I said:
>     3) The principle behind the improvements in synchronization speed 
> has been published (in the ASPLOS VII proceedings, I think).

And I should learn to check before opening my mouth.  The article I was
referring to was written by people from IBM TJ Watson Labs, not from
Sun, and was actually published in the 1998 PLDI proceedings.  The
correct citation is:

   Thin Locks: Featherweight Synchronization for Java
   Bacon, et. al.
   ACM SIGPLAN '98 Conference on Programming Language Design and Implementation
   SIGPLAN Notices Volume 33, Number 5, May, 1998

They claim a speed improvement of 5x in the JDK 1.1.2 VM for AIX, for
whatever that signifies for other implementations.
-----------------------------------------------------------------------------
"'Bother,' said Pooh, as Cthulu rose up and ate him."
-----------------------------------------------------------------------------
Bruce Cohen,                               |  email: cohenb at gemstone.com
GemStone Systems, Inc.                     |  phone: (503)533-3602
20575 NW Von Neumann Drive                 |  fax:   (503)629-8556
Beaverton, OR USA 97006                    |  web:   http://www.gemstone.com