Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!samsung!munnari.oz.au!sirius.ucs.adelaide.edu.au!augean!sibyl!ian From: ian@sibyl.eleceng.ua.OZ (Ian Dall) Newsgroups: comp.arch Subject: Re: Alignment on RS/6000 Message-ID: <893@sibyl.eleceng.ua.OZ> Date: 21 Nov 90 00:16:33 GMT References: Reply-To: ian@sibyl.OZ (Ian Dall) Organization: Engineering, Uni of Adelaide, Australia Lines: 20 In article aglew@crhc.uiuc.edu (Andy Glew) writes: >I'll bet the misalignment slowdown noted in the last post was due to >trapping and handling the misalignment in the trap handler. > >But it makes you think: is it really worth being backward compatible >if there is such a huge performance penalty? In a similar vein, how many people have been caught by a floating point program taking "forever" on a sparc (no doubt othe machines as well) because it was spending all it's time doing NaN and Inf exception handling? I know the handling of these faults can be changed, but the point is that the result takes *so* much longer to calculate as to be useless. Much better to make the default to core dump on a floating point exception. -- Ian Dall life (n). A sexually transmitted disease which afflicts some people more severely than others.