Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!ucsd!ucbvax!bloom-beacon!eru!hagbard!sunic!mcsun!ukc!cam-cl!news
From: cet1@cl.cam.ac.uk (C.E. Thompson)
Newsgroups: comp.arch
Subject: Re: Alignment on RS/6000
Message-ID: <1990Nov23.144234.12191@cl.cam.ac.uk>
Date: 23 Nov 90 14:42:34 GMT
References: <AGLEW.90Nov19212926@cobra.crhc.uiuc.edu> <AGLEW.90Nov19234534@lasso.crhc.uiuc.edu> <MCCALPIN.90Nov20092451@pereland.cms.udel.edu>
Reply-To: cet1@cl.cam.ac.uk (C.E. Thompson)
Organization: U of Cambridge Comp Lab, UK
Lines: 16

In article <MCCALPIN.90Nov20092451@pereland.cms.udel.edu> mccalpin@perelandra.cms.udel.edu (John D. McCalpin) writes:
>
>The example given was not terribly representative of real codes.  A
>*much* more likely cause of poor alignment is placing a
>double-precision entity after an odd number of integer items in a
>common block.  In these cases the user is given an informative error
>message by the compiler (or loader?).
>
But this specific case (double floating point load/store access on a
multiple of 4 which is not a multiple of 8) *is* dealt with in hardware on
the RS/6000, isn't it? Reference not immediately to hand, but surely I didn't
dream it?

Chris Thompson
JANET:    cet1@uk.ac.cam.phx
Internet: cet1%phx.cam.ac.uk@nsfnet-relay.ac.uk