From: leathm AT solwarra DOT gbrmpa DOT gov DOT au (Leath Muller)
Message-Id: <199709092239.IAA08502@solwarra.gbrmpa.gov.au>
Subject: Re: fread slowstart
To: eliz AT is DOT elta DOT co DOT il (Eli Zaretskii)
Date: Wed, 10 Sep 1997 08:39:00 +1000 (EST)
Cc: sandmann AT clio DOT rice DOT edu, dj AT delorie DOT com, djgpp-workers AT delorie DOT com
In-Reply-To: <Pine.SUN.3.91.970909143740.7143N-100000@is> from "Eli Zaretskii" at Sep 9, 97 02:38:09 pm
Content-Type: text
Precedence: bulk

> > I normally don't comment too much - but I would vote for even a
> > larger starting size and then quadruple... ie: start at say 4k, then
> > 16, then 32 etc. Why not do this?
 
> I think we should time several programs with different options and see
> the results, instead of guessing what would be the best initial size
> and the multiple.  Voting will not help here.
 
Consider it a figure of speech - not a vote... I checked through a lot of
my old code last night and found that about half of my headers/.c files
are around 2-3k, which means the whole file would be loaded in one hit...
The rest of the files were proportionally larger....

Maybe a test could be something like Allegro... it's probably one of the
bigger currently available real world testing suites that are suitable as
a good test bed. Of course the other would be gcc itself... ;)

Leathal.