[m-dev.] WAM and PARMA style variable bindings

[Fergus Henderson]

On 28-Apr-2004, Ralph Becket <rafe at cs.mu.OZ.AU> wrote:
> I seem to recall reading somewhere that the average pointer chain length
> in either style is virtually one for most programs anyway.  In which
> case I reckon there's going to be little to choose between the two
> approaches.  Can anyone confirm or refute me on this one (with a
> reference)?

Touati and Despain [1] said that most chains are typically of length 0
and that there is almost no chain of length > 1.

However, in Mercury presumably you would use non-solver types in most
cases, which would skew the statistics.

But even if the chain length is was always one,
there may well be a performance difference between having a dereference
loop which always executes one iteration and having a single dereference.

[1] H. Touati and A. Despain, An Empirical Study of the Warren Abstract
Machine, Proceedings of the 1987 Symposium on Logic Programming, San
Francisco, August/September 1987, IEEE Computer Society Press.



-- 
Fergus Henderson                    |  "I have always known that the pursuit
                                    |  of excellence is a lethal habit"
WWW: <http://www.cs.mu.oz.au/~fjh>  |     -- the last words of T. S. Garp.
--------------------------------------------------------------------------
mercury-developers mailing list
Post messages to:       mercury-developers at cs.mu.oz.au
Administrative Queries: owner-mercury-developers at cs.mu.oz.au
Subscriptions:          mercury-developers-request at cs.mu.oz.au
--------------------------------------------------------------------------