
DATA *dbuffer, DATA step, ushort nh, ushort nx)
ushort oflag = dlmsfast (DATA *x, DATA *h, DATA *r, DATA Adaptive delayed LMS filter (fast implemented)
*des, DATA *dbuffer, DATA step, ushort nh, ushort nx)
4-4
(d) Correlation
ushort acorr (DATA *x, DATA *r, ushort nx, ushort nr, type) Autocorrelation (positive side only) − MACRO
ushort corr (DATA *x, DATA *y, DATA *r, ushort nx, ushort Correlation (full-length)
ny, type)
(e) Trigonometric
ushort sine (DATA *x, DATA *r, ushort nx) sine of a vector
ushort atan2_16 (DATA *q, DATA *i, DATA *r, ushort nx) Four quadrant inverse tangent of a vector
ushort atan16 (DATA *x, DATA *r, ushort nx) Arctan of a vector
(f) Math
ushort add (DATA *x, DATA *y, DATA *r, ushort nx, Optimized vector addition
ushort scale)
ushort expn (DATA *x, DATA *r, ushort nx) Exponent of a vector
short bexp (DATA *x, ushort nx) Exponent of all values in a vector
ushort logn (DATA *x, LDATA *r, ushort nx) Natural log of a vector
ushort log_2 (DATA *x, LDATA *r, ushort nx) Log base 2 of a vector
ushort log_10 (DATA *x, LDATA *r, ushort nx) Log base 10 of a vector
short maxidx (DATA *x, ushort ng, ushort ng_size) Index for maximum magnitude in a vector
short maxidx34 (DATA *x, ushort nx) Index of the maximum element of a vector ≤ 34
short maxval (DATA *x, ushort nx) Maximum magnitude in a vector
void maxvec (DATA *x, ushort nx, DATA *r_val, Index and value of the maximum element of a
DATA *r_idx) vector
short minidx (DATA *x, ushort nx) Index for minimum magnitude in a vector
short minval (DATA *x, ushort nx) Minimum element in a vector
void minvec (DATA *x, ushort nx, DATA *r_val, Index and value of the minimum element of a
DATA *r_idx) vector
ushort mul32 (LDATA *x, LDATA *y, LDATA *r, ushort nx) 32-bit vector multiply
short neg (DATA *x, DATA *r, ushort nx) 16-bit vector negate
short neg32 (LDATA *x, LDATA *r, ushort nx) 32-bit vector negate

short power (DATA *x, LDATA *r, ushort nx) sum of squares of a vector (power)
void recip16 (DATA *x, DATA *r, DATA *rexp, ushort nx) Vector reciprocal
void ldiv16 (LDATA *x, DATA *y, DATA *r, DATA *rexp, 32-bit by 16-bit long division
ushort nx)
(f) Math (Continued)
ushort sqrt_16 (DATA *x, DATA *r, short nx) Square root of a vector
short sub (DATA *x, DATA *y, DATA *r, ushort nx, Vector subtraction
ushort scale)
(g) Matrix
ushort mmul (DATA *x1, short row1, short col1, matrix multiply
DATA *x2, short row2, short col2, DATA *r)
ushort mtrans (DATA *x, short row, short col, DATA *r) matrix transponse
(h) Miscellaneous
ushort fltoq15 (float *x, DATA *r, ushort nx) Floating-point to Q15 conversion
ushort q15tofl (DATA *x, float *r, ushort nx) Q15 to floating-point conversion
ushort rand16 (DATA *r, ushort nr) Random number generation
void rand16init(void) Random number generation initialization
4-6
acorr
acorr Autocorrelation
Function ushort oflag = acorr (DATA *x, DATA *r, ushort nx, ushort nr, type)
(defined in araw.asm, abias.asm , aubias.asm)
Arguments
x [nx] Pointer to real input vector of nx real elements. nx ≥ nr
r [nr] Pointer to real output vector containing the first nr elements
of the positive side of the autocorrelation function of vector x.
r must be different than x (in-place computation is not
allowed).
nx Number of real elements in vector x

nr Number of real elements in vector r
type Autocorrelation type selector. Types supported:
- If type = raw, r contains the raw autocorrelation of x
- If type = bias, r contains the biased autocorrelation of x
- If type = unbias, r contains the unbiased autocorrelation of
x
oflag Overflow flag.

- If oflag = 1, a 32-bit overflow has occurred
- If oflag = 0, a 32-bit overflow has not occurred
Description Computes the first nr points of the positive side of the autocorrelation of the
real vector x and stores the results in real output vector r. The full-length auto-
correlation of vector x will have 2*nx−1 points with even symmetry around the
lag 0 point (r[0]). This routine provides only the positive half of this for memory
and computational savings.
Algorithm Raw Autocorrelation
nx*j*1
r [j] + ȍ x [j ) k] x [k] 0 v j v nr
k+0
Biased Autocorrelation
nx*j*1
r [j] + 1
nx
ȍ x[j ) k] x [k] 0 v j v nr
k+0
Unbiased Autocorrelation
nx*j*1
r [j] + 1
(nx * abs(j))
ȍ x[j ) k] x[k] 0 v j v nr
k+0

acorr
Overflow Handling Methodology No scaling implemented for overflow prevention
Special Requirements x array in internal memory (coefficient pointer CDP used to address it)
Implementation Notes
- Special debugging consideration: This function is implemented as a mac-
ro that invokes different autocorrelation routines according to the type
selected. As a consequence the acorr symbol is not defined. Instead the
acorr_raw, acorr_bias, acorr_unbias symbols are defined.
- Autocorrelation is implemented using time-domain techniques
Example See examples/abias, examples/aubias, examples/araw subdirectories
Benchmarks (preliminary)
Cycles† Abias:
Core:
nr even: [(4 * nx − nr * (nr + 2) + 20) / 8] * nr
nr odd: [(4 * nx − (nr − 1) * (nr + 1) + 20) / 8] * (nr − 1) + 10
nr = 1: (nx + 2)
Overhead:
nr even: 90
nr odd: 83
nr = 1: 59
Araw:
Core:
nr even: [(4 * nx − nr * (nr + 2) + 28) / 8] * nr
nr odd: [(4 * nx − (nr − 1) * (nr + 1) + 28) / 8] * (nr − 1) + 13
nr = 1: (nx + 1)
Overhead:
nr even: 34
nr odd: 35
nr = 1: 30
† Assumes all data is in on-chip dual-access RAM and that there is no bus conflict due to twiddle
table reads and instruction fetches (provided linker command file reflects those conditions).
4-8
add
Cycles† Aubias:
Core:
nreven: [(8 * nx − 3 * nr * (nr + 2) + 68) / 8] * nr
nr odd: [(8 * nx − 3 * (nr−1) * (nr+1) + 68)/8] * (nr − 1) + 33
nr = 1: nx + 26
Overhead:
nr even: 64
nr odd: 55
nr = 1: 47
Code size Abias: 226
(in bytes) Araw: 178
Aubias: 308
add Vector Add
Function ushort oflag = add (DATA *x, DATA *y, DATA *r, ushort nx, ushort scale)
(defined in add.asm)
Arguments
x[nx] Pointer to input data vector 1 of size nx. In-place processing
allowed (r can be = x = y)
y[nx] Pointer to input data vector 2 of size nx
r[nx] Pointer to output data vector of size nx containing
- (x+y) if scale = 0
- (x+y) /2 if scale = 1
nx Number of elements of input and output vectors.

nx ≥ 4
scale Scale selection
- If scale = 1, divide the result by 2 to prevent overflow
- If scale = 0, do not divide by 2


atan2_16
Description This function adds two vectors, element by element.
Algorithm for (i + 0; i t nx; i ) )) z(i) + x(i) ) y(i)
Overflow Handling Methodology Scaling implemented for overflow prevention (user selectable)
Special Requirements none
Implementation Notes none
Example See examples/add subdirectory
Cycles† Core: 3 * nx
Overhead: 23
Code size 60
(in bytes)
atan2_16 Arctangent 2 Implementation
Function ushort oflag = atan2_16 (DATA *q, DATA *i, DATA *r, ushort nx)
(defined in arct2.asm)
Arguments
q[nx] Pointer to quadrature input vector of size nx.
i[nx] Pointer to in-phase input vector of size nx
r[nx] Pointer to output data vector (in Q15 format) number
representation of size nx containing. In-place processing
allowed (r can be equal to x ) on output, r contains the
arctangent of (i/q) /π

4-10
atan16
Description This function calculates the arctangent of the ratio i/q, where −1 <= atan2_16
(i/q) <= 1 representing an actual range of −π < atan2_16 (i/q) < π. The result
is placed in the resultant vector r. Output scale factor correction = π. For
example, if:
y = [0x1999, 0x1999, 0x0, 0xe667, 0x1999] (equivalent to [0.2, 0.2, 0, −0.2,
0.2] float)
x = [0x1999, 0x3dcc, 0x7ffff, 0x3dcc c234] (equivalent to [0.2, 0.4828, 1,
0.4828, –0.4828] float)
atan2_16(y, x, r,4) should give:
r = [0x2000, 0x1000, 0x0, 0xf000, 0x7000] equivalent to [0.25, 0.125, 0,
–0.125, 0.875]*π
Algorithm for (j + 0; j t nx; j ) )) r(j) + atan2(i[j], q[j])
Overflow Handling Methodology Not applicable
Special Requirements Linker command file: you must allocate .data section (for polynomial
coefficients)
Example See examples/arct2 subdirectory
Cycles† 18 + 62 * nx
Code size 170 program; 10 data; 4 stack
(in bytes)
atan16 Arctangent Implementation
Function ushort oflag = atan16 (DATA *x, DATA *r, ushort nx)
(defined in atant.asm)
Arguments
x[nx] Pointer to input data vector of size nx. x contains the tangent
of r, where |x| < 1.
r[nx] Pointer to output data vector of size nx containing the

arctangent of x in the range [−π/4, π/4] radians. In-place
processing allowed (r can be equal to x)
atan(1.0) = 0.7854 or 6478h

atan16

Description This function calculates the arc tangent of each of the elements of vector x. The
result is placed in the resultant vector r and is in the range [−π/2 to π/2] radians.
For example,
if x = [0x7fff, 0x3505, 0x1976, 0x0] (equivalent to tan(π/4), tan(π/8), tan(π/16),
0 in float):
atan16(x,r,4) should give
r = [0x6478, 0x3243, 0x1921, 0x0] equivalent to [π/4, π/8, π/16, 0]
Algorithm for (i + 0; i t nx; i ) )) r(i) + atan(x(i))
Special Requirements Linker command file: you must allocate .data section (for polynomial
coefficients)
- atan(x), with 0 v x v 1, output scaling factor + p.
- Uses a polynomial to compute the arctan (x) for |x| <1. For |x| > 1, you can
express the number x as a ratio of 2 fractional numbers and use the
atan2_16 function.
Example See examples/atant subdirectory
Cycles† 14 + 8 * nx
Code size 43 program; 6 data
(in bytes)
4-12
bexp
bexp Block Exponent Implementation
Function short r = bexp (DATA *x, ushort nx)

(defined in bexp.asm)
Arguments
x [nx] Pointer to input vector of size nx
r Return value. Maximum exponent that may be used in
scaling.
nx Length of input data vector
Description Computes the exponents (number of extra sign bits) of all values in the input
vector and returns the minimum exponent. This will be useful in determining
the maximum shift value that may be used in scaling a block of data.
Algorithm Not applicable
Example See examples/bexp subdirectory
Cycles Core: 3 * nx
Overhead: 4
Code size 19
(in bytes)

cbrev
cbrev Complex Bit Reverse
Function void cbrev (DATA *, DATA *r, ushort)

(defined in cbrev.asm)
Arguments
x[2*nx] Pointer to complex input vector x.
r[2*nx] Pointer to complex output vector r.
nx Number of complex elements of vectors x and r.
- To bit-reverse the input of a complex FFT, nx should be the
complex FFT size.
- To bit-reverse the input of a real FFT, nx should be half the
real FFT size.
Description This function bit-reverses the position of elements in complex vector x into out-
put vector r. In-place bit-reversing is allowed. Use this function in conjunction
with FFT routines to provide the correct format for the FFT input or output data.
If you bit-reverse a linear-order array, you obtain a bit-reversed order array. If
you bit-reverse a bit-reversed order array, you obtain a linear-order array.
Special Requirements
- Input vector x[ ] and output vector r[ ] must be aligned on 32−bit boundary.
(2 LSBs of byte address must be zero)
- Ensure that the entire array fits within a 64K boundary (the largest possible
array addressable by the 16-bit auxiliary register).
- in place bit−reversal has better performance.
Example See examples/cfft and examples/rfft subdirectories
4-14
cbrev32
FFT Size Cycles†

8 107
16 128
32 150
64 222
128 310
256 554
512 918
1024 1794
† Assumes all data is in on-chip dual-access RAM and that there is no bus conflict due to twiddle table reads and instruction
fetches (provided linker command file reflects those conditions).
cbrev32 32-Bit Complex Bit Reverse
Function void cbrev32(LDATA *, LDATA *r, ushort)

(defined in cbrev32.asm)
Arguments
x[2*nx] Pointer to complex input vector x.
r[2*x] Pointer to complex output vector r.
nx Number of complex elements in vector x.
- To bit-reverse the output of a complex (i)FFT, nx should be
the complex (i)FFT size.
- To bit-reverse the output of a real (i)FFT, nx should be half
the real (i)FFT size.
Description This function bit-reverses the position of elements in complex vector x into out-
put vector r. In-place bit-reversing is allowed. Use this function in conjunction
with (i)FFT routines to provide the correct format for the (i)FFT input or output
data. If you bit-reverse a linear-order array, you obtain a bit-reversed order
array. If you bit-reverse a bit-reversed order array, you obtain a linear-order
array.

cfft

- in place bit−reversal has better performance.
Implementation Notes x is read in normal linear addressing and r is written with bit-reversed address-
ing.
Example See example/c(i)fft subdirectory
Benchmarks
Cycles† Core:
5*nx (off-place)
11*nx (in-place)
Code size 75 (includes support for both in-place and off-place

(in bytes) bit-reverse)
cfft Forward Complex FFT
Function void cfft (DATA *x, ushort nx, type);

(defined in cfft.asm)
Arguments
x [2*nx] Pointer to input vector containing nx complex elements (2*nx
real elements) in normal order. On output, vector contains
the nx complex elements of the FFT(x) in bit-reversed order.
Complex numbers are stored in interleaved Re-Im format.
nx Number of complex elements in vector x. Must be between 8

and 1024.
type FFT type selector. Types supported:
- If type = SCALE, scaled version selected
- If type = NOSCALE, non-scaled version selected
Description Computes a complex nx-point FFT on vector x, which is in normal order. The
original content of vector x is destroyed in the process. The nx complex ele-
ments of the result are stored in vector x in bit-reversed order. The twiddle table
is in bit-reversed order.
4-16
cfft
Algorithm (DFT)
ȍ x [i] < ǒcos ǒ* 2 *nxp * i * kǓ ) j sin ǒ* 2 *nxp * i * kǓǓ

nx*1
y [k] + 1 <
(scale factor) i+0
Overflow Handling Methodology If type = SCALE is selected, scaling before each stage is imple-
mented for overflow prevention
- The twiddle table must be located in internal memory since it is accerred

by the C55x coefficient bus.
- Input data section is aligned on 32-bit boundary.
- For the best performance:
J Input data in DARAM block

J Twiddle table in SARAM block or DARAM block different than the
DARAM clock that contains the input data.
- Ensure that the entire input array fits within a 64K boundary (the largest
possible array addressable by the 16-bit auxiliary register).
- If the twiddle table and the data buffer are in the same block then the ra-
dix-2 kernal is 7 cycles and the radix-4 kernel is not affected.
- The implementations are optimized for MIPS, not for code size. They im-
plement the decimation-in-time (DIT) FFT algorithm.
- The NOSCALE version is implemented using radix-2 butterflies. The first

two stages are replaced by a single radix-4 stage.
- The SCALE version is implemented using only radix-2 stages. This routine
prevents overflow by scaling by 2 before each FFT stage.
Example See examples/cfft subdirectory
Benchmarks
- 5 cycles (radix-2 butterfly − used in both SCALE and NOSCALE versions)
- 10 cycles (radix-4 butterfly – used in the first 2 stages of a non-scaled

version)

cfft
Comparing the results to MATLAB:

J NOSCALE version
C55 DSPLIB MATLAB
Cfft( ) xN
Cfft( )
NOSCALE
The MATLAB cfft results need to be multiplied by the cfft size, N, in order to
be compared to the C55 DSPLIB cfft results.
J SCALE version
C55 DSPLIB MATLAB
Cfft( )
Cfft( )
SCALE
The C55 DSPLIB cfft results can be compared to the unmodified MATLAB cfft
results.
CFFT − SCALE
FFT Size Cycles† Code Size (in bytes)

8 208 493
16 358 493
32 624 493
64 1210 493
128 2516 493
256 5422 493
512 11848 493
1024 25954 493
4-18
cfft32
CFFT − NOSCALE

16 286 359
32 517 359
64 1036 359
128 2211 359
256 4858 359
512 10769 359
1024 23848 359

cfft32 32-Bit Forward Complex FFT
Function void cfft32 (LDATA *x, ushort nx, type);
Arguments
x[2*nx] Pointer to input vector containing nx complex elements (2*nx
real elements) in normal-order. On output, vector x contains
the nx complex elements of the FFT(x) in bit-reversed order.
Complex numbers are stored in the interleaved Re-Im
format.
and 1024.

Description Computes a complex nx-point FFT on vector x, which is in normal order. The
ments of the result are stored in vector x in bit-reversed order.
Algorithm (DFT)
ȍ x[i] < ǒcosǒ2 * pnx* i * kǓ * j sinǒ2 * pnx* i * kǓǓ
nx*1
y[k] + 1 <
(scale factor) i+0

cfft32
Overflow Handling Methodology If scale==1, scaling before each stage is implemented for over-
flow prevention.
- The twiddle table must be located in the internal memory since it is ac-
cerred by the C55x coefficient bus.
- For best performance, the data buffer has to be in a DARAM block.
- For best performance, the coefficient buffer can be in SARAM block or a

DARAM different from the DARAM block that contains the data buffer.
- Radix-2 DIT version of the FFT algorithm is implemented. The imple-
mentation is optimized for MIPS, not for code size.
Example See example/cfft32 subdirectory
Benchmarks
- 12 cycles for radix-2 butterfly in non-scaled version; 15 cycles for radix-2
butterfly in scaled version
- 21 cycles for radix-4 butterfly in non-scaled version
- 10 cycles for stage 1 loop in scaled version; 10 cycles for group 1 of stage
2 loop in scaled version; 13 cycles for group 2 of stage 2 in scaled version
CFFT32 − SCALE

16 715 504
32 1712 504
64 4038 504
128 9412 504
256 21618 504
512 48960 504

4-20
cfir
CFFT – NOSCALE

16 601 337
32 1461 337
64 3460 337
128 8083 337
256 18594 337
512 42161 337

cfir Complex FIR Filter
Function ushort oflag = cfir (DATA *x, DATA *h, DATA *r, DATA *dbuffer, ushort nx,
ushort nh)
Arguments
x[2*nx] Pointer to input vector of nx complex elements.
h[2*nh] - Pointer to complex coefficient vector of size nh in
normal order. For example, if nh=6, then h[nh] =
{h0r, h0i, h1r, h1i h2r, h2i, h3r, h3i, h4r, h4i, h5r, h5i}
where h0 resides at the lowest memory address in
the array.
- This array must be located in internal memory since
it is accessed by the C55x coefficient bus.
r[2*nx] Pointer to output vector of nx complex elements.
In-place computation (r = x) is allowed.

cfir
dbuffer[2*nh + 2] Pointer to delay buffer of length 2 * nh + 2

- In the case of multiple-buffering schemes, this array
should be initialized to 0 for the first filter block only.
Between consecutive blocks, the delay buffer pre-
serves the previous r output elements needed.
- The first element in this array is present for align-
ment purposes, the second element is special in
that it contains the array index−1 of the oldest input
entry in the delay buffer. This is needed for multiple-
buffering schemes, and should be initialized to 0
(like all the other array entries) for the first block only.
nx Number of complex input samples
nh The number of complex coefficients of the filter. For
example, if the filter coefficients are {h0, h1, h2, h3,
h4, h5}, then nh = 6. Must be a minimum value of 3.
For smaller filters, zero pad the coefficients to meet
the minimum value.
oflag Overflow error flag (returned value)
- If oflag = 1, a 32-bit data overflow has occurred in an
intermediate or final result.
- If oflag = 0, a 32-bit overflow has not occurred.
Description Computes a complex FIR filter (direct-form) using the coefficients stored in
vector h. The complex input data is stored in vector x. The filter output result
is stored in vector r. This function maintains the array dbuffer containing the
previous delayed input values to allow consecutive processing of input data
blocks. This function can be used for both block-by-block (nx ≥ 2) and sample-
by-sample filtering (nx = 1). In-place computation (r = x) is allowed.
ȍ h [k] x [j * k]
nh*1
Algorithm r [j] + 0 v j v nx
k+0
Overflow Handling Methodology No scaling implemented for overflow prevention.
- nh must be a minimum value of 3. For smaller filters, zero pad the h[ ] array.
- Coefficient array h[ ] is located in the internal memory.
- Input array x[ ] must be aligned on a 32−bit boundary (2 LSBs of byte ad-

dress must be zero).
4-22
cfir
- Delay buffer dbuffer[ ] must be aligned on a 32−bit boundary (2 LSBs of

byte address must be zero).
Implementation Notes The first element in the dbuffer array is present only for alignment purposes.
The second element in this array (index=0) is the entry index for the input
history. It is treated as an unsigned 16-bit value by the function even though
it has been declared as signed in C. The value of the entry index is equal to
the index − 1 of the oldest input entry in the array. The remaining elements
make up the input history. Figure 4−1 shows the array in memory with an entry
index of 2. The newest entry in the dbuffer is denoted by x(j−0), which in this
case would occupy index = 3 in the array. The next newest entry is x(j−1), and
so on. It is assumed that all x() entries were placed into the array by the
previous invocation of the function in a multiple-buffering scheme.
Figure 4−1, Figure 4−2, and Figure 4−3 show the dbuffer, x, and r arrays as
they appear in memory.

cfir
Figure 4−1. dbuffer Array in Memory at Time j

dummy value lowest memory address
entry index = 2
xr(j−nh−2)
xi(j−nh−2)
xr(j−nh−1)
xi(j−nh−1)
oldest x( ) entry xr(j−nh)
xi(j−nh)
newest x( ) entry xr(j−0)
xi(j−0)
xr(j−1)
xi(j−1)
xr(j−2)
xi(j−2)
•
•
•
xr(j−nh−3)
xi(j−nh−3)
xr(j−nh−4)
xi(j−nh−4)
xr(j−nh−3)
xi(j−nh−3) highest memory address
4-24
cfir
Figure 4−2. x Array in Memory
oldest x( ) entry xr(0) lowest memory address

xi(0)
xr(1)
xi(1)
•
•
•
xr(nx−2)
xi(nx−2)
xr(nx−1)
newest x( ) entry xi(nx−1) highest memory address
Figure 4−3. r Array in Memory
oldest x( ) entry rr(0) lowest memory address

ri(0)
rr(1)
ri(1)
•
•
•
rr(nx−2)
ri(nx−2)
rr(nx−1)
newest x( ) entry ri(nx−1) highest memory address
Example See examples/cfir subdirectory
Cycles† Core: nx * [8 + 2(nh−2)]
Overhead: 51
Code size 136
(in bytes)

cifft
cifft Inverse Complex FFT
Function void cifft (DATA *x, ushort nx, type);

(defined in cifft.asm)
Arguments
x [2*nx] Pointer to input vector containing nx complex elements (2*nx
real elements) in normal order. On output, vector contains
the nx complex elements of the IFFT(x) in bit-reversed order.
Complex numbers are stored in interleaved Re-Im format.

and 1024.

Description Computes a complex nx-point IFFT on vector x, which is in normal order. The
Algorithm (IDFT)
ȍ x [i] < ǒcosǒ2 * pnx* i * kǓ ) j sin ǒ2 * pnx* i * kǓǓ

nx*1
y [k] + 1 <
(scale factor) i+0
Overflow Handling Methodology If type = SCALE is selected, scaling before each stage is imple-
mented for overflow prevention
- The twiddle table must be located in internal memory since it is accessed

by the C55x coefficient bus.
- Input data section is aligned on 32-bit boundary.
- For the best performance:

J Input data in DARAM block.
J Twiddle table in SARAM block or DARAM block different than the DA-
RAM clock that contains the input data.
- If the twiddle table and the data buffer are in the same block then the ra-
dix-2 kernal is 7 cycles and the radix-4 kernel is not affected.
4-26
cifft
- The implementations are optimized for MIPS, not for code size. They im-
plement the decimation-in-time (DIT) FFT algorithm.
- The NOSCALE version is implemented using radix-2 butterflies. The first

two stages are replaced by a single radix-4 stage.
- The SCALE version is implemented using only radix-2 stages. This routine
prevents overflow by scaling by 2 before each FFT stage.
Example See examples/cifft subdirectory
- 5 cycles (radix-2 butterfly − used in both SCALE and NOSCALE versions)
- 10 cycles (radix-4 butterfly – used in NOSCALE version)
CIFFT − SCALE

8 208 494
16 358 494
32 624 494
64 1210 494
128 2516 494
256 5422 494
512 11848 494
1024 25954 494


cifft32
CFFT − NOSCALE

16 281 355
32 512 355
64 1031 355
128 2206 355
256 4853 355
512 10764 355
1024 23843 355

cifft32 32-Bit Inverse Complex FFT
Function void cifft32 (LDATA *x, ushort nx, type);
Arguments
x[2*nx] Pointer to input vector containing nx complex elements (2*nx
real elements) in normal-order. On output, vector x contains
the nx complex elements of the iFFT(x) in bit-reversed order.
Complex numbers are stored in the interleaved Re-Im
format.
and 1024.

Description Computes a complex nx-point iFFT on vector x, which is in normal-order. The

Algorithm (iDFT)
ȍ x[i] < ǒcosǒ2 * pnx* i * kǓ ) j sinǒ2 * pnx* i * kǓǓ

nx*1
y[k] + 1 <
(scale factor) i+0
4-28
cifft32
Overflow Handling Methodology If scale == 1, scaling before each stage is implemented for over-
flow prevention.
- The twiddle table must be located in the internal memory since it is ac-
cerred by the C55x coefficient bus.
- For best performance, the data buffer has to be in a DARAM block.
- For best performance, the coefficient buffer can be in an SARAM block or

a DARAM different from the DARAM block that contains the data buffer.
- Radix-2 DIT version of the iFFT algorithm is implemented. The imple-
mentation is optimized for MIPS, not for code size.
Example See example/cifft32 subdirectory
Benchmarks
- 12 cycles for radix-2 butterfly in non-scaled version; 15 cycles for radix-2
butterfly in scaled version
- 21 cycles for radix-4 butterfly in non-scaled version
- 10 cycles for stage 1 loop in scaled version; 10 cycles for group 1 of stage
2 loop in scaled version; 13 cycles for group 2 of stage 2 in scaled version
CIFFT32 − SCALE
iFFT Size Cycles† Code Size (in bytes)

16 715 504
32 1712 504
64 4038 504
128 9412 504
256 21618 504
512 48960 504


cifft32
CFFT32 − NOSCALE
iFFT Size Cycles† Code Size (in bytes)

16 601 337
32 1461 337
64 3460 337
128 8083 337
256 18594 337
512 42161 337

4-30
convol
convol Convolution
Function ushort oflag = convol (DATA *x, DATA *h, DATA *r, ushort nr, ushort nh)
Arguments
x[nr+nh−1] Pointer to input vector of nr + nh − 1 real elements.
h[nh] Pointer to input vector of nh real elements.
r[nr] Pointer to output vector of nr real elements.
nr Number of elements in vector r. In-place computation (r = x)
is allowed (see Description section for comment).
nh Number of elements in vector h.
- If oflag = 1, a 32-bit data overflow occurred in an inter-
mediate or final result.
Description Computes the real convolution of two real vectors x and h, and places the
results in vector r. Typically used for block FIR filter computation when there
is no need to retain an input delay buffer. This function can also be used to
implement single-sample FIR filters (nr = 1) provided the input delay history
for the filter is maintained external to this function. In-place computation (r = x)
is allowed, but be aware that the r output vector is shorter in length than the
x input vector; therefore, r will only overwrite the first nr elements of the x.
ȍ h [k] x [j * k]
nh*1
Algorithm r [j] + 0 v j v nr
k+0
Implementation Notes Figure 4−4, Figure 4−5, and Figure 4−6 show the x, r, and h arrays as they
appear in memory.

convol

x(0) lowest memory address
x(1)
•
•
•
x(nr+nh−2)
x(nr+nh−1) highest memory address

r(0) lowest memory address
r(1)
•
•
•
r(nr−2)
r(nr−1) highest memory address
Figure 4−6. h Array in Memory

h(0) lowest memory address
h(1)
•
•
•
h(nh−2)
h(nh−1) highest memory address
Example See examples/convol subdirectory
Cycles† Core: nr * (1 + nh)
Overhead: 44
Code size 88
(in bytes)
4-32
convol1
convol1 Convolution (fast)
Function ushort oflag = convol1 (DATA *x, DATA *h, DATA *r, ushort nr, ushort nh)
Arguments
x[nr+nh−1] Pointer to input vector of nr+nh−1 real elements.
r[nr] Pointer to output vector of nr real elements. In-place
computation (r = x) is allowed (see Description section for
comment).
nr Number of elements in vector r. Must be an even number.
results in vector r. This function utilizes the dual-MAC capability of the C55x
to process in parallel two output samples for each iteration of the inner function
loop. It is, therefore, roughly twice as fast as CONVOL, which implements only
a single-MAC approach. However, the number of output samples (nr) must be
even. Typically used for block FIR filter computation when there is no need to
retain an input delay buffer. This function can also be used to implement single-
sample FIR filters (nr = 1) provided the input delay history for the filter is main-
tained external to this function. In-place computation (r = x) is allowed, but be
aware that the r output vector is shorter in length than the x input vector; there-
fore, r will only overwrite the first nr elements of the x.
ȍ h [k] x [j * k]
nh*1
k+0

- nr must be an even value.
- The vector h[nh] must be located in internal memory since it is accessed

using the C55x coefficient bus, and that bus does not have access to exter-
nal memory.
appear in memory.

convol1

x(1)
•
•
•
x(nr+nh−2)

r(1)
•
•
•
r(nr−2)

h(1)
•
•
•
h(nh−2)
Example See examples/convol1 subdirectory
Cycles† Core: nr/2 * [3+(nh−2)]
Overhead: 58
Code size 101
(in bytes)
4-34
convol2
convol2 Convolution (fastest)
Function ushort oflag = convol2 (DATA *x, DATA *h, DATA *r, ushort nr, ushort nh)
Arguments
x[nr+nh−1] Pointer to input vector of nr + nh − 1 real elements.
r[nr] Pointer to output vector of nr real elements. In-place
computation (r = x) is allowed (see Description section for
comment). This array must be aligned on a 32-bit boundary
in memory.
nr Number of elements in vector r. Must be an even number.

- If oflag = 1, a 32-bit data overflow has occurred in an inter-
results in vector r. This function utilizes the dual-MAC capability of the C55x
to process in parallel two output samples for each iteration of the inner function
loop. It is, therefore, roughly twice as fast as CONVOL, which implements only
a single-MAC approach. However, the number of output samples (nr) must be
even. In addition, this function offers a small performance improvement over
CONVOL1 at the expense of requiring the r array to be 32-bit aligned in memo-
ry. Typically used for block FIR filter computation when there is no need to
retain an input delay buffer. This function can also be used to implement single-
sample FIR filters (nr = 1) provided the input delay history for the filter is main-
tained external to this function. In-place computation (r = x) is allowed, but be
aware that the r output vector is shorter in length than the x input vector; there-
fore, r will only overwrite the first nr elements of the x.
ȍ h [k] x [j * k]
nh*1
k+0

convol2
- nr must be an even value.
- The vector h[nh] must be located in internal memory since it is accessed

using the C55x coefficient bus, and that bus does not have access to exter-
nal memory.
- The vector r[nr] must be 32-bit aligned in memory.
appear in memory.

x(1)
•
•
•
x(nr+nh−2)
Figure 4−11.r Array in Memory

r(1)
•
•
•
r(nr−2)

h(1)
•
•
•
h(nh−2)
4-36
corr
Example See examples/convol2 subdirectory
Cycles† Core: nr/2 * (1 + nh)
Overhead: 24
Code size 100
(in bytes)
corr Correlation, full-length
Function ushort oflag = corr (DATA *x, DATA *y, DATA *r, ushort nx, ushort ny, type)
Arguments
x [nx] Pointer to real input vector of nx real elements.
y [ny] Pointer to real input vector of ny real elements.
r[nx+ny−1] Pointer to real output vector containing the full-length
correlation (nx + ny − 1 elements) of vector x with y. r
must be different than both x and y (in-place
computation is not allowed).

ny Number of real elements in vector y
type Correlation type selector. Types supported:
- If type = raw, r contains the raw correlation
- If type = bias, r contains the biased-correlation
- If type = unbias, r contains the unbiased-correlation
oflag Overflow flag
Description Computes the full-length correlation of vectors x and y and stores the result
in vector r. using time-domain techniques.

corr
Algorithm Raw correlation

nr*j*1
r[j] + ȍ x [j ) k] * y [k] 0 v j v nr + nx ) ny * 1
k+o
Biased correlation
nr*j*1
1
r[j] + nr ȍ x [j ) k] * y[k] 0 v j v nr + nx ) ny * 1
k+o
Unbiased correlation
nr*j*1
r[j] + 1
(nx * abs(j))
ȍ x [j ) k] * y [k] 0 v j v nr + nx ) ny * 1
k+o
- x array located in the internal memory because it is accessed by the C55

coefficient bus.
- Requirements for nx,ny
J nx y
J ny nx
- Special debugging consideration: This function is implemented as a
macro that invokes different correlation routines according to the type
selected. As a consequence the corr symbol is not defined. Instead the
corr_raw, corr_bias, corr_unbias symbols are defined.
- Correlation is implemented using time-domain techniques
Cycles Raw: 2 times faster than C54x
Unbias: 2.14 times faster than C54x
Bias: 2.1 times faster than C54x
Code size Raw: 318
(in bytes) Unbias: 417
Bias: 356
4-38
dlms
dlms Adaptive Delayed LMS Filter
Function ushort oflag = dlms (DATA *x, DATA *h, DATA *r, DATA *des, DATA *dbuffer,
DATA step, ushort nh, ushort nx)
(defined in dlms.asm)
Arguments
x[nx] Pointer to input vector of size nx
h[nh] Pointer to filter coefficient vector of size nh.
- h is stored in reversed order : h(n−1), ... h(0) where h[n]
is at the lowest memory address.
- Memory alignment: h is a circular buffer and must start
in a k-bit boundary(that is, the k LSBs of the starting ad-
dress must be zeros) where k = log2(nh)
r[nx] Pointer to output data vector of size nx. r can be equal to
x.
des[nx] Pointer to expected output array

dbuffer[nh+2] Pointer to the delay buffer structure.
The delay buffer is a structure comprised of an index
register and a circular buffer of length nh + 1. The index
register is the index into the circular buffer of the oldest
data sample.
nh Number of filter coefficients. Filter order = nh − 1.
nh ≥ 3
nx Length of input and output data vectors
Description Adaptive delayed least-mean-square (LMS) FIR filter using coefficients stored
in vector h. Coefficients are updated after each sample based on the LMS
algorithm and using a constant step = 2*μ. The real data input is stored in vec-
tor dbuffer. The filter output result is stored in vector r .
LMS algorithm uses the previous error and the previous sample (delayed) to
take advantage of the C55x LMS instruction.

dlms
The delay buffer used is the same delay buffer used for other functions in the
C55x DSP Library. There is one more data location in the circular delay buffer
than there are coefficients. Other C55x DSP Library functions use this delay
buffer to accommodate use of the dual-MAC architecture. In the DLMS func-
tion, we make use of the additional delay slot to allow coefficient updating as
well as FIR calculation without a need to update the circular buffer in the interim
operations.
The FIR output calculation is based on x(i) through x(i−nh+1). The coefficient
update for a delayed LMS is based on x(i−1) through x(i−nh). Therefore, by
having a delay buffer of nh+1, we can perform all calculations with the given
delay buffer containing delay values of x(i) through x(i−nh). If the delay buffer
was of length nh, the oldest data sample, x(i−nh), would need to be updated
with the newest data sample, x(i), sometime after the calculation of the first co-
efficient update term, but before the calculation of the last FIR term.
Algorithm FIR portion
ȍ h [k] * x [i * k]
nh*1
r [j] + 0 v i v nx * 1
k+0
Adaptation using the previous error and the previous sample:

e (i) + des (i * 1) * r (i * 1)
h k (i ) 1) + h k (i) ) 2 * m * e(i * 1) * x(i * k * 1)
Special Requirements Minimum of 2 input and desired data samples. Minimum of 2 coefficients
- Delayed version implemented to take advantage of the C55x LMS instruc-
tion.
- Effect of using delayed error signal on convergence minimum:

For reference, the following is the algorithm for the regular LMS (non-
delayed):
FIR portion
ȍ h [k] * x [i * k]
nh*1
r [j] + 0 v i v nx * 1
k+0
Adaptation using the current error and the current sample:

e (i) + des (i) * r (i)
h k (i ) 1) + h k (i) ) 2 * m * e(i) * x(i * k)
4-40
dlmsfast
Example See examples/dlms subdirectory
Cycles† Core: nx * (7 + 2*(nh − 1)) = nx * (5 + 2 * nh)
Overhead: 26
Code size 122
(in bytes)
dlmsfast Adaptive Delayed LMS Filter (fast implemented)
Function ushort oflag = dlmsfast (DATA *x, DATA *h, DATA *r, DATA *des, DATA *dbuffer,
DATA step, ushort nh, ushort nx)
This function is implemented for better performance on large number of filter
orders.
(defined in dlmsfast.asm)
Arguments
x[nx] Pointer to input vector of size nx.
h[2*nh] Pointer to filter coefficient array of size 2*nh. This array
contains two coefficient buffers h_coef and h_scratch.
The upated coefficients in different time slot are stored
into these two buffers alternatively. The final updated
coefficients are stored in h_coef.
- h_coef is stored in reversed order: h_coef(n−1), ...
h_coef(0) where h_coef(n−1) is at the lowest memory
address of the first half of array h.
- h_scratch is stored in reversed order : h_scratch(n−1),
... h_scratch(0) where h_scratch(n−1) is at the lowest
memory address of the second half of array h.
- Memory alignment: h must be aligned in 32 bytes
boundary.
r[nx] Pointer to output data vector of size nx. r can be equal to
x.
des[nx] Pointer to expected output array.

dlmsfast
dbuffer[nh+3] Pointer to the delay buffer structure.

- The delay buffer is a structure comprised of an index
register and a circular buffer of length nh+2. The index
register is the index into the circular buffer of the oldest
data sample.
- Memory alignment: dbuffer must be aligned in 32 bytes
boundary.
nh Number of filter coefficients. Filter order = nh−1. nh has
to be a even number. nh ≥ 10.
nx Length of input and output data vectors. nx has to be a

even number.

Description Adaptive delayed least-mean-square (LMS) FIR filter using coefficients stored
in vector h. Coefficients are updated after each sample based on the LMS al-
gorithm and using a constant step = 2*μ. The real data input is stored in vector
dbuffer. The filter output result is stored in vector r.
Unlike the DLMS function in DSPLIB, which uses C55x LMS instruction to do
partial filtering and addition of delta h to the coefficient, this fast LMS algorithm
is implemented by doing coefficient updating and filtering separately to get bet-
ter cycle count.
In this implementation, two input data are processed as a pair. The filtering op-
eration uses dual-MAC to process two time slots of data and two set of coeffi-
cients are updated corresponding to these two time slots.
The delay buffer used is the same delay buffer used for other functions in the
C55x DSP Library. There is two more data location in the circular delay buffer
than there are coefficients. Other C55x DSP Library functions use this delay
buffer to accommodate use of the dual-MAC architecture. In the DLMS func-
tion, we make use of the additional delay slots to allow coefficient updating as
well as FIR calculation without a need to update the circular buffer in the interim
operations.
4-42
dlmsfast
The first time slot of FIR output calculation is based on x(i) through x(i−nh+1).
While the coefficient update for a delayed LMS is based on x(i−1) through
x(i−nh). The second time slot of FIR output is based on x(i+1) through
x(i−nh+2). While the coefficient update for the delayed LMS is based on x(i)
through x(i−nh+1). Therefore, by having a delay buffer of nh+2, we can per-
form all calculations with the given delay buffer containing delay values of x(i)
through x(i−nh+1).
Algorithm FIR portion:

nh*1
r[i] + ȍ h[k] < x[i * k] 0 v i v nx * 1
k+0
Adaptation using the previous error and the previous sample:

e(i) + des(i * 1) * r(i * 1)
h k (i ) 1) + h k (i) ) 2 * m * e(i * 1) * x(i * k * 1)
- Delay buffer array dbugger[ ] must be locaed in the internal memory.
- Minimum of 10 coefficients. Coefficient buffer need to be aligned on 32−bit

boundary (2 LSBs of byte address must be zero).
- dbuffer need to be aligned on 32 bytes memory boundary.
- Coefficient buffer and dbuffer need to be put into different block of memory
for the best performance.
- Filtering and coefficient updating are implemented separately.
Figure 4−13, Figure 4−14, and Figure 4−15 show the x buffer, dbuffer, and
h buffers.
Figure 4−13. x Buffer
oldest x( ) entry x(0) lowest memory address

x(1)
•
•
•
x(nx−2)
newest x( ) entry x(nx−1) highest memory address

dlmsfast
Figure 4−14. dbuffer
entry index = 0
newest x( ) entry x(1)
oldest x( ) entry x(−(nh+1))
x(−nh)
•
•
•
x(−2)
x(−1) highest memory address
Figure 4−15. h Buffers
h_coefs h_coef(nh−1) lowest memory address

h_coef(nh−2)
•
•
•
h_coef(0)
h_scratch h_scratch(nh−1)
h_scratch(nh−2)
•
•
•
h_scratch(0) highest memory address
- Effect of using delayed error signal on convergence minimum. For refer-

ence, the following is the algorithm for the regular LMS (non-delayed):
FIR portion
ȍ h[k] < x[i * k]

nh*1
r[i] + 0 v i v nx * 1
k+0
Adaptation using the current error and the current sample

e(i) + des(i) * r(i)
h k (i ) 1) + h k (i) ) 2 * m * e(i) * x(i * k)
Example See examples/dlmsfast subdirectory
4-44
expn
Benchmarks
Cycles† Core: nx/2 * (26 + 3*nh)
Overhead: 71
Code size 322
(in bytes)
expn Exponential Base e
Function ushort oflag = expn (DATA *x, DATA *r, ushort nx)
(defined in expn.asm)
Arguments
x[nx] Pointer to input vector of size nx. x contains the numbers
normalized between (−1,1) in q15 format
r[nx] Pointer to output data vector (Q3.12 format) of size nx. r can
be equal to x.

Description Computes the exponent of elements of vector x using Taylor series.
Algorithm for (i + 0; i t nx; i ) )) y(i) + e x(i) where −1 t x(i) t 1
Special Requirements Linker command file: you must allocate .data section (for polynomial coeffi-
cients) on a 32−bit boundary (2 LSBs of byte address must be zero).

fir
Implementation Notes Computes the exponent of elements of vector x. It uses the following Taylor
series:
exp(x) + c0 ) (c1 * x) ) (c2 * x 2) ) (c3 * x 3) ) (c4 * x 4) ) (c5 * x 5)
where
c0 = 1.0000
c1 = 1.0001
c2 = 0.4990
c3 = 0.1705
c4 = 0.0348
c5 = 0.0139
Example See examples/expn subdirectory
Overhead: 18
Code size 57
(in bytes)
fir FIR Filter
Function ushort oflag = fir (DATA *x, DATA *h, DATA *r, DATA *dbuffer, ushort nx,
ushort nh)
Arguments
x[nx] Pointer to input vector of nx real elements.
h[nh] - Pointer to coefficient vector of size nh in normal order.
For example, if nh=6, then h[nh] = {h0, h1, h2, h3, h4,
h5} where h0 resides at the lowest memory address in
the array.
r[nx] Pointer to output vector of nx real elements. In-place

computation (r = x) is allowed.
4-46
fir
dbuffer[nh+2] Pointer to delay buffer of length nh = nh + 2

should be initialized to 0 for the first filter block only. Be-
tween consecutive blocks, the delay buffer preserves
the previous elements needed.
- The first element in this array is special in that it con-
tains the array index−1 of the oldest input entry in the
delay buffer. This is needed for multiple-buffering
schemes, and should be initialized to 0 (like all the oth-
er array entries) for the first block only.
nx Number of input samples
nh The number of coefficients of the filter. For example, if
the filter coefficients are {h0, h1, h2, h3, h4, h5}, then nh
= 6. Must be a minimum value of 3. For smaller filters,
zero pad the coefficients to meet the minimum value.

Description Computes a real FIR filter (direct-form) using the coefficients stored in vector
h. The real input data is stored in vector x. The filter output result is stored in
vector r. This function maintains the array dbuffer containing the previous
delayed input values to allow consecutive processing of input data blocks. This
function can be used for both block-by-block (nx ≥ 2) and sample-by-sample
filtering (nx = 1). In place computation (r = x) is allowed.
ȍ h [k] x [j * k]
nh*1
k+0
Special Requirements nh must be a minimum value of 3. For smaller filters, zero pad the h[ ] array.

fir
Implementation Notes The first element in the dbuffer array (index = 0) is the entry index for the input
make up the input history. Figure 4−16 shows the array in memory with an
entry index of 2. The newest entry in the dbuffer is denoted by x(j−0), which
in this case would occupy index = 3 in the array. The next newest entry is
x(j−1), and so on. It is assumed that all x() entries were placed into the array
by the previous invocation of the function in a multiple-buffering scheme.
The dbuffer array actually contains one more history value than is needed to
implement this filter. The value x(j−nh) does not enter into the calculations for
for the output r(j). However, this value is required in other DSPLIB filter func-
tions that utilize the dual-MAC units on the C55x, such as FIR2. Including this
extra location ensures compatibility across all filter functions in the C55x
DSPLIB.
Figure 4−16, Figure 4−17, and Figure 4−18 show the dbuffer, x, and r arrays
as they appear in memory.
entry index = 2 lowest memory address

x(j−nh−2)
x(j−nh−1)
x(j−nh−1)
oldest x( ) entry x(j−nh)
newest x( ) entry x(j−0)
x(j−1)
x(j−2)
•
•
•
x(j−nh−5)
x(j−nh−4)
x(j−nh−3) highest memory address
4-48
fir2

x(1)
•
•
•
x(nx−2)
newest x( ) entry x(nx−1)
highest memory address

oldest x( ) entry r(0) lowest memory address
r(1)
•
•
•
r(nx−2)
newest x( ) entry r(nx−1) highest memory address
Example See examples/fir subdirectory

Cycles† Core: nx * (2 + nh)
Overhead: 25
Code size 107
(in bytes)
† Assumes all data is in on-chip dual-access RAM (provided linker command file reflects those
conditions).
fir2 FIR2 Filter
Function ushort oflag = fir (DATA *x, DATA *h, DATA *r, DATA *dbuffer, ushort nx,
ushort nh)
Arguments
h[nh] - Pointer to coefficient vector of size nh in normal order.
For example, if nh=6, then h[nh] = {h0, h1, h2, h3, h4,
h5} where h0 resides at the lowest memory address in
the array.

fir2

dbuffer[nh+2] Pointer to delay buffer of length nh = nh + 2
should be initialized to 0 for the first filter block only. Be-
the previous elements needed.
- The first element in this array is special in that it con-
tains the array index−1 of the oldest input entry in the
delay buffer. This is needed for multiple-buffering
schemes, and should be initialized to 0 (like all the oth-
er array entries) for the first block only.

nh The number of coefficients of the filter. For example, if
the filter coefficients are {h0, h1, h2, h3, h4, h5}, then nh
= 6. Must be a minimum value of 3. For smaller filters,
zero pad the coefficients to meet the minimum value.

function can be used for both block-by-block (nx ≥ 2) and sample-by-sample
ȍ h [k] x [j * k]
nh*1
k+0
- nh must be a minimum value of 3. For smaller filters, zero pad the h[ ] array.
- array r[ ] must be aligned on 32-bit boundary.
- array h[ ] must be located in internal memory because it is accessed with

the coefficient data pointer, CDP.
4-50
fir2
x(j−1), and so on. Every iteration two entries are updated in the dbuffer array.
It is assumed that all x() entries were placed into the array by the previous in-
vocation of the function in a multiple-buffering scheme.

x(j−nh−2)
x(j−nh−1)
x(j−nh−1)
x(j−1)
x(j−2)
•
•
•
x(j−nh−5)
x(j−nh−4)

x(1)
•
•
•
x(nx−2)

firdec

r(1)
•
•
•
r(nx−2)
Example See examples/fir2 subdirectory

Cycles† Core: nx * (3 + nh/2)
Overhead: 25
Code size 107
(in bytes)
† Assumes all data is in on-chip dual-access RAM (provided linker command file reflects those
conditions).
firdec Decimating FIR Filter
Function ushort oflag = firdec (DATA *x, DATA *h, DATA *r, DATA *dbuffer , ushort nh,
ushort nx, ushort D)
(defined in decimate.asm)
Arguments
h[nh] Pointer to coefficient vector of size nh in normal order:
H = b0 b1 b2 b3 …
r[nx/D] Pointer to real input vector of nx/D real elements.
In-place computation (r = x) is allowed
dbuffer[nh+1] Delay buffer
should be initialized to 0 for the first block only. Be-
previous delayed input samples. It also preserves a
ptr to the next new entry into the dbuffer. This ptr is
preserved across function calls in dbuffer[0].
4-52
firdec

nh Number of coefficients
D Decimation factor. For example a D = 2 means you drop
every other sample. Ideally, nx should be a multiple of
D. If not, the trailing samples will be lost in the process.
oflag Overflow error flag
Description Computes a decimating real FIR filter (direct-form) using coefficient stored in
vector h. The real data input is stored in vector x. The filter output result is
stored in vector r. This function retains the address of the delay filter memory
d containing the previous delayed values to allow consecutive processing of
blocks. This function can be used for both block-by-block and sample-by-
sample filtering (nx = 1).
ȍ h[k]x [j * D * k]
nh
Algorithm r[j] + 0 v j v nx
k+0
Example See examples/decim subdirectory
Cycles Core: (nx/D)*(10+nh+(D−1))
Overhead 67
Code size 144
(in bytes)

firinterp
firinterp Interpolating FIR Filter
Function ushort oflag = firinterp (DATA *x, DATA *h, DATA *r, DATA *dbuffer , ushort nh,
ushort nx, ushort I)
(defined in interp.asm)
Arguments
h[nh] Pointer to coefficient vector of size nh in normal
order:
H = b0 b1 b2 b3 …
r[nx*I] Pointer to real output vector of nx real elements.
In-place computation (r = x) is allowed
dbuffer[(nh/I)+1] Delay buffer of (nh/I)+1 elements
- In the case of multiple-buffering schemes, this
array should be initialized to 0 for the first block
only. Between consecutive blocks, the delay buff-
er preserves delayed input samples in dbuf-
fer[1…(nh/I)+1]. It also preserves a ptr to the next
new entry into the dbuffer. This ptr is preserved
across function calls in dbuffer[0].
- The delay buffer is only nh/I elements and holds
only delayed x inputs. No zero-samples are in-
serted into dbuffer (since only non-zero products
contribute to the filter output)
nx Number of real elements in vector x and r

nh Number of coefficients, with (nh/I) w 3
I Interpolation factor. I is effectively the number of
output samples for every input sample. This routine
can be used with I=1.
- If oflag = 1, a 32-bit data overflow occurred in an
4-54
firinterp
Description Computes an interpolating real FIR filter (direct-form) using coefficient stored
in vector h. The real data input is stored in vector x. The filter output result is
sample filtering (nx = 1).
ȍ h[k]xƪI *t kƫ
nh
Algorithm r[t] + 0 v j v nr
k+0
Special Requirements nh has to be a multiple of I, such nh/I ≥ 3.
Example See examples/decimate subdirectory
Cycles Core:
If I > 1
nx*(2+I*(1+(nh/I)))
If I=1 :
nx*(2+nh)
Overhead 72
Code size 164

(in bytes)

firlat
firlat Lattice Forward (FIR) Filter
Function ushort oflag = firlat (DATA *x, DATA *h, DATA *r, DATA *pbuffer, int nx, int nh)
Arguments
x [nx] Pointer to real input vector of nx real elements in normal
order:
x[0]
x[1]
.
.
x[nx−2]
x[nx−1]
h[nh] Pointer to lattice coefficient vector of size nh in normal
order:
h[0]
h[1]
.
.
h[nh−2]
h[nh−1]
r[0]
r[1]
.
.
r[nx−2]
r[nx−1]
4-56
firlat
pbuffer [nh] Delay buffer

should be initialized to 0 for the first block only. Between
consecutive blocks, the delay buffer preserves the pre-
vious r output elements needed.
- pbuffer: procession buffer of nh length in order:
e′0[n−1]
e′1[n−1]
.
.
e′nh−2[n−1]
e′nh−1[n−1]
nx Number of real elements in vector x (input samples)
intermediate or final result
Description Computes a real lattice FIR filter implementation using coefficient stored in
vector h. The real data input is stored in vector x. The filter output result is
sample filtering (nx=1)
Algorithm e 0[n] + eȀ 0[n] + x[n],

e i[n] + e i*1[n] * hieȀ i*1[n * 1], i + 1, 2, AAA , N
eȀ i[n] + h ie i*1[n] ) eȀ i*1[n * 1], i + 1, 2, AAA , N
y[n] + e N[n]
Example See examples/firlat subdirectory

firs
Cycles† Core: nx{4 + 4(nh−1)]
Overhead: 23
Code size 53
(in bytes)
firs Symmetric FIR Filter
Function ushort oflag = firs (DATA *x, DATA *h, DATA *r, DATA *dbuffer, ushort nx,
ushort nh2)
Arguments
h[nh2] - Pointer to coefficient vector containing the first half
of the symmetric filter coefficients. For example, if
the filter coefficients are {h0, h1, h2, h2, h1, h0},
then h[nh2] = {h0, h1, h2} where h0 resides at the
lowest memory address in the array.
- This array must be located in internal memory
since it is accessed by the C55x coefficient bus.
dbuffer[2*nh2 + 2] Pointer to delay buffer of length nh = 2*nh2 + 2
array should be initialized to 0 for the first filter
block only. Between consecutive blocks, the delay
buffer preserves the previous r output elements
needed.
- The first element in this array is special in that it
contains the array index of the oldest input entry in
the delay buffer. This is needed for multiple-buffer-
ing schemes, and should be initialized to 0 (like all
the other array entries) for the first block only.
4-58
firs
nh2 Half the number of coefficients of the filter (due to

symmetry there is no need to provide the other half).
For example, if the filter coefficients are {h0, h1, h2,
h2, h1, h0}, then nh2 = 3. Must be a minimum value
of 3. For smaller filters, zero pad the coefficients to
meet the minimum value.

Description Computes a real FIR filter (direct-form) with nh2 symmetric coefficients using
the FIRS instruction approach. The filter is assumed to have a symmetric im-
pulse response, with the first half of the filter coefficients stored in the array h.
The real input data is stored in vector x. The filter output result is stored in vec-
tor r. This function maintains the array dbuffer containing the previous delayed
input values to allow consecutive processing of input data blocks. This function
can be used for both block-by-block (nx ≥ 2) and sample-by-sample filtering
(nx = 1). In-place computation (r = x) is allowed.
ȍ h, . .. , [k] * ( x[j * k] ) x [j ) k * 2 * nh2 ) 1] )

nh2*1
k+0
- nh must be a minimum value of 3. For smaller filters, zero pad the h[] array.
- Coefficient array h[nh2] must be located in internal memory since it is ac-

cessed using the C55x coefficient bus, and that bus does not have access
to external memory.

firs
implement this filter. The value x(j−2*nh2) does not enter into the calculations
for for the output r(j). However, this value is required in other DSPLIB filter func-
DSPLIB.

x(j−2*nh2−2)
x(j−2*nh2−1)
oldest x( ) entry x(j−2*nh2)
x(j−1)
x(j−2)
•
•
•
x(j−2*nh2−5)
x(j−2*nh2−4)
x(j−2*nh2−3) highest memory address

x(1)
•
•
•
x(nx−2)
newest x( ) entry x(nx−1) highest memory address
4-60
fltoq15

r(1)
•
•
•
r(nx−2)
Example See examples/firs subdirectory
Cycles† Core: nx[5 + (nh−2)]
Overhead: 72
Code size 133
(in bytes)
fltoq15 Floating-point to Q15 Conversion
Function ushort errorcode = fltoq15 (float *x, DATA *r, ushort nx)
(defined in fltoq15.asm)
Arguments
x[nx] Pointer to floating-point input vector of size nx. x should
contain the numbers normalized between (−1,1). The
errorcode returned value will reflect if that condition is not
met.
r[nx] Pointer to output data vector of size nx containing the q15

equivalent of vector x.
errorcode The function returns the following error codes:
- 1 − if any element is too large to represent in Q15 format
- 2 − if any element is too small to represent in Q15 format
- 3 − both conditions 1 and 2 were encountered

hilb16
Description Convert the IEEE floating-point numbers stored in vector x into Q15 numbers
stored in vector r. The function returns the error codes if any element x[i] is not
representable in Q15 format.
All values that exceed the size limit will be saturated to a Q15 1 or −1 depend-
ing on sign (0x7fff if value is positive, 0x8000 if value is negative). All values
too small to be correctly represented will be truncated to 0.
Overflow Handling Methodology Saturation implemented for overflow handling
Example See examples/expn subdirectory
Cycles† Core: 17 * nx (if x[n] ==0)
23 * nx (if x[n] is too small for Q15
representation)
32 * nx (if x[n] is too large for Q15
representation)
38 * nx (otherwise)
Overhead: 23
Code size 157
(in bytes)
hilb16 FIR Hilbert Transformer
Function ushort oflag = hilb16 (DATA *x, DATA *h, DATA *r, DATA *dbuffer, ushort nx,
ushort nh)
Arguments
h[nh] - Pointer to coefficient vector of size nh in normal
order. H= {h0, h1, h2, h3, h4, …} Every odd valued
filter coefficient has to 0, i.e. h1 = h3 = … = 0. And
H = {h0, 0, h2, 0, h4, 0, …} where h0 resides at the
lowest memory address in the array.
4-62
hilb16
r[nx] Pointer to output vector of nx real elements.

In-place computation (r = x) is allowed.
dbuffer[nh + 2] Pointer to delay buffer of length nh = nh + 2
array should be initialized to 0 for the first filter
block only. Between consecutive blocks, the delay
buffer preserves the previous r output elements
needed.
- The first element in this array is special in that it
contains the array index-1 of the oldest input entry
in the delay buffer. This is needed for multiple-
buffering schemes, and should be initialized to
zero (like all the other array entries) for the first
block only.

nh The number of coefficients of the filter. For example
if the filter coefficients are {h0, h1, h2, h3, h4, h5},
then nh = 6. Must be a minimum value of 6. For
smaller filters, zero pad the coefficients to meet the
minimum value.
function can be used for both block-by-block (nx >= 2) and sample-by-sample
ȍ h[k] x [j * k]
nh*1
Algorithm r[j] + 0 v j v nx
k+0

hilb16
- Every odd valued filter coefficient has to be 0. This is a requirement for the
Hilbert transformer. For example, a 6 tap filter may look like this: H = [0.867
0 –0.324 0 –0.002 0]
- Always pad 0 to make nh as a even number. For example, a 5 tap filter with
a zero pad may look like this: H = [0.867 0 –0.324 0 –0.002 0]
- nh must be a minimum value of 6. For smaller filters, zero pad the H[] array.
implement this filter. The value x(j−nh) does not enter into the calculations for
for the output r(j). However, this value is required in other DSPLIB filter func-
DSPLIB.

x(j−nh−2)
x(j−nh−1)
x(j−1)
x(j−2)
•
•
•
x(j−nh−5)
x(j−nh−4)
4-64
hilb16

x(1)
•
•
•
x(nx−2)

r(1)
•
•
•
r(nx−2)
Example See examples/hilb16 subdirectory
Cycles Core: nx*(2+nh/2)
Overhead: 28
Code size 108
(in bytes)

iir32
iir32 Double-precision IIR Filter
Function ushort oflag = iir32 (DATA *x, LDATA *h, DATA *r, LDATA *dbuffer, ushort nbiq,
ushort nr)
(defined in iir32.asm)
Arguments
x [nr] Pointer to input data vector of size nr
h[5*nbiq] Pointer to the 32-bit filter coefficient vector with the
following format. For example for nbiq= 2, h is equal
to:
b21 – high beginning of biquad 1
b21 – low
b11 – high
b11 – low
b01 – high
b01 – low
a21 – high
a21 – low
a11 – high
a11 – low
b22 – high beginning of biquad 2 coefs
b22 – low
b12 – high
b12 – low
b02 – high
b02 – low
a22 – high
a22 – low
a12 – high
a12 – low
r[nr] Pointer to output data vector of size nr. r can be
equal or less than x.
4-66
iir32
dbuffer[2*nbiq+2] Pointer to address of 32-bit delay line dbuffer. Each

biquad has 3 consecutive delay line elements. For
example for nbiq = 2:
d1(n−2) − low beginning of biquad 1
d1(n−2) − high
d1(n−1) – low
d1(n−1) – high
d2(n−2) − low beginning of biquad 2
d2(n−2) − high
d2(n−1) – low
d2(n−1) – high
only. Between consecutive blocks, the delay buffer
preserves the previous elements needed.
- Memory alignment: none required for C5510. This

is a group of circular buffers. Each biquad’s delay
buffer is treated separately. The Buffer Start Ad-
dress (BSAxx) updated to a new location for each
biquad.
nbiq Number of biquads

nr Number of elements of input and output vectors
Description Computes a cascaded IIR filter of nbiquad biquad sections using 32-bit coeffi-
cients and 32-bit delay buffers. The input data is assumed to be single-preci-
sion (16 bits).
Each biquad section is implemented using Direct-form II. All biquad coeffi-
cients (5 per biquad) are stored in vector h. The real data input is stored in vec-
tor x. The filter output result is stored in vector r .
This function retains the address of the delay filter memory d containing the
previous delayed values to allow consecutive processing of blocks. This func-
tion is more efficient for block-by-block filter implementation due to the C-call-
ing overhead. However, it can be used for sample-by-sample filtering (nx = 1).

iircas4
Algorithm (for biquad)

d(n) + x(n) * a1 * d(n * 1) * a2 * d(n * 2)
y(n) + b0 * d(n) ) b1 * d(n * 1) ) b2 * d(n * 2)
Implementation Notes See program iircas32.asm
Example See examples/iir32 subdirectory
Cycles Core: nx*(7+ 31*nbiq)
Overhead: 77
Code size 203
(in bytes)
iircas4 Cascaded IIR Direct Form II Using 4 Coefficients per Biquad
Function ushort oflag = iircas4 (DATA *x, DATA *h, DATA *r, DATA *dbuffer, ushort nbiq,
ushort nx)
(defined in iir4cas4.asm)
Arguments
x [nx] Pointer to input data vector of size nx
h[4*nbiq] Pointer to filter coefficient vector with the following
format:
h = a11 b11 a21 b21 ....a1i b1i a2i b2i
where i is the biquad index (a21 is the a2 coefficient
of biquad 1). Pole (recursive) coefficients = a. Zero
(non-recursive) coefficients = b
r[nx] Pointer to output data vector of size nx. r can be

equal than x.
4-68
iircas4
dbuffer[2*nbiq+1] First location of dbuffer is reserved for the index.

Each biquad has 2 delay line elements separated by
nbiq locations in the following format:
d1(n−1), d2(n−1),..di(n−1) d1(n−2), d2(n−2)...di(n−2)
where i is the biquad index (d2(n−1) is the (n−1)th
delay element for biquad 2).
preserves the previous r output elements needed.
nx Number of elements of input and output vectors
Description Computes a cascade IIR filter of nbiq biquad sections. Each biquad section is
implemented using Direct-form II. All biquad coefficients (4 per biquad) are
stored in vector h. The real data input is stored in vector x. The filter output
result is stored in vector r.

d(n) + x(n) * a1 * d(n * 1) * a2 * d(n * 2)
y(n) + d(n) ) b1 * d(n * 1) ) b2 * d(n * 2)
Special Requirements Number of biquads, nbiq, must be even.
Example See examples/iircas4 subdirectory

iircas5
Cycles† Core: nx * (2 + 3 * nbiq)
Overhead: 44
Code size 122
(in bytes)
iircas5 Cascaded IIR Direct Form II (5 Coefficients per Biquad)
ushort nx)
(defined in iircas5.asm)
Arguments
format:
h = a11 b11 a21 b21 b01 ... a1i b1i a2i b2i b0i
where i is the biquad index a21 is the a2 coefficient

equal than x.
dbuffer[2*nbiq+1] Pointer to address of delay line d. Each biquad has 2
delay line elements separated by nbiq locations in
the following format:
d1(n−1), d2(n−1),..di(n−1) d1(n−2), d2(n−2)...di(n−2)
where i is the biquad index(d2(n−1) is the (n−1)th
preserves the previous elements needed.
4-70
iircas5

implemented using Direct-form II. All biquad coefficients (5 per biquad) are
The usage of 5 coefficients instead of 4 facilitates the design of filters with a

unit gain of less than 1 (for overflow avoidance), typically achieved by filter
coefficient scaling.

d(n) + x(n) * a1 * d(n * 1) * a2 * d(n * 2)
y(n) + b0 * d(n) ) b1 * d(n * 1) ) b2 * d(n * 2)
Overhead: 60
Code size 126
(in bytes)

iircas51
iircas51 Cascaded IIR Direct Form I (5 Coefficients per Biquad)
ushort nx)
(defined in iircas51.asm)
Arguments
format:
h = b01 b11 b21 a11 a21 ....b0i b1i b2i a1i a2I
where i is the biquad index (a21 is the a2 coefficient

equal to x.
dbuffer[4*nbiq+1] Pointer to address of delay line dbuffer. Each biquad
has 4 delay line elements stored consecutively in
memory in the following format:
x1(n−1) ... xi(n−1), x1(n−2) ... xi(n−2)
y1(n−1) ... yi(n−1), yi(n−2) ... yi(n−2)
where i is the biquad index(x1(n−1) is the (n−1)th
preserves the previous r output elements needed.
- If oflag = 1, a 32-bit overflow has occurred.
implemented using Direct-form I. All biquad coefficients (5 per biquad) are
4-72
iircas51
The usage of 5 coefficients instead of 4 facilitates the design of filters with a

unit gain of less than 1 (for overflow avoidance), typically achieved by filter
coefficient scaling.

y(n) + b0 * x(n) ) b1 * x(n * 1) ) b2 * x(n * 2) * a1 * y(n * 1) * a2 * y(n * 2)
Overhead: 68
Code size 154
(in bytes)

iirlat
iirlat Lattice Inverse (IIR) Filter
Function ushort oflag = iirlat (DATA *x, DATA *h, DATA *r, DATA *dbuffer, int nx, int nh)
Arguments
x [nx] Pointer to real input vector of nx real elements in normal
order:
x[0]
x[1]
.
.
x[nx−2]
x[nx−1]
h[nh] Pointer to lattice coefficient vector of size nh in normal
order with the first element zero-padded:
0
h[0]
h[1]
.
.
h[nh−2]
h[nh−1]

r[0]
r[1]
.
.
r[nx−2]
r[nx−1]
dbuffer[nh] Delay buffer
In the case of multiple-buffering schemes, this array should
be initialized to 0 for the first block only. Between
consecutive blocks, the delay buffer preserves the previous
r output elements needed.
4-74
iirlat
Description Computes a real lattice IIR filter implementation using coefficient stored in vec-
tor h. The real data input is stored in vector x. The filter output result is stored
in vector r . This function retains the address of the delay filter memory d con-
taining the previous delayed values to allow consecutive processing of blocks.
This function can be used for both block-by-block and sample-by-sample filter-
ing (nx = 1)
Algorithm e N[n] + x[n],

e i*1[n] + e i[n] * hieȀ i*1[n * 1], i + N, (N*1), AAA , 1
eȀ i[n] + −k ie i*1 ) eȀ i*1[n * 1], i + N, (N*1), AAA , 1
y[n] + e 0[n] + eȀ 0[n]
Example See examples/iirlat subdirectory
Cycles† Core: 4 * (nh − 1) * nx
Overhead: 24
Code size 54
(in bytes)

ldiv16
ldiv16 32-bit by 16-bit Long Division Function
Function void ldiv16 (LDATA *x, DATA *y, DATA *r, DATA *rexp, ushort nx)
Arguments
x [nx] Pointer to input data vector 1 of size nx
x[0]
x[1]
.
.
x[nx−2]
x[nx−1]
r[nx] Pointer to output data buffer

r[0]
r[1]
.
.
r[nx−2]
r[nx−1]
rexp[nx] Pointer to exponent buffer for output values. These exponent

values are in integer format.
rexp[0]
rexp[1]
.
.
rexp[nx−2]
rexp[nx−1]
Description This routine implements a long division function of a Q31 value divided by a
Q15 value. The reciprocal of the Q15 value, y, is calculated then multiplied by
the Q31 value, x. The result is returned as an exponent such that:
r[i] * rexp[i] = true reciprocal in floating-point
Algorithm The reciprocal of the Q15 number is calculated using the following equation:
Ym + 2 * Ym * Ym 2 * Xnorm
If we start with an initial estimate of Ym, the equation converges to a solution

very rapidly (typically 3 iterations for 16-bit resolution).
4-76
log_10
The initial estimate can be obtained from a look-up table, from choosing a mid-
point, or simply from linear interpolation. The method chosen for this problem
is linear interpolation and is accomplished by taking the complement of the
least significant bits of the Xnorm value.
The reciprocal is multiplied by the Q31 number to generate the output.
Overflow Handling Methodology none
Example See examples/ldiv16 subdirectory
Overhead: 14
Code size 91
(in bytes)
log_10 Base 10 Logarithm
Function ushort oflag = log_10 (DATA *x, LDATA *r, ushort nx)
(defined in log_10.asm)
Arguments
r[nx] Pointer to output data vector (Q31 format) of size nx.
Description Computes the log base 10 of elements of vector x using Taylor series.
Algorithm for (i + 0; i t nx; i ) )) y(i) + log 10x(i) where −1 t x(i) t 1

log_10
Implementation Notes y = 0.4343 * ln(x) with x = M(x)*2^P(x) = M*2^P

y = 0.4343 * (ln(M) + ln(2)*P)
y = 0.4343 * (ln(2*M) + (P−1)*ln(2))
y = 0.4343 * (ln((2*M−1)+1) + (P−1)*ln(2))
y = 0.4343 * (f(2*M−1) + (P−1)*ln(2))
with f(u) = ln(1+u).
We use a polynomial approximation for f(u) :

f(u) = (((((C6*u+C5)*u+C4)*u+C3)*u+C2)*u+C1)*u+C0
for 0<= u <= 1.
The polynomial coefficients Ci are as follows :

C0 = 0.000 001 472
C1 = 0.999 847 766
C2 = −0.497 373 368
C3 = 0.315 747 760
C4 = −0.190 354 944
C5 = 0.082 691 584
C6 = −0.017 414 144
The coefficients Bi used in the calculation are derived from the Ci as follows:
B0 Q30 1581d 0062Dh

B1 Q14 16381d 03FFDh
B2 Q15 −16298d 0C056h
B3 Q16 20693d 050D5h
B4 Q17 −24950d 09E8Ah
B5 Q18 21677d 054ADh
B6 Q19 −9130d 0DC56h
Example See examples/log_10 subdirectory
Overhead: 36
Code size 162
(in bytes)
4-78
log_2
log_2 Base 2 Logarithm
Function ushort oflag = log_2 (DATA *x, LDATA *r, ushort nx)
(defined in log_2.asm)
Arguments
Description Computes the log base 2 of elements of vector x using Taylor series.
Algorithm for (i + 0; i t nx; i ) )) y(i) + log 12x(i) where 0 t x(i) t 1

y = 1.4427 * (ln(M) + ln(2)*P)
y = 1.4427 * (ln(2*M) + (P−1)*ln(2))
y = 1.4427 * (ln((2*M−1)+1) + (P−1)*ln(2))
y = 1.4427 * (f(2*M−1) + (P−1)*ln(2))
We use a polynomial approximation for f(u) :

for 0<= u <= 1.
The polynomial coefficients Ci are as follows:

C0 = 0.000 001 472
C1 = 0.999 847 766
C2 = −0.497 373 368
C3 = 0.315 747 760
C4 = −0.190 354 944
C5 = 0.082 691 584
C6 = −0.017 414 144

logn
B0 Q30 1581d 0062Dh

B1 Q14 16381d 03FFDh
B2 Q15 −16298d 0C056h
B3 Q16 20693d 050D5h
B4 Q17 −24950d 09E8Ah
B5 Q18 21677d 054ADh
B6 Q19 −9130d 0DC56h
Example See examples/log_2 subdirectory
Overhead: 37
Code size 166
(in bytes)
logn Base e Logarithm (natural logarithm)
Function ushort oflag = logn (DATA *x, LDATA *r, ushort nx)
(defined in logn.asm)
Arguments
Description Computes the log base e of elements of vector x using Taylor series.
Algorithm for (i + 0; i t nx; i ) )) y(i) + log nx(i) where −1 t x(i) t 1
4-80
maxidx

y = 0.4343 * (ln(M) + ln(2)*P)
y = 0.4343 * (ln(2*M) + (P−1)*ln(2))
y = 0.4343 * (ln((2*M−1)+1) + (P−1)*ln(2))
y = 0.4343 * (f(2*M−1) + (P−1)*ln(2))
We use a polynomial approximation for f(u):
for 0<= u <= 1.
The polynomial coefficients Ci are as follows:
C0 = 0.000 001 472
C1 = 0.999 847 766
C2 = −0.497 373 368
C3 = 0.315 747 760
C4 = −0.190 354 944
C5 = 0.082 691 584
C6 = −0.017 414 144
B0 Q30 1581d 0062Dh
B1 Q14 16381d 03FFDh
B2 Q15 −16298d 0C056h
B3 Q16 20693d 050D5h
B4 Q17 −24950d 09E8Ah
B5 Q18 21677d 054ADh
B6 Q19 −9130d 0DC56h
Example See examples/logn subdirectory
Overhead: 36
Code size 132
(in bytes)
maxidx Index of the Maximum Element of a Vector

Function short r = maxidx (DATA *x, ushort ng, ushort ng_size);
(defined in maxidx.asm)
Arguments

maxidx

r Index for vector element with maximum value.
ng Number of groups.
ng_size Size of group.
Description The vector x is divided in ng groups of size ng_size.
Size of x = ng x ng_size. ng_size must be an even number between 2 and 34.

The larger ng_size, the better the performance.Returns the index of the maxi-
mum element of a vector x. The index is a number between 0 and nx − 1. In
case of multiple maximum elements, r contains the index of the first maximum
element found.
Example 1: size of x is 64.

Choose ng_size = 32, ng = 2


- ng_size is an even number between 2 and 34.
- nx is an even number.
- Input vector has to be 32-bit aligned.
- Algorithm uses two locations in .bss section.
Example See examples/maxidx subdirectory
4-82
maxidx34
Cycles† Core: nx/2 + ng16
Overhead: 40
Code size 143
(in bytes)
maxidx34 Index of the Maximum Element of a Vector ≤ 34
Function short r = maxidx34 (DATA *x, ushort nx)

(defined in maxidx34.asm)
Arguments
r Index for vector element with maximum value.
nx Lenght of input data vector (nx ≤ 34).
Description Returns the index of the maximum element of a vector x. The index is a number
between 0 and nx − 1. In case of multiple maximum elements, r contains the
index of the first maximum element found.
Special Requirements Size of the vector, nx ≤ 34
nx is an even number.
Input vector has to be 32-bit aligned.
Example See examples/maxidx34 subdirectory
Cycles† Core: nx/2
Overhead: 42
Code size 26
(in bytes)

maxval
maxval Maximum Value of a Vector
Function short r = maxval (DATA *x, ushort nx)

(defined in maxval.asm)
Arguments
r Maximum value of a vector
Description Returns the maximum element of a vector x.

Special Requirements nx is an even number.
Input vector has to be 32-bit aligned.
Example See examples/maxval subdirectory
Cycles† Core: nx
Overhead: 3
Code size 20
(in bytes)
maxvec Index and Value of the Maximum Element of a Vector
Function void maxvec (DATA *x, ushort nx, DATA *r_val, DATA *r_idx)
(defined in maxvec.asm)
Arguments
r_val maximum value
r_idx Index for vector element with maximum value
nx Lenght of input data vector (nx w 6)
4-84
minidx
Description This function finds the index for vector element with maximum value. In case
of multiple maximum elements, r_idx contains the index of the first maximum
element found. r_val contains the maximum value.
Example See examples/maxvec subdirectory
Cycles Core: nx*3
Overhead: 8
Code size 26
(in bytes)
minidx Index of the Minimum Element of a Vector
Function short r = minidx (DATA *x, ushort nx)

(defined in minidx.asm)
Arguments
r Index for vector element with minimum value
Description Returns the index of the minimum element of a vector x. In case of multiple
minimum elements, r contains the index of the first minimum element found.
- Input vector must be 321−bit aligned.
- Algorithm uses two locations in .bss section.

minval

Example See examples/minidx subdirectory
Cycles† Core: nx * 3
Overhead: 7
Code size 26
(in bytes)
minval Minimum Value of a Vector
Function short r = minval (DATA *x, ushort nx)

(defined in minval.asm)
Arguments
r Minimum value of a vector
Description Returns the minimum element of a vector x.

- nx is an even number.
- Input vector has to be 32−bit aligned.

Example See examples/minval subdirectory
Cycles† Core: nx/2
Overhead: 7
Code size 20
(in bytes)
4-86
minvec
minvec Index and Value of the Minimum Element of a Vector
Function void minvec (DATA *x, ushort nx, DATA *r_val, DATA *r_idx)
(defined in minvec.asm)
Arguments
r_val Minimum value
r_idx Index for vector element with minimum value
nx Length of input data vector (nx w 6)
Description This function finds the index for vector element with minimum value. In case
of multiple minimum elements, r_idx contains the index of the first minimum
element found. r_val contains the minimum value.
Example See examples/minvec subdirectory
Cycles Core: nx*3
Overhead: 8
Code size 26
(bytes)

mmul
mmul Matrix Multiplication
Function ushort oflag = mmul (DATA *x1,short row1,short col1,DATA *x2,short

row2,short col2,DATA *r)
(defined in mmul.asm)
Arguments
x1[row1*col1]: Pointer to input vector of size nx
Pointer to input matrix of size row1*col1
; row1 :
; :
; :
; r[row1*col2] : Pointer to output data vector of size
row1*col2
row1 number of rows in matrix 1
col1 number of columns in matrix 1
x2[row2*col2]: Pointer to input matrix of size row2*col2
row2 number of rows in matrix 2
col2 number of columns in matrix 2
r[row1*col2] Pointer to output matrix of size row1*col2
Description This function multiplies two matrices
Algorithm Multiply input matrix A (M by N) by input matrix B (N by P) using 2 nested loops:

for i = 1 to M
for k = 1 to P
{
temp = 0
for j = 1 to N
temp = temp + A(i,j) * B(j,k)
C(i,k) = temp
}
- Verify that the dimensions of input matrices are legal, i.e. col1 == row2
- x2[ ] matrix must be located in the internal memory.
4-88
mtrans
Implementation Notes In order to take advantage of the dual MAC architecture of the C55x, this imple-
mentation checks the size of the matrix x1. For small matrices x1 (row1 < 4 or
col1 < 2), single MAC loops are used. For larger matrices x1 (row1 ≥ 4 and
col1 ≥ 2), Dual MAC loops are more efficient and quickly make up for the addi-
tional initialization overhead.
Example See examples/mmul subdirectory
Cycles† Core:
- if(row1 < 4 || col1 < 2), use single MAC
((col1 + 2)*row1 + 4)*col2
- if((row1==even)&&(row1 ≥ 4)&&(col1 ≥ 2)), use dual MAC
((col1 + 4)*0.5*row1 + 10)col2
- if((row1==odd)&&(row1 ≥ 4)&&(col1 ≥ 2), use dual MAC
((col1 + 4)*0.5*(row1 − 1) + col1 + 12)col2
Overhead: 30
Code size 215
(in bytes)
mtrans Matrix Transpose
Function ushort oflag = mtrans (DATA *x, short row, short col, DATA *r)
(defined in mtrans.asm)
Arguments
x[row*col] Pointer to input matrix. In-place processing is not allowed.
row number of rows in matrix
col number of columns in matrix
r[row*col] Pointer to output data vector
Description This function transposes matrix x
Algorithm for i = 1 to M
for j = 1 to N
C(j,i) = A(i,j)

mul32
Example See examples/mtrans subdirectory
Cycles† Core: (1 + col) * row
Overhead: 23
Code size 65
(in bytes)
mul32 32-bit Vector Multiplication
Function ushort oflag = mul32 (LDATA *x, LDATA *y, LDATA *r, ushort nx)
(defined in mul32.asm)
Arguments
x[nx] Pointer to input data vector of size nx. In-place processing
y[nx] Pointer to input data vector of size nx

r[nx] Pointer to output data vector of size nx
nx Number of elements of input and output vectors. Nx w 4
oflag Overflow flag
Description This function multiplies two 32-bit Q31 vectors, element by element, and
produces a 32-bit Q31 vector.
Algorithm for (i + 0; i t nx; i ) ))

z(i) + x(i) * y(i)
- Input and Output vectors must be 32−bit aligned.
4-90
neg

Example See examples/add subdirectory
Benchmarks
Cycles Core: 4*nx + 4
Overhead 21
Code size 73
(in bytes)
neg Vector Negate
Function ushort oflag = neg (DATA *x, DATA *r, ushort nx)
(defined in neg.asm)
Arguments
r[nx] Pointer to output data vector of size nx. In-place processing
allowed
Special cases:
- if x[I] = −1 = 32768 , then r = 1 = 321767 with oflag = 1
- if x= 1 = 32767 , then r = −1 = 321768 with oflag = 1

nx ≥ 4
Caution: overflow in negation of a Q15 number can happen
naturally when negating (−1).
Description This function negates each of the elements of a vector (fractional values).
Algorithm for (i + 0; i t nx; i ) )) x(i) + −x(i)

Example See examples/neg subdirectory

neg32
Overhead: 13
Code size 61
(in bytes)
neg32 Vector Negate (double-precision)
Function ushort oflag = neg32 (LDATA *x, LDATA *r, ushort nx)
(defined in neg.asm)
Arguments
r[nx] Pointer to output data vector of size nx. In-place processing

allowed
Special cases:
- if x = −1 = 32768 * 216, then r = 1 = 321767 * 216 with oflag
=1
- if x= 1 = 32767 * 216, then r = −1 = 321768 * 216 with oflag
=1

nx ≥ 4
Caution: overflow in negation of a Q31 number can happen
naturally when negating (−1).
Description This function negates each of the elements of a vector (fractional values).
Algorithm for (i + 0; i t nx; i ) )) x(i) + −x(i)
4-92
power
- Input and Output vectors must be 32−bit aligned.
Example See examples/neg32 subdirectory
Overhead: 13
Code size 61
(in bytes)
power Vector Power
Function ushort oflag = power (DATA *x, LDATA *r, ushort nx)
(defined in power.asm)
Arguments
r[1] Pointer to output data vector element in Q31 format
Special cases:
- if x= −1 = 32768*216 , then r = 1 = 321767*216
with oflag = 1
- if x= 1 = 32767*216 , then r = −1 = 321768*216
with oflag = 1
nx Number of elements of input vectors.
nx ≥ 4
Description This function calculates the power (sum of products) of a vector.
Algorithm Power = 0 for (i + 0; i t nx; i ) )) power )+ x(i) * x(I)
Overflow Handling Methodology No scaling implemented for overflow handling

q15tofl
- Output vector must be 32−bit aligned.
Example See examples/power subdirectory
Cycles† Core: nx − 1
Overhead: 12
Code size 54
(in bytes)
q15tofl Q15 to Floating-point Conversion
Function ushort q15tofl (DATA *x, float *r, ushort nx)

(defined in q152fl.asm)
Arguments
x[nx] Pointer to Q15 input vector of size nx.
r[nx] Pointer to floating-point output data vector of size nx
containing the floating-point equivalent of vector x.
Description Converts the Q15 stored in vector x to IEEE floating-point numbers stored in
vector r.
- Output vector must be 32−bit aligned.
Example See examples/ug subdirectory
4-94
rand16
Cycles† Core: 7 * nx (if x[n] ==0)

32 * nx
Overhead: 18
Code size 124
(in bytes)
rand16 Random Number Generation Algorithm
Function ushort oflag= rand16 (DATA *r, ushort nr)
Arguments
*r Pointer to the array where the 16-bit random numbers are
stored
nr Number of random numbers that are generated
Description This algorithm computes an array of random numbers based on the linear con-
gruential method introduced by D. Lehmer in 1951. This is one of the fastest
and simplest techniques of generating random numbers. The code shown
here generates 16-bit integers, however, if a 32-bit value is desired the code
can be modified to perform 32-bit multiplies using the defined constants
RNDMULT and RNDINC. The disadvantage of this technique is that it is very
sensitive to the choice of RANDMULT and RNDINC.
Algorithm r[n] + [( r[n * 1] < RNDMULT ) ) RNDINC] % M

where 0 v n v nr and 0 v M v 65536
Special Requirements No special requirements.

rand16
Implementation Notes Rand16() is written so that it can be called from a C program. Prior to calling
rand16(), rand16i() can be called to initialize the random number generator
seed value. The C routine passes two parameters to rand16(): A pointer to the
random number array *r and the count of random numbers (nr) desired. The
random numbers are declared as short or 16 bit values. Two constants
RNDMULT and RNDINC are defined in the function. The algorithm is sensitive
to the choice of RNDMULT and RNDINC so exercise caution when changing
these.
M This value is based on the system that the routine runs. This
routine returns a random number from 0 to 65536 (64K) and
is NOT internally bounded. If you need a min/max limit, this
must be coded externally to this routine.
RNDSEED An arbitrary constant that can be any value between 0 and

64K. If 0 (zero) is chosen, then RNDINC should be some
value greater than 1. Otherwise, the first two values will be 0
and 1. To change the set of random numbers generated by
this routine, change the RNDSEED value. In this routine,
RNDSEED is initialized to 21845, which is 65536/3.
RNDMULT Should be chosen such that the last three digits fall in the
pattern even_digit−2−1 such as xx821, xx421 etc.
RNDMULT = 31821 is used in this routine.
RNDINC In general, this constant can be any prime number related to
M. Research shows that RNDINC (the increment value)
should be chosen by the following formula:
RNDINC = ((1/2 − (1/6 * SQRT(3))) * M). Using M=65536,
RNDINC was picked as 13849.
The random seed initialized in rand16i() is used to generate the first random
number. Each random number generated is used to generate the next number
in the series. The random number is generated in the accumulator (32 bits) by
using the multiply-accumulate (MAC) unit to do the computation. In the course
of the algorithm if there is intermediate overflow, the overflow flag bit in status
register is set. At the end of the algorithm, the overflow flag is tested for any
intermediate overflow conditions.
Example See examples/rand16 subdirectory
4-96
rand16init
Benchmarks
Cycles Core: 13 + nr*2
Overhead: 10
Code size 49
(in bytes)
C54x Benchmark for Comparison

Cycles Core: 10 + nr*4
Overhead: 16
Code size 56
(in bytes)
rand16init Random Number Generation Initialization
Function void rand16init(void)
Arguments none
Description Initializes seed for 16-bit random number generator.
Special Requirements Allocation of .bss section is required in linker command file.
Implementation Notes This function initializes a global variable rndseed in global memory to be used
for the 16 bit random number generation routine (rand16)
Example See examples/rand16i subdirectory
Benchmarks
Cycles 6
Code size 9
(in bytes)
C54x Benchmark for Comparison

Cycles 7
Code size 10
(in bytes)

recip16
recip16 16-bit Reciprocal Function
Function void recip16 (DATA *x, DATA *r, DATA *rexp, ushort nx)
Arguments
x[nx] Pointer to input data vector 1 of size nx.
x[0]
x[1]
.
.
x[nx−2]
x[nx−1]
r[nx] Pointer to output data buffer

r[0]
r[1]
.
.
r[nx−2]
r[nx−1]
rexp[nx] Pointer to exponent buffer for output values. These exponent

values are in integer format.
rexp[0]
rexp[1]
.
.
rexp[nx−2]
rexp[nx−1]
Description This routine returns the fractional and exponential portion of the reciprocal of
a Q15 number. Since the reciprocal is always greater than 1, it returns an expo-
nent such that:
r[i] < r exp[i] + true reciprocal in floating-point
Algorithm Ym + 2 * Ym * Ym 2 * Xnorm
If we start with an initial estimate of Ym, the equation converges to a solution

very rapidly (typically 3 iterations for 16-bit resolution).
4-98
rfft
is linear interpolation and is accomplished by taking the complement of the
least significant bits of the Xnorm value.
Overflow Handling Methodology none
Example See examples/recip16 subdirectory
Overhead: 12
Code size 69
(in bytes)
rfft Forward Real FFT (in-place)
Function void rfft (DATA *x, ushort nx, type);

(reference cfft.asm, cbrev.asm, unpack.asm)
Arguments
x [nx] Pointer to input vector containing nx real elements. On output,
vector x contains the first half (nx/2 complex elements) of the
FFT output in the following order. Real FFT is a symmetric
function around the Nyquist point, and for this reason only half
of the FFT(x) elements are required.
On output x will contain the FFT(x) = y in the following format:
y(0)Re y(nx/2)im → DC and Nyquist
y(1)Re y(1)Im
y(2)Re y(2)Im
....
y(nx/2)Re y(nx/2)Im
Complex numbers are stored in Re-Im format
nx Number of real elements in vector x. can take the following
values.
nx = 16, 32, 64, 128, 256, 512, 2048

rfft
type RFFT type selector. Types supported:

Description Computes a Radix-2 real DIT FFT of the nx real elements stored in vector x
in normal order. The original content of vector x is destroyed in the process.
The first nx/2 complex elements of the RFFT(x) are stored in vector x in nor-
mal-order.
Algorithm (DFT)
See CFFT
- Input vector must be aligned on a 32−bit boundary.
- Twiddle table must be located in the internal memory.
J Ensure that the entire data buffer fits within a 64K boundary (the larg-
est possible array addressable by the 16-bit auxiliary register).
J For best performance, the data buffer has to be in a DARAM block.
J If the twiddle table and the data buffer are in the same DARAM block,
then the radix-2 kernel is 7 cycles and the radix-4 kernel is not af-
fected.
Implementation Notes Implemented as a complex FFT of size nx/2 followed by an unpack stage to
unpack the real FFT results. Therefore, Implementation Notes for the cfft func-
tion apply to this case.
Notice that normally an FFT of a real sequence of size N, produces a complex

sequence of size N (or 2*N real numbers) that will not fit in the input sequence.
To accommodate all the results without requiring extra memory locations, the
output reflects only half of the spectrum (complex output). This still provides
the full information because an FFT of a real sequence has even symmetry
around the center or nyquist point(N/2).
When scale = 1, this routine prevents overflow by scaling by 2 at each FFT

intermediate stages and at the unpacking stage.
4-100
rfft
Comparing the results to MATLAB:

The C55 DSPLIB rfft( ) function is implemented as follows using the cfft ( ) and
unpack ( ) functions. (N denotes the size of the real fft)
C55 DSPLIB C55 DSPLIB
Cfft( ), N/2
rfft( ) Unpack( )
J NOSCALE version
C55 DSPLIB MATLAB
rfft( )
Cfft( ), N x N/2
NOSCALE
The unpack routine in the DSPLIB always scales by two the data independent-
ly of the scaled or non-scaled rfft ( ). In order to compare the results to the MAT-
LAB results, the MATLAB results need to be multiplied by a factor of N/2 (N is
the rfft size).
J SCALE version
C55 DSPLIB MATLAB
rfft( )
cfft( ), N
SCALE
The C55 DSPLIB scaled rfft results can be compared to the unmodified MAT-
LAB cfft results.

rfft32
rfft32 Forward 32-bit Real FFT (in-place)
Function void rfft32 (LDATA *x, ushort nx, type);

(reference cfft32.asm, cbrev32.asm, unpack32.asm)
Arguments
x [nx] Pointer to input vector containing nx 32-bit real elements. On
output, vector x contains the first half (nx/2 complex elements)
of the FFT output in the following order. Real FFT is a
symmetric function around the Nyquist point, and for this
reason only half of the FFT(x) elements are required.
On output x will contain the FFT(x) = y in the following format:
y(0)Re y(nx/2)im → DC and Nyquist
y(1)Re y(1)Im
y(2)Re y(2)Im
....
y(nx/2)Re y(nx/2)Im
Complex numbers are stored in Re-Im format
nx Number of real elements in vector x. can take the following
values.
nx = 16, 32, 64, 128, 256, 512,1024,2048
Description Computes a Radix-2 real DIT FFT of the nx real elements stored in vector x
in normal order. The original content of vector x is destroyed in the process.
The first nx/2 complex elements of the RFFT(x) are stored in vector x in nor-
mal-order.
Algorithm (DFT)
See CFFT
4-102
rifft
J For best performance, the coefficient buffer can be in an SARAM

block or a DARAM different from the DARAM clock that contains the
data buffer.
Implementation Notes Implemented as a complex FFT of size nx/2 followed by an unpack stage to
unpack the real FFT results. Therefore, Implementation Notes for the cfft func-
tion apply to this case.
Notice that normally an FFT of a real sequence of size N, produces a complex
the full information because an FFT of a real sequence has even symmetry
When scale = 1, this routine prevents overflow by scaling by 2 at each FFT
Example See examples/rfft32 subdirectory
rifft Inverse Real FFT (in-place)
Function void rifft (DATA *x, ushort nx, type);

(reference cifft.asm, cbrev.asm, and unpacki.asm)
Arguments
x [nx] Pointer to input vector x containing nx real elements. The
unpacki routine should be called to unpack the rfft sequence
before calling the bit reversal routine. (See examples directory
for calling sequence)
On output, the vector x contains nx complex elements
corresponding to RIFFT(x) or the signal itself.
nx Number of real elements in vector x. nx can take the following
values.
nx =16, 32, 64, 128, 256, 512, 1024, 2048
Description Computes a Radix-2 real DIT IFFT of the nx real elements stored in vector x
in bit−reversed order. The original content of vector x is destroyed in the
process. The first nx/2 complex elements of the IFFT(x) are stored in vector
x in normal-order.

rifft
Algorithm (IDFT)
See CIFFT
J If the twiddle table and the data buffer are in the same DARAM block,
then the radix-2 kernel is 7 cycles and the radix-4 kernel is not af-
fected.
Implementation Notes Implemented as a complex IFFT of size nx/2 followed by an unpack stage to
unpack the real IFFT results. Therefore, Implementation Notes for the cfft
function apply to this case.
Notice that normally an IFFT of a real sequence of size N, produces a complex

the full information because an IFFT of a real sequence has even symmetry
When scale = 1, this routine prevents overflow by scaling by 2 at each IFFT

Example See examples/rifft subdirectory
4-104
rifft32
rifft32 Inverse 32-bit Real FFT (in-place)
Function void rifft32 (LDATA *x, ushort nx, type);

(reference cifft32.asm, cbrev32.asm, and unpacki32.asm)
Arguments
x [nx] Pointer to input vector x containing nx 32-bit real elements.
On output, the vector x contains nx complex elements
corresponding to RIFFT(x) or the signal itself.
nx Number of real elements in vector x. nx can take the following
values.
nx =16, 32, 64, 128, 256, 512, 1024, 2048
Description Computes a Radix-2 real DIT IFFT of the nx real elements stored in vector x
in bit−reversed order. The original content of vector x is destroyed in the
process. The first nx/2 complex elements of the IFFT(x) are stored in vector
x in normal-order.
Algorithm (IDFT)
See CIFFT
J For best performance, the coefficient buffer can be in an SARAM clock

or a DARAM different from the DARAM block that contains the data
buffer.
Implementation Notes Implemented as a complex IFFT of size nx/2 followed by an unpack stage to
unpack the real IFFT results. Therefore, Implementation Notes for the cift32
function apply to this case.

sine
Notice that normally an IFFT of a real sequence of size N, produces a complex

the full information because an IFFT of a real sequence has even symmetry
When scale = 1, this routine prevents overflow by scaling by 2 at each IFFT

Example See examples/rifft subdirectory
sine Sine
Function ushort oflag = sine (DATA *x, DATA *r, ushort nx)
(defined in sine.asm)
Arguments
x[nx] Pointer to input vector of size nx. x contains the angle in
radians between [−π, π] normalized between (−1,1) in q15
format
x = xrad /π
For example:
45o = π/4 is equivalent to x = 1/4 = 0.25 = 0x200 in q15
format.
r[nx] Pointer to output vector containing the sine of vector x in q15
format
nx ≥ 4
Description Computes the sine of elements of vector x. It uses Taylor series to compute
the sine of angle x.
Algorithm for (i + 0; i t nx; i ) )) y(i) + sin(x(i)) where x(i) * xrad

p
4-106
sqrt_16
Implementation Notes Computes the sine of elements of vector x. It uses the following Taylor series
to compute the angle x in quadrant 1 (0−π/2).
sin(x) = c1*x + c2*x^2 + c3*x^3 + c4*x^4 + c5*x^5
c1 = 3.140625x
c2 = 0.02026367
c3 = − 5.3251
c4 = 0.5446778
c5 = 1.800293
The angle x in other quadrant is calculated by using symmetries that map the
angle x into quadrant 1.
Example See examples/sine subdirectory
Overhead: 17
(in bytes)
sqrt_16 Square Root of a 16-bit Number
Function ushort oflag = sqrt_16 (DATA *x, DATA *r, short nx)
(defined in sqrtv.asm)
Arguments
r[nx] Pointer to output vector of size nx containing the sqrt(x).
Description Calculates the square root for each element in input vector x, storing results
in output vector r.
Algorithm for (i + 0; i t nx; i ) )) r [i] + Ǹ(x(i) 0 v i v nx

sub
Implementation Notes The square root of a number(x) can be calculated using Newton’s method. An
initial approximation is guessed and then the approximation gets recomputed
using the formula,
(old approximation 2 * x)
new approximation + old approximation * .
2
The new approximation then becomes the old approximation and the process
is repeated until the desired accuracy is reached.
Example See examples/sqrtv subdirectory
Overhead: 14
(in bytes)
sub Vector Subtract
Function short oflag = sub (DATA *x, DATA *y, DATA *r, ushort nx, ushort scale)
(defined in sub.asm)
Arguments
y[nx] Pointer to input data vector 2 of size nx

r[nx] Pointer to output data vector of size nx containing
- (x−y) if scale =0
- (x−y)/2 if scale =1

nx ≥ 4
4-108
sub
scale Scale selection

- If scale = 1, divide the result by 2 to prevent overflow.
- If scale = 0, do not divide by 2.

Description This function subtracts two vectors, element by element.
Algorithm for (i + 0; i t nx; i ) )) z(i) + x(i) * y(i)
Example See examples/sub subdirectory
Overhead: 23
Code size 60
(in bytes)

Chapter 5
DSPLIB Benchmarks and Performance Issues
All functions in the DSPLIB are provided with execution time and code size
benchmarks. While developing the included functions, we tried to compromise
between speed, code size, and ease of use. However, with few exceptions, the
highest priority was given to optimize for speed and ease of use, and last for
code size.
Even though DSPLIB can be used as a first estimation of processor perfor-

mance for a specific function, you should know that the generic nature of
DSPLIB may add extra cycles not required for customer specific usage.
Topic Page
5.1 What DSPLIB Benchmarks are Provided . . . . . . . . . . . . . . . . . . . . . . . . 5-2

5.2 Performance Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-2
5-1
What DSPLIB Benchmarks are Provided
5.1 What DSPLIB Benchmarks are Provided

DSPLIB documentation includes benchmarks for instruction cycles and
memory consumption. The following benchmarks are typically included in the
assembly source files:
- Calling and register initialization overhead
- Number of cycles in the kernel code: Typically provided in the form of an

equation that is a function of the data size parameters. We consider the
kernel (or core) code, the instructions contained between the _start and
_end labels that you can see in each of the functions.
- Memory consumption: Typically program size in bytes is reported. For

functions requiring significant internal data allocation, data memory con-
sumption is also provided. When stack usage for local variables is mini-
mum, that data consumption is not reported.
For functions in which it is difficult to determine the number of cycles in the ker-
nel code as a function of the data size parameters, we have included direct
cycle count for specific data sizes.
5.2 Performance Considerations

Benchmark cycles presented assume best case conditions, typically assum-
ing:
- 0 wait-state memory external memory for program and data
- data allocation to on-chip DARAM
- no pipeline hits
A linker command file showing the memory allocation used during testing and
benchmarking in the Code Composer C55x Simulator is included under the
example subdirectory.
Remember, execution speed in a system is dependent on where the different

sections of program and data are located in memory. Be sure to account for
such differences, when trying to explain why a routine is taking more time that
the reported DSPLIB benchmarks.
5-2
Chapter 6
Software Updates and Customer Support
This chapter details the software updates and customer support issues for the
TMS320C55x DSPLIB.
Topic Page
6.1 DSPLIB Software Updates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-2

6.2 DSPLIB Customer Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-2
6-1
DSPLIB Software Updates
6.1 DSPLIB Software Updates

C55x DSPLIB Software updates will be periodically released, incorporating
product enhancement and fixes.
DSPLIB Software Updates will be posted as they become available in the

same location you download this information. Source Code for previous re-
leases will be kept public to prevent any customer problem in case we decide
to discontinue or change the functionality of one of the DSPLIB functions.
Make sure to read the readme.1st file available in the root directory of every
release.
6.2 DSPLIB Customer Support

If you have any questions or want to report problems or suggestions regarding
the C55x DSPLIB, contact Texas Instruments at dsph@ti.com.
We encourage the use of the software report form (report.txt) contained in the
DSPLIB root directory to report any problem associated with the C55x
DSPLIB.
6-2
Appendix
AppendixAA
Overview of Fractional Q Formats
Unless specifically noted, DSPLIB functions use Q15 format or to be more ex-
act Q0.15. In a Qm.n format, there are m bits used to represent the two’s com-
plement integer portion of the number, and n bits used to represent the two’s
complement fractional portion. m+n+1 bits are needed to store a general Qm.n
number. The extra bit is needed to store the sign of the number in the most-sig-
nificant bit position. The representable integer range is specified by (−2 m, 2 m)
and the finest fractional resolution is 2 *n .
For example, the most commonly used format is Q.15. Q.15 means that a
16-bit word is used to express a signed number between positive and negative
1. The most-significant binary digit is interpreted as the sign bit in any Q format
number. Thus in Q.15 format, the decimal point is placed immediately to the
right of the sign bit. The fractional portion to the right of the sign bit is stored
in regular two’s complement format.
Topic Page
A.1 Q3.12 Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-2

A.2 Q.15 Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-2
A.3 Q.31 Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-2
A-1
Q3.12 Format
A.1 Q3.12 Format

Q.3.12 format places the sign bit after the fourth binary digit from the right, and
the next 12 bits contain the two’s complement fractional component. The
approximate allowable range of numbers in Q.3.12 representation is (−8,8)
and the finest fractional resolution is 2 *12 + 2.441 10 *4 .
Table A−1. Q3.12 Bit Fields

Bit 15 14 13 12 11 10 9 … 0
Value S I3 I2 I1 Q11 Q10 Q9 … Q0
A.2 Q.15 Format

Q.15 format places the sign bit at the leftmost binary digit, and the next 15 leftmost
bits contain the two’s complement fractional component. The approximate allowable
range of numbers in Q.15 representation is (−1,1) and the finest fractional resolution
is 2 *15 + 3.05 10 *5 .
Table A−2. Q.15 Bit Fields

Bit 15 14 13 12 11 10 9 … 0
Value S Q14 Q13 Q12 Q11 Q10 Q9 … Q0
A.3 Q.31 Format

Q.31 format spans two 16-bit memory words. The 16-bit word stored in the lower
memory location contains the 16 least-significant bits, and the higher memory
location contains the most-significant 15 bits and the sign bit. The approximate
allowable range of numbers in Q.31 representation is (−1,1) and the finest fractional
resolution is 2 *31 + 4.66 10 *10.
Table A−3. Q.31 Low Memory Location Bit Fields

Bit 15 14 13 12 … 3 2 1 0
Value Q15 Q14 Q13 Q12 … Q3 Q2 Q1 Q0
Table A−4. Q.31 High Memory Location Bit Fields

Bit 15 14 13 12 … 3 2 1 0
Value S Q30 Q29 Q28 … Q19 Q18 Q17 Q16
A-2
Appendix
AppendixBA
Calculating the Reciprocal of a Q15 Number
The most optimal method for calculating the inverse of a fractional number
(Y=1/X) is to normalize the number first. This limits the range of the number
as follows:
0.5 v Xnorm t 1
−1 v Xnorm v −0.5 (1)
The resulting equation becomes
Y+ 1
(Xnorm * 2 *n )
or
Y+ 2n (2)
Xnorm
where n = 1, 2, 3, …, 14, 15
Letting Ye + 2 n :
Ye + 2 n (3)
Substituting (3) into equation (2):
Y + Ye * 1 (4)
Xnorm
Letting Ym + 1 :
Xnorm
Ym + 1 (5)
Xnorm
Substituting (5) into equation (4):
Y + Ye * Ym (6)
For the given range of Xnorm, the range of Ym is:
1 v Ym t 2
−2 v Ym v −1 (7)
To calculate the value of Ym, various options are possible:
a) Taylor Series Expansion
b) 2nd,3rd,4th,.. Order Polynomial (Line Of Best Fit)
c) Successive Approximation
B-1
The method chosen in this example is (c). Successive approximation yields

the most optimum code versus speed versus accuracy option. The method
outlined below yields an accuracy of 15 bits.
Assume Ym (new) + exact value of 1 :

Xnorm
Ym (new) + 1 (c1)
Xnorm
or
Ym (new) * X + 1 (c2)
Assume Ym (old) + estimate of value 1 :

X
Ym (old) * Xnorm + 1 ) Dyx
or
Dxy + Ym (old) * Xnorm * 1 (c3)
where Dyx = error in calculation
Assume that Ym(new) and Ym(old) are related as follows:
Ym (new) + Ym (old) * Dy (c4)
where Dy = difference in values
Substituting (c2) and (c4) into (c3):
Ym (old) * Xnorm + Ym(new) * Xnorm ) Dxy
(Ym (new) ) Dy) * Xnorm + Ym (new) * Xnorm ) Dxy
Ym (new) * Xnorm ) Dy * Xnorm + Ym (new) * Xnorm ) Dxy
Dy * Xnorm + Dxy
Dy + Dxy * 1 (c5)
Xnorm
Assume that 1/Xnorm is approximately equal to Ym(old):
Dy + Dxy * Ym (old) (approx) (c6)
Substituting (c6) into (c4):
Ym (new) + Ym (old) * Dxy * Ym (old) (c7)
Substituting for Dxy from (c3) into (c7):
Ym (new) + Ym (old) * (Ym (old) * Xnorm * 1) * Ym(old)
Ym (new) + Ym (old) * Ym (old)2 * Xnorm ) Ym (old)
Ym (new) + 2 * Ym (old) * Ym (old)2 * Xnorm (c8)
B-2
If after each calculation we equate Ym(old) to Ym(new):
Ym (old) + Ym (new) + Ym
Then equation (c8) evaluates to:
Ym + 2 * Ym * Ym 2 * Xnorm (c9)
If we start with an initial estimate of Ym, then equation (c9) converges to a solu-
tion very rapidly (typically 3 iterations for 16-bit resolution).
is linear interpolation and accomplished by taking the complement of the least
significant bits of the Xnorm value.
Calculating the Reciprocal of a Q15 Number B-3

B-4
Index
Index
A conversion
floating-point to Q15 (fltoq15) 4-62
Q15 to floating-point (q15tofl) 4-95
acorr 4-7
adaptive delayed LMS filter 4-39 convol 4-31
fast implemented 4-41 convol1 4-33
add 4-9 convol2 4-35
arctangent 2 implementation 4-10 convolution 4-31
arctangent implementation 4-11 convolution (fast) 4-33
atan16 4-11 convolution (fastest) 4-35
atan2_16 4-10 corr 4-37
autocorrelation 4-7 correlation
auto (acorr) 4-7
full-length (corr) 4-37
B
base 10 logarithm 4-78 D
base 2 logarithm 4-80
base e logarithm 4-81 decimating FIR filter 4-52
bexp 4-13 dlms 4-39
block exponent implementation 4-13 dlmsfast 4-41
double-precision IIR filter 4-67
C DSPLIB
arguments 3-2
cascaded IIR direct form I 4-73 calling a function from assembly language source
code 3-3
cascaded IIR direct form II 4-69, 4-71
calling a function from C 3-3
cbrev 4-14 content 2-2
cbrev32 4-15 data types 3-2
cfft 4-16 dealing with overflow and scaling issues 3-4
cfft32 4-19 how to install 2-3
cfir 4-21 how to rebuild 2-4
cifft 4-26
cifft32 4-28 E
complex bit reverse 4-14
32-bit 4-15 expn 4-45
complex FIR filter 4-21 exponential base e 4-45
Index-1
Index
F iircas51 4-73
iirlat 4-75
FFT index and value of maximum element of a vec-
forward complex tor 4-85
cfft 4-16 index and value of minimum element of a vec-
cfft32 4-19 tor 4-88
forward real, in-place (rfft) 4-100, 4-102 index of maximum element of a vector 4-82
inverse complex index of maximum element of a vector less than or
cifft 4-26 equal to 34 4-84
cifft32 4-28
index of minimum element of a vector 4-86
inverse real, in-place (rifft) 4-104, 4-105
interpolating FIR filter 4-55
fir 4-46
inverse complex FFT 4-26
FIR filter 4-46
32-bit 4-28
complex (cfir) 4-21
decimating (firdec) 4-52 inverse real FFT , in-place 4-104, 4-105
direct form (fir) 4-46
Hilbert Transformer 4-63
interpolating (firinterp) 4-55
L
lattice forward (firlat) 4-57 lattice forward (FIR) filter 4-57
symmetric (firs) 4-59 lattice inverse (IIR) filter 4-75
FIR Hilbert Transformer 4-63 ldiv16 4-77
fir2 4-49 log_10 4-78
firdec 4-52 log_2 4-80
firinterp 4-55 logarithm
firlat 4-57 base 10 (log_10) 4-78
firs 4-59 base 2 (log_2) 4-80
floating-point to Q15 conversion 4-62 base e (logn) 4-81
fltoq15 4-62 logn 4-81
forward complex FFT 4-16
32-bit 4-19 M
forward real FFT, in-place 4-100, 4-102
matrix multiplication 4-89
H matrix transpose 4-90

maxidx 4-82
hilb16 4-63 maxidx34 4-84
maximum element of a vector
I index (maxidx) 4-82

index and value (maxvec) 4-85
IIR filter maximum element of a vector less than or equal to
cascaded, direct form I (iircas51) 4-73 34, index (maxidx34) 4-84
cascaded, direct form II (iircas4) 4-69 maximum value of a vector 4-85
cascaded, direct form II (iircas5) 4-71 maxval 4-85
double-precision (iir32) 4-67 maxvec 4-85
lattice inverse (iirlat) 4-75 minidx 4-86
iir32 4-67 minimum element of a vector
iircas4 4-69 index (minidx) 4-86
iircas5 4-71 index and value (minvec) 4-88
Index-2
Index
minimum value of a vector 4-87 random number generation

minval 4-87 algorithm 4-96
minvec 4-88 initialization 4-98
mmul 4-89 recip16 4-99
mtrans 4-90 rfft 4-100, 4-102
mul32 4-91 rifft 4-104, 4-105
N S
natural logarithm (logn) 4-81 sine 4-106
neg 4-92 sqrt_16 4-107
neg32 4-93 square root of a 16-bit number 4-107
sub 4-108
P symmetric FIR filter 4-59
power 4-94
V
Q vector add 4-9
Q15 to floating-point conversion 4-95 vector negate 4-92
q15tofl 4-95 vector negate, double-precision 4-93
vector power 4-94
R vector subtract 4-108

16-bit reciprocal function 4-100
rand16 4-96 32-bit by 16-bit long division function 4-79
rand16init 4-98 32-bit vector multiplication 4-93
Index-3
IMPORTANT NOTICE
Texas Instruments Incorporated and its subsidiaries (TI) reserve the right to make corrections, enhancements, improvements and other
changes to its semiconductor products and services per JESD46, latest issue, and to discontinue any product or service per JESD48, latest
issue. Buyers should obtain the latest relevant information before placing orders and should verify that such information is current and
complete. All semiconductor products (also referred to herein as “components”) are sold subject to TI’s terms and conditions of sale
supplied at the time of order acknowledgment.
TI warrants performance of its components to the specifications applicable at the time of sale, in accordance with the warranty in TI’s terms
and conditions of sale of semiconductor products. Testing and other quality control techniques are used to the extent TI deems necessary
to support this warranty. Except where mandated by applicable law, testing of all parameters of each component is not necessarily
performed.
TI assumes no liability for applications assistance or the design of Buyers’ products. Buyers are responsible for their products and
applications using TI components. To minimize the risks associated with Buyers’ products and applications, Buyers should provide
adequate design and operating safeguards.
TI does not warrant or represent that any license, either express or implied, is granted under any patent right, copyright, mask work right, or
other intellectual property right relating to any combination, machine, or process in which TI components or services are used. Information
published by TI regarding third-party products or services does not constitute a license to use such products or services or a warranty or
endorsement thereof. Use of such information may require a license from a third party under the patents or other intellectual property of the
third party, or a license from TI under the patents or other intellectual property of TI.
Reproduction of significant portions of TI information in TI data books or data sheets is permissible only if reproduction is without alteration
and is accompanied by all associated warranties, conditions, limitations, and notices. TI is not responsible or liable for such altered
documentation. Information of third parties may be subject to additional restrictions.
Resale of TI components or services with statements different from or beyond the parameters stated by TI for that component or service
voids all express and any implied warranties for the associated TI component or service and is an unfair and deceptive business practice.
TI is not responsible or liable for any such statements.
Buyer acknowledges and agrees that it is solely responsible for compliance with all legal, regulatory and safety-related requirements
concerning its products, and any use of TI components in its applications, notwithstanding any applications-related information or support
that may be provided by TI. Buyer represents and agrees that it has all the necessary expertise to create and implement safeguards which
anticipate dangerous consequences of failures, monitor failures and their consequences, lessen the likelihood of failures that might cause
harm and take appropriate remedial actions. Buyer will fully indemnify TI and its representatives against any damages arising out of the use
of any TI components in safety-critical applications.
In some cases, TI components may be promoted specifically to facilitate safety-related applications. With such components, TI’s goal is to
help enable customers to design and create their own end-product solutions that meet applicable functional safety standards and
requirements. Nonetheless, such components are subject to these terms.
No TI components are authorized for use in FDA Class III (or similar life-critical medical equipment) unless authorized officers of the parties
have executed a special agreement specifically governing such use.
Only those TI components which TI has specifically designated as military grade or “enhanced plastic” are designed and intended for use in
military/aerospace applications or environments. Buyer acknowledges and agrees that any military or aerospace use of TI components
which have not been so designated is solely at the Buyer's risk, and that Buyer is solely responsible for compliance with all legal and
regulatory requirements in connection with such use.
TI has specifically designated certain components as meeting ISO/TS16949 requirements, mainly for automotive use. In any case of use of
non-designated products, TI will not be responsible for any failure to meet ISO/TS16949.
Products Applications
Audio www.ti.com/audio Automotive and Transportation www.ti.com/automotive
Amplifiers amplifier.ti.com Communications and Telecom www.ti.com/communications
Data Converters dataconverter.ti.com Computers and Peripherals www.ti.com/computers
DLP® Products www.dlp.com Consumer Electronics www.ti.com/consumer-apps
DSP dsp.ti.com Energy and Lighting www.ti.com/energy
Clocks and Timers www.ti.com/clocks Industrial www.ti.com/industrial
Interface interface.ti.com Medical www.ti.com/medical
Logic logic.ti.com Security www.ti.com/security
Power Mgmt power.ti.com Space, Avionics and Defense www.ti.com/space-avionics-defense
Microcontrollers microcontroller.ti.com Video and Imaging www.ti.com/video
RFID www.ti-rfid.com
OMAP Applications Processors www.ti.com/omap TI E2E Community e2e.ti.com
Wireless Connectivity www.ti.com/wirelessconnectivity
Mailing Address: Texas Instruments, Post Office Box 655303, Dallas, Texas 75265
Copyright © 2013, Texas Instruments Incorporated

TMS320VC5502 Library Reference

Uploaded by

TMS320VC5502 Library Reference

Uploaded by

TMS320C55x DSP Library

SPRU422J − May 2000

Read This First

About This Manual

The Texas Instruments TMS320C55x™ DSPLIB is an optimized DSP Function

- Lehmer, D.H. “Mathematical Methods in large-scale computing units.”

- Oppenheim, Alan V. and Ronald W Schafer. Discrete-Time Signal Proces-

- Digital Signal Processing with the TMS320 Family (SPR012)

- TMS320C55x DSP CPU Reference Guide (SPRU371)

- TMS320C55x Optimizing C Compiler User’s Guide (SPRU281)

TMS320, TMS320C55x, and C55x are trademarks of Texas Instruments.

Matlab is a trademark of Mathworks, Inc.

Read This First iii

5 DSPLIB Benchmarks and Performance Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1

5.2 Performance Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-2

6 Software Updates and Customer Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1

7 Overview of Fractional Q Formats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-1

8 Calculating the Reciprocal of a Q15 Number . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-1

4−1 dbuffer Array in Memory at Time j . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-24

4−1 Function Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-2

The Texas Instruments TMS320C55x DSP Library (DSPLIB) is an optimized

1.1 DSP Routines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2

1.1 DSP Routines

- Fast-Fourier Transforms (FFT)

- Filtering and convolution

1.2 Features and Benefits

- C-callable routines fully compatible with the TI C55x compiler

- Fractional Q15-format operand supported

- Complete set of examples on usage provided

- Benchmarks (time and code) provided

- Tested against Matlab™ scripts

This chapter describes how to install the DSPLIB.

2.1 DSPLIB Content . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2

2.1 DSPLIB Content

2) One object library under the ”lib” sub−directory:

2.2 How to Install DSPLIB

2.2.1 De-Archive DSPLIB

2.2.2 Relocate Library File

For example, if your TI C5500 tools are located in c:\ti\c5500\cgtools\bin and

Installing DSPLIB 2-3

2.3 How to Rebuild DSPLIB

2.3.1 For Full Rebuild of 55xdsp.lib

2.3.2 For Partial Rebuild of 55xdsp.lib

2) Re-assemble your new fir.asm assembly source file:

This chapter describes how to use the DSPLIB.

3.1 DSPLIB Arguments and Data Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2

3.1 DSPLIB Arguments and Data Types

3.1.1 DSPLIB Arguments

- Vector stride is always equal 1: vector operands are composed of vector

- Complex elements are assumed to be stored in a Re-Im format.

- In-place computation is allowed (unless specifically noted): Source

3.1.2 DSPLIB Data Types

- Q.3.12 : Contains 3 integer bits and 12 fractional bits.

Unless specifically noted, DSPLIB operates on Q15-fractional data type

3.2 Calling a DSPLIB Function from C

- Include the dsplib.h include file

- Use a correct linker command file describing the memory configuration

Refer to the TMS320C55x Optimizing C Compiler User’s Guide (SPRU281).

3.3 Calling a DSPLIB Function from Assembly Language Source Code

Realize that the TI DSPLIB is not an optimal solution for assembly-only

3.4 Where to Find Sample Code

Using DSPLIB 3-3

3.5 How DSPLIB is Tested − Allowable Error

We consider this testing methodology a good first pass approximation. Further

3.6 How DSPLIB Deals with Overflow and Scaling Issues

It’s important to note that a TMS320C55x architectural feature that makes

- Scaling implemented for overflow prevention: In this type of function,

- No scaling implemented for overflow prevention: In this type of func-

- Saturation implemented for overflow handling: In this type of function,

- DSPLIB reporting of overflow conditions (overflow flag): Due to the

As a final note, DSPLIB is provided also in source format to allow customiza-

Using DSPLIB 3-5

3.7 Where DSPLIB Goes From Here

- Increased number of functions: We anticipate the number of functions

- Increased Code portability: DSPLIB looks to enhance code portability

This chapter provides descriptions for the TMS330C55x DSPLIB functions.

Function void cbrev (DATA , DATA r, ushort)

Function void cbrev32(LDATA , LDATA r, ushort)

ȍ x [i] < ǒcos ǒ* 2 nxp i * kǓ ) j sin ǒ* 2 nxp i * kǓǓ