Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Keil C51 Cross Compiler: Embedded Systems 2 2 - 1 C Programming

Download as pdf or txt
Download as pdf or txt
You are on page 1of 35

Keil C51 Cross Compiler

ANSI C Compiler Generates fast compact code for the 8051 and its derivatives Advantages of C over Assembler Do not need to know the microcontroller instruction set Register allocation and addressing modes are handled by the compiler Programming time is reduced Code may be ported easily to other microcontrollers (not 100% portable) C51 supports a number of C language extensions that have been added to support the 8051 microcontroller architecture e.g. Data types Memory types Pointers Interrupts
2-1 C Programming

Embedded Systems 2

C vs Assembly Language
Code efficiency can be defined by 3 factors: 1. How long does it take to write the code 2. What size is the code? (how many Bytes of code memory required?) 3. How fast does the code run? C is much quicker to write and easier to understand and debug than assembler. Assembler will normally produce faster and more compact code than C Dependant on C compiler A good knowledge of the micro architecture and addressing modes will allow a programmer to produce very efficient C code

Embedded Systems 2

2-2

C Programming

C51 Keywords

Embedded Systems 2

2-3

C Programming

C51 Data Types

= extension to ANSI C
Embedded Systems 2 2-4 C Programming

C51 Memory Models


Small All variables stored in internal data memory Be careful stack is placed here too Generates the fastest, most efficient code Default model for Keil uVision Projects Compact All variables stored in 1 page (256 bytes) of external data memory Accessed using MOVX @R0 Slower than small model, faster than large model Large All variables stored in external data memory (up to 64KByte) Accessed using MOVX @DPTR Generates more code than small or compact models

Embedded Systems 2

2-5

C Programming

C51 Memory Models

Select Model

Embedded Systems 2

2-6

C Programming

C51 Memory Types


Memory type extensions allow access to all 8051 memory types. A variable may be assigned to a specific memory space The memory type may be explicitly declared in a variable declaration variable_type <memory_type> variable_name; e.g. int data x; Program Memory CODE memory type Up to 64Kbytes (some or all may be located on 8051 chip) Data Memory 8051 derivatives have up to 256 bytes of internal data memory Lower 128 bytes can be directly or indirectly addressed Upper 128 bytes shares the same address space as the SFR registers and can only be indirectly addressed Can be expanded to 64KBytes off-chip

Embedded Systems 2

2-7

C Programming

C51 Memory Types


code Program memory (internal or external). unsigned char code const1 = 0x55; //define a constant char code string1[] = hello; //define a string data Lower 128 bytes of internal data memory Accessed by direct addressing (fastest variable access) unsigned int data x; //16-bit variable x idata All 256 bytes of internal data memory (8052 micro) Accessed by indirect addressing (slower)

Embedded Systems 2

2-8

C Programming

C51 Memory Types


bdata Bit addressable area of internal data memory (addresses 20H to 2FH) Allows data types that can be accessed at the bit level unsigned char bdata status; sbit flag1 = status^0; xdata External data memory Slower access than internal data memory unsigned char xdata var1;

Embedded Systems 2

2-9

C Programming

8051 Memory Usage


In a single chip 8051 application data memory is scarce 128 or 256 bytes on most 8051 derivatives Always declare variables as the smallest possible data type Declare flags to be of type bit Use chars instead of ints when a variables magnitude can be stored as 8 bits Use code memory to store constants and strings

Embedded Systems 2

2 - 10

C Programming

Example Code (1)


sbit input = P1^0; void main() { unsigned char status; int x; for(x=0;x<10;x++) { status = input; } } How can the variable declarations be improved?

Embedded Systems 2

2 - 11

C Programming

Program Branching
Most embedded programs need to make some decisions e.g. perform a certain task only if a switch input is high An if-else statement is the most common method of coding a decision box of a flowchart. if (P1^0 == 1) { //task A } else { //task B } Note that the brackets are only required if the if or else statement contains more then 1 line of code
Embedded Systems 2 2 - 12

P1.0 = 1?

Task A

Task B

C Programming

Wait Loops
In some applications the program must wait until a certain condition is true before progressing to the next program task (wait loop) e.g. in a washing machine wait for the water to heat before starting the wash cycle A while statement is often used here //task A while(P1^0 == 0); //will stay here while P1.0 is low //can also write as while(P1^0 == 0) {} //task B
Task A

P1.0 = 1? Y Task B

Embedded Systems 2

2 - 13

C Programming

Continuous Loops
An embedded program never ends It must contain a main loop than loops continuously e.g. a washing machine program continually tests the switch inputs A continuous loop is programmed using a loop with no condition while(1){ } while(1); for(;;){ } for(;;);

Embedded Systems 2

2 - 14

C Programming

C Bitwise Operators
NOT ~ char data x = 0x05; x = ~x; //x=0xFA AND & char data x = 0x45; x &= 0x0F; //x = 0x05 Useful for clearing specific bits within a variable (masking) OR | char data x = 0x40; x |= 0x01; //x = 0x41 Useful for setting individual bits within a variable EXCLUSIVE OR ^ char data x = 0x45; x ^= 0x0F; //x = 0x4A Useful for inverting individual bits within a variable
2 - 15 C Programming

Embedded Systems 2

Question
What is the value of variable x after the following code executes? unsigned char data x; x = 21; x &= 0xF0; x |= 0x03; Write C code to set the upper 4 bits of a char to 1100.

Embedded Systems 2

2 - 16

C Programming

C Logical Operators
Bitwise operators will change certain bits within a variable Logical operators produce a true or false answer They are used for looping and branching conditions C Logical Operators AND OR Equals Not Equal Less Less or Equal Greater Greater or Equal

&& || == != < <= > >=


2 - 17 C Programming

Embedded Systems 2

Example Code (2)


//program to add 2 8-bit variables //a flag should be set if the result exceeds 100 void main() { unsigned char data num1, num2; unsigned int data result; bit overflow; num1 = 10; num2 = 25; result = num1 + num2; if (result >100) overflow = 1; }
Embedded Systems 2 2 - 18 C Programming

Accessing Port Pins


Writing to an entire port P2 = 0x12; //Port 2 = 12H (00010010 binary) P2 &= 0xF0; //clear lower 4 bits of Port 2 P2 |= 0x03; //set P2.0 and P2.1

Reading from a port unsigned char data status; status = P1; //read from Port 1 Before reading from a port pin, always write a 1 to the pin first.

Embedded Systems 2

2 - 19

C Programming

Reading 8051 Port Pins


We can only read the proper logic value of a pin if the latch contains a 1. If the latch contains a 0 the FET is on and a 0 will always be read.

Embedded Systems 2

2 - 20

C Programming

Example Code (3)


//Program to read from 8 switches connected to Port 1. The status of the switches //is written to 8 LEDs connected to Port 2. #include <reg51.h> //SFR address definitions void main() { unsigned char data status; //variable to hold switch status //continuous loop to read switches and display the status while(1) { status = P1; P2 = status; } }

Embedded Systems 2

2 - 21

C Programming

Accessing Individual Port Pins


Writing to a port pin //write a logic 1 to port 0 pin 1 P0^1 = 1; In a program using lots of individual port pins it is better coding practice to name the pins according to their function sbit power_led = P0^1; power_led = 1; Reading from a port pin //Turn on LED if switch is active sbit switch_input = P2^0; if (switch_input) power_led = 1; else power_led = 0;
2 - 22 C Programming

Embedded Systems 2

Example Code (4)


//program to flash an LED connected to Port 2 pin 0 every 0.2 seconds #include <reg51.h> sbit LED = P2^0; void delay(); void main() { while (1) { LED = 0; //LED off delay(); LED = 1; //LED on delay(); } } //Delay function void delay() { . } Embedded Systems 2 2 - 23 C Programming

Generating Delays
Software Delays Uses looping mechanisms in C or assembly Does not use any microcontroller hardware resources Ties up the CPU while the delay is running Delay produced depends on the compiler Hardware Delays Uses a microcontroller timer Uses very little CPU resources (runs in background)

Embedded Systems 2

2 - 24

C Programming

Software Delay
Use a for loop Does not use any timer resources Uses CPU resources while running i.e. no other task can be performed during delay loop Can result in large amounts of machine code being generated Results may be different for different compilers The following code results in a 204 usecond delay for the 8051 operating off the 12MHz oscillator For loops can be nested to produce longer delays

void delay() { unsigned char x; for (x=100; i > 0; x--); }


Embedded Systems 2 2 - 25 C Programming

Software Delay
Using a decrementing for loop will generate a DJNZ loop. Loop execution time = 204 machine cycles [1 + 1 + (100 *2) + 2] = 204usec for a 12MHz crystal. Change starting value to 98 to get a 200usec delay

Embedded Systems 2

2 - 26

C Programming

Hardware Delay
Use timer 0 or timer 1 Very efficient mechanism of producing accurate delays Timer mode, control and counter registers must be configured Timer run bit is used to start/stop the timer The timer flag can be polled to determine when required time delay has elapsed Using timer in interrupt mode uses very little CPU resources See timer notes for more details

Embedded Systems 2

2 - 27

C Programming

C51 Functions
Can specify Register bank used Memory Model Function as an interrupt Re-entrancy [return_type] func_name([args]) [model] [re-entrant] [interrupt n] [using n] int square(int x) { return(x * x); }

Embedded Systems 2

2 - 28

C Programming

Re-entrant Functions
A re-entrant function is a function that can be called while it is still running. e.g. an interrupt occurs while the function is running and the service routine calls the same function. Keil compiler supports re-entrant functions. Beware of stack limitations

Embedded Systems 2

2 - 29

C Programming

Scope of Variables
A variable defined within a function will default to an automatic variable An automatic variable may be overlaid i.e. the linker may use the variables memory space for a variable in another function call. This will cause the variable to lose its value between function calls. If you want a variable to maintain its value between function calls, declare it as a static variable. static int x; Static variables cannot be overlaid by the linker Declare a variable as volatile if you want its value to be read each time it is used

Embedded Systems 2

2 - 30

C Programming

Function Parameter Passing


Up to 3 arguments may be passed in registers This improves system performance as no memory accesses are required If no registers are available fixed memory locations are used

Embedded Systems 2

2 - 31

C Programming

Function Return Values


Function return values are always passed in registers

Embedded Systems 2

2 - 32

C Programming

Interrupt Functions
Interrupt vector number is provided in the function declaration void timer0_int() interrupt 1 using 2 { . } Contents of A, B, DPTR and PSW are saved on stack when required All working registers used in the ISR are stored on the stack if the using attribute is not used to specify a register bank All registers are restored before exiting the ISR

Embedded Systems 2

2 - 33

C Programming

Absolute Variable Location


The _at_ keyword is used to assign an absolute memory address to a variable Useful for accessing memory mapped peripherals or specific memory locations Syntax type [memory_space] variable_name _at_ constant unsigned char xdata lcd _at_ 0x8000 Absolute variables may not be initialised

Embedded Systems 2

2 - 34

C Programming

C51 Pointers
Generic pointers Same as ANSI C pointers char *ptr; Stored using 3 bytes Can be used to access any memory space Code generated executes more slowly than memory specific pointers Memory Specific Pointers Specify the memory area pointed to char data *ptr1; //pointer to char in data memory Stored as 1 (data, idata) or 2 bytes (code or xdata) Can also specify where pointer is stored char data * xdata ptr2; //pointer to data stored in xdata

Embedded Systems 2

2 - 35

C Programming

You might also like