>>> Dan Magenheimer <dan.magenheimer@oracle.com> 13.01.09 00:29
>>>
>I''m guessing the gcc optimizer for the memcpy code was tuned
>for an Intel pipeline... Jan, were you measuring on an
>AMD processor?
Oh, actually my previous reply was without pushing my thinking fully back
to what I was doing (and measuring) back then. I really measured on quad
core Xeons, as what I was looking at were the highmem helpers. Pretty
likely this almost exclusively covered the cold cache (all levels) case, but I
also think this is the most likely scenario.
Jan
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel
