henry@utzoo.uucp (Henry Spencer) (04/27/89)
In article <8055@boring.cwi.nl> dik@cwi.nl (Dik T. Winter) writes: >... I said something else: Cray's choice predates C; you know, >you will find Seymour Cray's choice in such relicts as the CDC 7600. The choice being discussed here is not the size of floating point on the machine, but the mapping from C's sizes to machine sizes. There's nothing wrong with making "float" and "double" the same size, and this is clearly what should have been done. -- Mars in 1980s: USSR, 2 tries, | Henry Spencer at U of Toronto Zoology 2 failures; USA, 0 tries. | uunet!attcan!utzoo!henry henry@zoo.toronto.edu
davidsen@steinmetz.ge.com (Wm. E. Davidsen Jr) (04/28/89)
In article <1989Apr26.170635.3933@utzoo.uucp> henry@utzoo.uucp (Henry Spencer) writes: | The choice being discussed here is not the size of floating point on the | machine, but the mapping from C's sizes to machine sizes. There's nothing | wrong with making "float" and "double" the same size, and this is clearly | what should have been done. As a matter of interest, the only thing on the Cray2 C which has really bitten me is the 64 bit short. When porting code to the C2 I found a number of places in which the size of int could vary without problem, but short>16 bits caused trouble. The recent net posting of the lzhuf program suffers from this... There's still at least one place in which truncation to 16 bits is assumed implicitly, and I hope to find it. -- bill davidsen (wedu@crd.GE.COM) {uunet | philabs}!steinmetz!crdos1!davidsen "Stupidity, like virtue, is its own reward" -me