This discussion has been locked.

You can no longer post new replies to this discussion. If you have a question you can start a new discussion

PUSH and POP in Keil C

dear all, how can I use PUSH and POP instruction in Keil C ?
Thanks

Parents

0 Graham Cole over 21 years ago in reply to Ingvar Kreft

The only way to do this just in C is to create a pseudo stack (an array) and a pseudo pointer. Simply write your own pseudo_push() and pseudo_pop() functions.

There is a relatively safe way to do this using 8051 assembly language functions that can be called from a C51 function. You will need something like this:

#pragma ASM

    $REGUSE _push_byte( A, DPH, DPL )

#pragma ENDASM

void push_byte( char x )
{                               //
    x = x;                      // Supress UNUSED warning.
                                //
    #pragma ASM

        POP     DPH             ;Get return address from stack
        POP     DPL             ;and store in DPTR
        MOV     A,R7            ;Get x into accumulator...
        PUSH    Acc             ;...and push it onto the stack.
        CLR     A               ;
        JMP     @A+DPTR         ;Return to caller.

    #pragma ENDASM
}

#pragma ASM

    $REGUSE _pop_byte( A, DPH, DPL, R7 )

#pragma ENDASM

char pop_byte( void )
{                               //
                                //
    #pragma ASM

        POP     DPH             ;Get return address from stack
        POP     DPL             ;and store in DPTR
        POP     Acc             ;Get the return value from stach
        MOV     R7,A            ;and put it in R7.
        CLR     A               ;
        JMP     @A+DPTR         ;Return to caller.

    #pragma ENDASM

    return(0);                  // Dummy return.
}

And use it like this:


void push_byte( char x );
char pop_byte( void );

void main( void )
{
    unsigned char loop;
    char          a;
    char          b;

    a = 0;

    loop = 255;

    do
    {
        push_byte( a );
        b = pop_byte();
        a++;
    }while( --loop != 0 );

    loop = 255;
}

This is much safer than placing pushes and pops directly in-line because the compiler expects a function call to disrupt register values etc.

From my C51 experience, my guess is that this will work fairly reliably. But, and its a big BUT, there are no garantees. Even if it works for now, it could stop working at the next compiler update!

Anyway, why do you want to do this?

Reply

0 Graham Cole over 21 years ago in reply to Ingvar Kreft

#pragma ASM

    $REGUSE _push_byte( A, DPH, DPL )

#pragma ENDASM

void push_byte( char x )
{                               //
    x = x;                      // Supress UNUSED warning.
                                //
    #pragma ASM

        POP     DPH             ;Get return address from stack
        POP     DPL             ;and store in DPTR
        MOV     A,R7            ;Get x into accumulator...
        PUSH    Acc             ;...and push it onto the stack.
        CLR     A               ;
        JMP     @A+DPTR         ;Return to caller.

    #pragma ENDASM
}

#pragma ASM

    $REGUSE _pop_byte( A, DPH, DPL, R7 )

#pragma ENDASM

char pop_byte( void )
{                               //
                                //
    #pragma ASM

        POP     DPH             ;Get return address from stack
        POP     DPL             ;and store in DPTR
        POP     Acc             ;Get the return value from stach
        MOV     R7,A            ;and put it in R7.
        CLR     A               ;
        JMP     @A+DPTR         ;Return to caller.

    #pragma ENDASM

    return(0);                  // Dummy return.
}

And use it like this:


void push_byte( char x );
char pop_byte( void );

void main( void )
{
    unsigned char loop;
    char          a;
    char          b;

    a = 0;

    loop = 255;

    do
    {
        push_byte( a );
        b = pop_byte();
        a++;
    }while( --loop != 0 );

    loop = 255;
}

Children

No data