wide arguments and/or argument wide

[shirono at hcx3.SSD.HARRIS.COM]

In comp.lang.c, gl8f at bessel.acc.Virginia.EDU writes:
> Why are arguments to C functions always widened to ints or doubles?

PDP 11.

> This
> makes sense when the argument is being widened to the buswidth (which
> ints are supposed to be), but doubles are often larger than that.

The reason for auto-type-conversion to double was because the FPP on the
PDP 11 could operate in one of two modes: single and double precision, but
going from one to the other was rather costly.  This (I believe) was the
reason why DMR specified that all floating point arithmetic was to be
performed in double: have the program startup routine set the FPP to double
mode and just keep going.

--Roberto
______________________________________________________________________________
                               ||   Internet: shirono at ssd.harris.com
     Roberto Shironoshita      ||
      Harris Corporation       ||             ...!novavax---\
   Computer Systems Division   ||   UUCP:     ...!uunet-------!hcx1!shirono
                               ||             ...!mit-eddie-/
------------------------------------------------------------------------------
DISCLAIMER: The opinions expressed here are my own; they in no way reflect the
            opinion or policies of Harris Corporation.