Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!crdgw1!crdos1!davidsen
From: davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr)
Newsgroups: comp.arch
Subject: Re: Bitfield instructions--a good idea?
Message-ID: <3357@crdos1.crd.ge.COM>
Date: 19 Apr 91 14:29:29 GMT
References: <1991Apr15.193425.3436@waikato.ac.nz> <1991Apr18.093804.18183@odin.diku.dk> <1991Apr18.195841.10588@zoo.toronto.edu>
Reply-To: davidsen@crdos1.crd.ge.com (bill davidsen)
Organization: GE Corp R&D Center, Schenectady NY
Lines: 19

In article <1991Apr18.195841.10588@zoo.toronto.edu> henry@zoo.toronto.edu (Henry Spencer) writes:

| Unless access to memory is seriously expensive, table lookup can implement
| both of these operations quite efficiently.  Agreed, it's not as fast as
| having them wired in...

  This is true, but the original topic was somewhat more general,
breaking N bit chunks into parts and assembling them. In the case where
N is a power of two I think the table lookup loses if I want the data
elsewhere. I can generate it faster than I can read it from memory (on
most machines). 

  The problem comes when I want to go from words back to byte packing.
In this case a lookup doesn't help, and I have to fetch, shift, and or
the bits regardless of wether I do it with an instruction or a loop.
-- 
bill davidsen	(davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen)
        "Most of the VAX instructions are in microcode,
         but halt and no-op are in hardware for efficiency"