---

Programming principle in a function call

---

• What exactly is a function definition (in any programming language)
  • Consider the following function definition:

    float f(float x) { return x*x; }

  • This function definition defines a black box:

    Specifically:

    The parameter definition: float f(float x) { return x*x; } means: The function will require an input value in the float encoding (= in IEEE 754 code) If you need to refresh on the IEEE 754 code, see: click here The return type: float f(float x) { return x*x; } means: The function will return a value is encoded using the float encoding (= IEEE 754)

    ---

    ---

    ---

  • Another example:

    int g(int a) { return a*a; }

    means:

    The function g( ) will require an input value in the 2s complement encoding The function will return a value is encoded using the 2s complement encoding

  ---

  ---

  ---

  ---

• Extremely important fact from the assembler lectures
  • Extremely important fact:

    Computer uses different codes to represent different types of values Example: int x = 2 Binary number (code) in x = 00000000000000000000000000000010 float y = 2; Binary number (code) in y = 01000000000000000000000000000000 It's a IEEE 754 code: 0 10000000 00000000000000000000000 (exponent)

    Here is a C program that shows the binary representation of the values stored in int and float variables:

    #include <stdio.h> int main(int argc, char *argv[] ) { int a = 2; float b = 2; printf("int %d representation in bits: ", a); printBits(&a); // Ignore the definition of this function for now... printf("\n\n"); b = 2; printf("float %2.0f representation in bits: ", b); printBits(&b); printf("\n\n"); } /* ------------------------------------------------------ Print bit representation (Ignore this function for now; it uses some advanced pointer stuff this will be discussed later) ------------------------------------------------------- */ void printBits( void *ptr ) { int x = *((int *)ptr); int i; for ( i = 31; i >= 0; i-- ) { if ( (x & (1 << i)) != 0 ) printf("1"); else printf("0"); } }

    Output:

    int 2 representation in bits: 00000000000000000000000000000010 float 2 representation in bits: 01000000000000000000000000000000

    This is exactly what you have learned in the first part of CS255 !!!

    ---

  • Example Program: (Demo above code)                                                
    • Prog file: /home/cs255001/demo/C/set1/data-repr1.c

    How to run the program:

    To compile:   gcc data-repr1.c          To run: ./a.out

  ---

  ---

  ---

  ---

• What happens when you call a function
  • Consider the following function call to float f(float x):

    float f(float x) { return x*x; } int main( int argc, char* argv[ ]) { .... b = f(a); }

    This is what happens in the function call b = f(a):

    The compiler must make sure that: b = f( a ); the value of parameter a is encoding using float (IEEE 754) The compiler will perform a type conversion operation if necessary !!!           The compiler must also make sure that: b = f( a ); the return value of f( ) is converted to the data type of the variable b !!! The compiler will perform a type conversion operation if necessary !!!

    ---

    ---

  • I will illustrate the conversion that happens in a function call in the next web page

    I will also high light the difference in behavior between C and Java....