Write an ARM assembly language subroutine that performs according to the following specification (Caesar cypher).

Given a mixed case alphabetic string from the C driver program, convert it to all upper case and create a new string for return to the C program. The new string has the characters from the old string replaced by upper case characters from the alphabet offset by the given amount in the alphabet.

If the character string is 'Abc' with an offset of -3 the new string would be 'XYZ'. For characters in the alphabet such as 'mno', the alphabet is 'JKL'. The beginning of the alphabet is wrapped to the end.

To convert a character to upper case, AND the character with 0xDF. Do not process the space character (leave it as a space) or the NULL byte terminator but be sure to include the NULL byte in the new string.

The C driver code:

/* C code for cipher */

#include

#include

extern char * cypher( char * s, int offset ) ;

int main( int argc, char * argv[] )

{

            char string[] = "The quick brown fox jumped over the lazy dog" ;

            char * result ;

            int offset = -5 ;

            result = cypher( string, offset ) ;

            printf( "Original:%s\n", string ) ;

            printf( "Coded:   %s\n", result ) ;

            exit( 0 ) ;

}

The assembly routines definition is:

            extern char * cypher( char * s, int offset )

                        where char * s is the pointer to the string,

                                    int offset is the integer distance to move

                        returns a pointer to a string.

To determine the length of the string in order to reserve space for the new string use the library routine strlen. Register a1 contains a pointer to the string to measure and on return register a1 has the number of characters in the string. You need to add one to this result since strlen does not account for the null byte at the end.

You can pass this length to the malloc library routine to obtain space for the new string.