> > > > > BTW, although malloc() returns a pointer "suitable aligned", that
> > > > > doesn't mean it is fast.
> > > I mean that the x86 cache is divided into 16-byte chunks and they are
> > > loaded using RAM burst mode starting at the beginning of the paragraph,
> Now I'm completely confused. The topic was aligning memory for faster
> access, but you now seem to be saying that a) it doesn't make any real
> difference and b) alignment for faster access is off topic.
Oh! I'm sorry! My point is that alignment on natural boundaries (i.e.
int at 4 bytes, etc.) and *recognition* of this fact by algorithms is
more than enough for OpenSSL. Alignment at cache line boundaries is
architecture dependent and is getting off topic as I don't beleive it'll
make hell of a difference in OpenSSL case. Sorry about
misunderstanding:-)
Andy.
______________________________________________________________________
OpenSSL Project http://www.openssl.org
Development Mailing List [EMAIL PROTECTED]
Automated List Manager [EMAIL PROTECTED]
