On Mon, Oct 01, 2007 at 03:09:16PM -0700, Davide Libenzi wrote:

On Mon, 1 Oct 2007, Paul E. McKenney wrote:

quoted

That would indeed be one approach that CPU designers could take to
avoid being careless or sadistic.  ;-)

That'd be the easier (unique maybe) approach too for them, from an silicon 
complexity POV. Distinguishing between different CPUs stores once inside a 
shared store buffer, would require tagging them in some way. That'd defeat 
most of the pros of having a shared store buffer ;)

Tagging requires but one bit per entry.  Depends on the workload -- if
lots of barriers, bursty stores and little sharing, tagging might win.
If lots of sharing, then your suggested approach might win.

						Thanx, Paul