Last update January 22, 2007

Doc Comments /


More Information

You need to talk about the representation of integers: 1's complement, 2's complement, sign magnitude, padding bits, trap representations, etc. What happens when you cast a signed negative integer to an unsigned integer, what operations can generate -0, and do the arithmetic operators use the value type or the object type? -- anonymous


Add your comments here...

real type size

real: largest hardware implemented floating point size (Implementation Note: 80 bits for Intel CPU's)

  • On the PowerPC, a "real" is either 64 bits (currently) or 128 bits (eventually, or on AIX) - implemented using two 64-bit registers ("double-double"). There are no 80-bit FPU registers. The double-double (AIX) type is not IEEE 754 compliant, which makes it unsuitable for D.
  • On the SPARC, a "real" type is 128 bits and using IEEE-compliant quadruple precision. There are no 80-bit FPU registers.

Integer Promotions

Integer Promotions are conversions of the following types:

 char 	int
 wchar 	int
 dchar 	uint
Why are not char and wchar promoted to uint as well ? Typo ?


FrontPage | News | TestPage | MessageBoard | Search | Contributors | Folders | Index | Help | Preferences | Edit

Edit text of this page (date of last change: January 22, 2007 21:26 (diff))