On Tue, 28 Dec 1999 15:50:10 -0500 
 David Edelsohn <dje@watson.ibm.com> wrote:

 >         So you want to decrease preformance for all programs because a few
 > explicitly want to define misaligned structures?  Seems backwards to me.

So, if unaligned accesses are in fact slower than aligned ones, why
is doing strict alignment a performance penalty?

        -- Jason R. Thorpe <thorpej@nas.nasa.gov>