SPLIT 16 BIT VAR IN 2 VAR OF ONE BYTE

[appavito]

hi i need to split a 16 bit variabile in 2 variabile of one byte on c language.. and viceversa.. the MSB in the first var and the LSB on the second var...

can hel me???

[slandrum]

Do you mean something like this?

Code: Select all

   short var;          // 16 bit variable
   char hi, lo;        // 8 bit variables
   hi = (var>>8)&255;  // set hi to the upper 8 bits of var
   lo = (var&255);     // set lo to the lower 8 bits of var
   var = (hi<<8) + lo; // set var to 16 bit value from hi and lo


Or do you mean something else?

[IceCube]

Yo could do something like this:

Code: Select all

typedef union {
   unsigned int Word;
   struct {
      unsigned char Low;
      unsigned char High;
   } Byte;
} WordByteType;

WordByteType Value, AnotherValue;

Value. Word = 0x1234;
UART_PutChar(Value.Byte.Low);  //Send 0x34
UART_PutChar(Value.Byte.High);  //Send 0x12

AnotherValue.Byte.Low = 0x78;
AnotherValue.Byte.High = 0x65;
PWM16_WritePeriod(AnotherValue.Word);   // Write 0x6578

Regards,
Rolf

[appavito]

good, but i not undestend something:

example var=1122; // 1122=0000010001100010 bynary
// hi =00000100 lo 01100010

hi = (var>>8)&255; // 0000000000000100 shift 8 position and insert 8 zero.
// typecast hi= var ( only the lsb ) hi= 00000100

BUT WHY USE &255 ????

[slandrum]

Because if the number is negative, zeros won't get shifted into the upper bits, ones will.

A smart compiler will not generate any code for the &255, but it also prevents complaints for implied casting and out of range values if the number you are shifting is a constant.