Newsgroups: comp.arch
Path: utzoo!henry
From: henry@zoo.toronto.edu (Henry Spencer)
Subject: Re: Bitfield instructions--a good idea?
Message-ID: <1991Apr23.152155.2298@zoo.toronto.edu>
Date: Tue, 23 Apr 1991 15:21:55 GMT
References: <1991Apr15.193425.3436@waikato.ac.nz> <SPOT.91Apr18123711@WOOZLE.GRAPHICS.CS.CMU.EDU> <PUWA8R4@xds13.ferranti.com> <1991Apr23.053619.13474@kithrup.COM>
Organization: U of Toronto Zoology

In article <1991Apr23.053619.13474@kithrup.COM> sef@kithrup.COM (Sean Eric Fagan) writes:
>>Wouldn't this tend to be a cache-basher? Particularly a problem when you're
>>already bashing the cache with the bitmap anyway?
>
>Perhaps that's where a 'don't cache' attribute on a load/store might be
>useful... 

Anybody who caches a frame buffer is crazy.  Especially if the cache isn't
write-through, in which case your frame-buffer updates show up on the screen
some arbitrary time later!

Table-lookup algorithms (the original topic) can be somewhat hard on caches,
although the impact is minimized if the tables are small (i.e. lookup by
bytes, not by shorts!) and the cache is large.
-- 
And the bean-counter replied,           | Henry Spencer @ U of Toronto Zoology
"beans are more important".             |  henry@zoo.toronto.edu  utzoo!henry
