Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watnot!watmath!clyde!rutgers!lll-lcc!ames!ucbcad!ucbvax!CORY.BERKELEY.EDU!dillon
From: dillon@CORY.BERKELEY.EDU.UUCP
Newsgroups: comp.sys.amiga
Subject: Re: Turbo Pascal Rumor
Message-ID: <8704031026.AA17840@cory.Berkeley.EDU>
Date: Fri, 3-Apr-87 05:26:19 EST
Article-I.D.: cory.8704031026.AA17840
Posted: Fri Apr  3 05:26:19 1987
Date-Received: Sun, 5-Apr-87 01:31:30 EST
Sender: daemon@ucbvax.BERKELEY.EDU
Lines: 21

>I am under the (possibly wrong!) impression that the Amiga's intuition,
>exec, and dos functions have been written with C interfaces. So if 
>Borland wants to enter the Amiga market, why not with a C compiler?
>Is he afraid there are too many C compilers already out there?
>I am under the impression that the various versions of Manx and Lattice
>that are out are a) expensive and b) bug ridden. There must be
>room in the market for an inexpensive (ala other Borland products)
>C compiler that works correctly.

	All of the Amiga's libraries were designed for easy access by
assembly language.  That is, all parameters are passed in registers 
rather than on the stack.  The libraries are also designed, however, to
make C interfacing easy in that a simple MOVEM.L from the stack to a set
of registers usually suffices to take C pushed arguments on the stacks and
place them in registers before making the library call.  This is handled by
the link time libraries.  E.G. you make a call to, say, FindTask(name),
and the function FindTask() moves the pushed pointer into the proper
register, loads A6 with the library vector base, and makes an A6 
relative call to the proper library vector table entry.

					-Matt