Path: utzoo!utgpu!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!bloom-beacon!mit-eddie!uw-beaver!cornell!rochester!pt.cs.cmu.edu!andrew.cmu.edu!jk3k+
From: jk3k+@andrew.cmu.edu (Joe Keane)
Newsgroups: comp.arch
Subject: Re: The & (address) operator and register allocation
Message-ID: 
Date: 5 Dec 88 20:48:24 GMT
References: <1224@cps3xx.UUCP> <1988Dec3.221843.28966@utzoo.uucp>,
	<1586@nmtsun.nmt.edu>
Organization: Carnegie Mellon
Lines: 10
In-Reply-To: <1586@nmtsun.nmt.edu>

Those were the good old days, when you could jump to a register for a really
fast loop.  But i don't think you want to do this now, because it'll slow the
machine down a good deal.  Have any new architectures proposed memory-mapping
the registers?

However, with caller saves, you can easily ensure that the register is always
saved to the same address, and this'd be `address' of the register.  Should be
no overhead.  Have any compilers done this?

--Joe