On Oct 14, 2011, at 15:27, John Ehrman wrote: > > (1) Addresses and offsets aren't known until the end of the first pass over > the source, when the values of all symbols are known. This means it's > difficult to capture the needed information in time for use by conditional > assembly. > How many passes does HLASM make over the source?
> (2) Suppose you find that an address can use a short displacement. > Subsequent instructions need long displacements, which makes the > displacement assigned to the first instruction too small. This could lead > to a situation where the displacements oscillate back and forth for each > pass. (Similar problems occur with text formatters, where shorter or > longer page numbers, or movement of figures from places where they fit, can > change with each pass over the text.) > Similar problems exist with the old-fashioned type attribute. I can code a test which, depending on the type attribute, defines that symbol with a different type. This really ought to cause an assembly error. -- gil
