Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watmath!clyde!rutgers!sri-spam!ames!amdahl!amdcad!tim
From: tim@amdcad.UUCP
Newsgroups: comp.lang.c,comp.sys.m68k
Subject: Ackermann's function is wrong in Moto Benchmark Report
Message-ID: <16731@amdcad.AMD.COM>
Date: Wed, 20-May-87 13:11:34 EDT
Article-I.D.: amdcad.16731
Posted: Wed May 20 13:11:34 1987
Date-Received: Fri, 22-May-87 01:05:10 EDT
Organization: Advanced Micro Devices, Inc., Sunnyvale, Ca.
Lines: 34
Xref: utgpu comp.lang.c:2120 comp.sys.m68k:459


In the "Apples to Oranges" MC68020 Benchmark Report Motorola put out in
1986, the C code for Ackermann's function was shown as:

/* Ackerman's function (in C) */

main()
{
	A(3,6);
}

A(x,y)
int x,y;
{
	if (x==0) return (++y);
	if (y==0) return (A(--x,1));
	return (A(--x,A(x,--y)));
}


The last line is incorrect.  Depending upon evaluation order, the inner
call to the A() function could use the undecremented value for x
(correct) or the decremented value of x (incorrect).  The last line
should be written:

	return (A(x-1,A(x,--y)));


Could someone from Motorola verify that the time to run the correct
version of the benchmark is the same as the time listed for the
incorrect version?

	-- Tim Olson
	Advanced Micro Devices