Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watnot!watmath!clyde!rutgers!husc6!sri-unix!hplabs!decwrl!labrea!Shasta!andy
From: andy@Shasta.UUCP
Newsgroups: comp.arch
Subject: RISC register windows
Message-ID: <1335@Shasta.STANFORD.EDU>
Date: Wed, 25-Feb-87 03:46:25 EST
Article-I.D.: Shasta.1335
Posted: Wed Feb 25 03:46:25 1987
Date-Received: Fri, 27-Feb-87 22:35:04 EST
References: <1881@homxc.UUCP> <898@moscom.UUCP> <476@mntgfx.MENTOR.COM>
Reply-To: andy@Shasta.UUCP (Andy Freeman)
Organization: Stanford University
Lines: 28

The sliding register window had an area for parameters passed
to the procedure, locals, and parameters passed to other procedures.
Calling lots of little procedures (that don't call other procedures)
that don't use their window fully is equivalent to doing a better
job allocating but having one less window.  The first call might
overflow, but subsequent ones won't.

There were also a set of global registers.

One argument for fixed window sizes is that register access
shouldn't require an addition.  Bit-wise or is one thing (or the
register number with the current base to determine the register);
addition is an expensive operation to put on the register access
data path.  This basically restricts window sizes to powers of two.
(Then again, if there are cheap functions that one could use to
implement variable size register windows, I'm sure someone on
the net will mention them.)  It then is probably easier to pick
a power than support a range of them.

Then again, I seem to remember that the window size wasn't a power
of two on the Berkeley RISC, which surprised me.  Maybe I'm wrong.

-andy
-- 
Andy Freeman
UUCP:  ...!decwrl!shasta!andy forwards to
ARPA:  andy@sushi.stanford.edu
(415) 329-1718/723-3088 home/cubicle