From: Eugene Leitl <eugene DOT leitl AT lrz DOT uni-muenchen DOT de>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Date: Wed, 14 Jul 1999 18:56:11 -0700 (PDT)
To: pgcc AT delorie DOT com
Cc: pcg AT goof DOT com
Subject: calling functions
In-Reply-To: <3.0.32.19990715013633.010d9ab0@pop.xs4all.nl>
References: <3 DOT 0 DOT 32 DOT 19990715013633 DOT 010d9ab0 AT pop DOT xs4all DOT nl>
X-Mailer: VM 6.43 under 20.4 "Emerald" XEmacs  Lucid
Message-ID: <14221.16161.37716.619890@lrz.de>
Reply-To: pgcc AT delorie DOT com
X-Mailing-List: pgcc AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

Vincent Diepeveen writes:
 > Good morning,
 > 
 > Reading through the ia-64 technical manual the merced 
 > seems to have many registers:
 
The downside of having many registers is of course the need to save a
lot of state when the need to switch context arises.

Out of curiousity, what do people think about viability of
architectures like http://www.cs.cmu.edu/~koopman/stack_computers/
these days?
 
 > 128 general purpose  (64 bits)
 > 128 floating point   (82 bits)
 > 64  predicate        (1 bit)
 > 128 special purpose
 > 
 > One of the reasons they give why it has so many
 > registers, they mention that they want to prevent
 > that calling a function gives a lot of overhead.
 > 
 > What is currently the overhad of a function call at the
 > PII, that this needs heavy optimization?
 > 
 > How many clocks for a typical call does pgcc need if we
 > have 2 integer parameters for it?
 > 
 > Greetings,
 > Vincent