Xref: utzoo comp.sys.ibm.pc:29817 comp.sys.intel:832
Path: utzoo!censor!geac!jtsv16!uunet!husc6!rutgers!apple!vsi1!daver!apt!brian
From: brian@apt.UUCP (Brian Litzinger)
Newsgroups: comp.sys.ibm.pc,comp.sys.intel
Subject: Re: 80286, 80386 LOADALL instructions.
Summary: LOADALL in the 386 may be gone, but it's not dead
Keywords: LOADALL 80386
Message-ID: <1657@apt.UUCP>
Date: 5 Jun 89 07:20:31 GMT
References: <30105@conexch.UUCP> <213@guardian.UUCP>
Followup-To: poster
Organization: APT Technology, Inc., San Jose, CA
Lines: 27

In article <213@guardian.UUCP>, peter@guardian.UUCP (peter) writes:
> In article <30105@conexch.UUCP> rob@conexch.UUCP (Robert Collins) writes:
> >What about the 80386 version of LOADALL (Opcode 0f07).  Does anybody have
> >the document on that?
> 
> I've heard from a reliable source that future steppings of the
> Intel386(tm) will NOT recognize the LOADALL instruction.  Intel is rewriting
> all 386 software which includes LOADALL; I suspect the gang in Redmond is
> doing the same, but that's pure speculation.  I'd strongly encourage everyone
> to avoid using LOADALL on a 386.

While the LOADALL instruction is, in fact, gone from the Intel386(tm),
it is definitely not dead.

In fact, on many 386 machines you can execute the 286 LOADALL opcode
and exactly the right things will happen.  Not because the particular
386 processor happens to have the instruction, but because the system
BIOS vendor choose to emulation the instruction in the BIOS.

Some BIOSes catch the invalid opcode interrupt, check for the 286
LOADALL instruction, and then emulate it it software.

I do, however, agree with peter...Avoid the LOADALL instruction on the 386!

<>  Brian Litzinger @ APT Technology Inc., San Jose, CA
<>  UUCP:  {apple,sun,pyramid}!daver!apt!brian    brian@apt.UUCP
<>  VOICE: 408 370 9077      FAX: 408 370 9291