AN669 查看數據表（PDF） - Microchip Technology

零件编号

产品描述 (功能)

生产厂家

AN669

Embedding Assembly Routines into C Language Using a Floating Point Routine as an Example

Microchip Technology 

SUMMARY

For this discussion only the 32-bit floating point multiply

is used. However, the same principles of embedded

assembly language routines into C code can be used

with other assembly language routines. A summary list

of a step- by- step process to embed assembly code

into your C code is below:

• Convert assembly register EQU equates to C vari-

able types such as unsigned int.

• Convert constants to #define in C.

• Place the assembly code into a subroutine using

#asm and #endasm

• To avoid paging issues in parts with multiple pro-

gram memory pages, force the code to an

address where it will not cross a page boundary.

For example:

#pragma memory ROM [MAXROM-0x800] @ 0x800;

• Macros and conditional assembly will have to be

rewritten in actual in-line assembly code. The

MPLAB-C compiler does not support these higher

level assembly options to the same degree as the

assembler, MPASM.

For your convenience, all the 32-bit floating point rou-

tines in application note AN575 are provided in a zip file

along with this application note. Each routine has been

separated to work as a stand-alone routine. There is a

AN669

separate file for each floating point routine. The files

may be included individually into your C code. Table 2

shows a list of all the files and routines included with

this application note.

TABLE 2: 32-BIT FLOATING POINT C FILES/FUNCTIONS INCLUDED WITH THIS APPLICATION

NOTE

AN575 Original

Assembly Routine/file *

Equivalent C

file/function

Purpose

-

FLO2432

FLO3232

FPD32

FPM32

FPA32

FPS32

INT3224

example.c

flo2432.inc

flo3232.inc

fpd32.inc

fpm32.inc

fpsa32.inc

fps32() 32-bit subtract

fpa32() 32-bit add

int3224.inc

The example main() routine calculating

the area given the radius. (uses fpm32)

24-bit integer to 32-bit floating point conversion

32-bit integer to 32-bit floating point conversion

32-bit floating point divide

32-bit floating point multiply

32-bit floating point add

32-bit floating point subtract

32-bit floating point to 24-bit integer conversion

INT3232

int3232.inc

32-bit floating point to 32-bit integer conversion

NRM3232

nrm3232.inc

32-bit normalization of unnormalized 32-bit

floating point numbers

NRM4032

nrm4032.inc

32-bit normalization of unnormalized 40-bit

floating point numbers

math16.inc

math16c.c

variables and constants need for the floating point functions

* Check Microchip web site and bulletin board for latest code.

© 1997 Microchip Technology Inc.

DS00669A-page 5

Share Link: