🚀 go-pugleaf

In article <405@ssp2.idca.tds.philips.nl> pb@idca.tds.PHILIPS.nl (Peter Brouwer) writes:
| elg@killer.Dallas.TX.US (Eric Green) writes:
| >davidsen@steinmetz.ge.com (Wm. E. Davidsen Jr) says:

	[ ... my introductory remarks stripped ... ]

| >>   I checked the size of image files compiled from the same source on
| >> Xenix/386 v2.3.1 and SunOS 3.4 (on a 3/280 if anyone cares). I noted
| >
| >A possible flaw: I seem to recall that SunOS's old compilers, at
| From this and previous postings I get the impression than one is comparing
| apples and oranges.

| I think that comparing code size should be done on code that is generated
| with the same operating system, probably unix.

  I realize that BSD bigots hate to recognize any other flavor of unix,
but for most of us SysV, Xenix, Ultrix, SunOS and BSD are all called
UNIX.

| If that is the case than the 386 compiler does not use the segment registers
| in the user code. All memory references will be 32 bits references.
| The compiler we have for the 386 on unix is from AT&T unix V release 3.1
| It does not use the segment registers, that leaves 8 registers. From
| these registers 3 are used for the c register declarations ( 4 in 3.2 ).
| For the motorola the compiler can hold ten register declarations.
| This make a big diffrence in code size and execution speed.
| 
| So before more comparisons are made lets do it in the same operating
| environment and not compare code generated under MSDOS with code generated
| under UNIX.

  This totally mystifies me. Did someone bring in some MS-DOS figures?
My figures were clearly for Xenix, and I clearly stated the versions and
stuff (I think the info was quoted and appears above).
-- 
	bill davidsen		(wedu@crd.GE.COM)
  {uunet | philabs}!steinmetz!crdos1!davidsen
"Stupidity, like virtue, is its own reward" -me