[JSR308] Annotations for worst-case execution time

[Igor Peshansky]

Michael Ernst wrote on 01/30/2007 07:44:08 PM:

> Trevor-
> 
> Annotations on loops and other statements are within the scope of JSR 
308,
> as noted by the current version of the proposal.  However, no one has 
yet
> put forward a specific technical proposal for the syntax of 
loop/statement
> annotations (in Java and in the class file).  Once someone does so, then 
we
> can evaluate that proposal and accept, improve, or reject it.
> 
> I know that several other people, including Wayne Carr and Eugene 
Kuleshov
> (who responded separately to your message), are also interested in this
> topic.  Perhaps you could join forces in preparing a proposal or at 
least
> preparing a statement that broaches the issues to the mailing list.

Hi, everyone,

While I can't yet contribute to the set of new ideas, there was a detailed
proposal for the syntax and bytecode representation for various types of
annotations mentioned above in my Ph.D. thesis (Section 6.3, pp. 84-95):
<http://cs.nyu.edu/pechtcha/pubs/thesis.html>.  This is quite dated, and
was implemented on top of Jikes rather than Javac, but can serve as a
basis for future discussion.
        Igor
-- 
Igor Peshansky  (note the spelling change!)
IBM T.J. Watson Research Center
XJ: No More Pain for XML's Gain (http://www.research.ibm.com/xj/)
X10: Parallel Productivity and Performance (http://x10.sf.net/)