Path: utzoo!attcan!uunet!ncrlnk!ncr-sd!hp-sdd!hplabs!hpl-opus!hpccc!brown
From: brown@hpccc.HP.COM (Jeffrey L. Brown)
Newsgroups: comp.dcom.lans
Subject: Re: Re: Microsoft C calls to Novell 2.11 SFT
Message-ID: <5820001@hpccc.HP.COM>
Date: 10 Oct 88 17:31:05 GMT
References: <124@avatar.UUCP>
Organization: Hewlett-Packard Corporate Computing Center
Lines: 31

/ hpccc:comp.dcom.lans / kory@avatar.UUCP (Kory Hamzeh) /  5:18 pm  Oct  6, 1988 /
In article <267@serene.CTS.COM>, rfarris@serene.CTS.COM (Rick Farris) writes:
> 	inreg.h.al = 0xdb;

>Shouldn't the function value 0xdb be put in the ah register as opposed to the
>al register? Thats the way dos works.
>
>
>
>-- 
>-------------------------------------------------------------------------------
>Kory Hamzeh			    UUCP:     ..!uunet!psivax!quad1!avatar!kory
>				    INTERNET: avatar!kory@quad.com

It's been a few months since I last wrote Microsoft C Code for Netware.  Where
I worked we were using V2.10.  Every Netware call I ever saw used the AH
register for passing the function code.  Considering that Novell seems to
develop a new way to use functions at every release, this little bit of
commonality (sp?) was amazing.  Since changing registers for the function call
would more than likely break a lot of software, I'm willing to bet, though,
that inreg.h.al should be inreg.h.ah.  Now the big question:  are you sure
that 0xdb is a function code and not a "sub-function selector"?  A lot of the
older (ie., non-packet) calls used al to specify a "sub-function" (I guess
that's as good a name as any; I don't know what it's actually called....)

Like I said, I used to write netware code with Microsoft C V5.0, but I haven't
done so in 2 months (new job), and I don't know how much the API in 2.11 (or
2.15) has changed.  Take all this with a beach (lots of grains :-) ) of salt.

Jeff Brown
brown%hpccc@hplabs.hp.com
...hplabs!hpccc!brown