Re: Interpreter vs hardware threads

From: Max Skaller (maxs@in.ot.com.au)
Date: Thu Mar 09 2000 - 00:30:16 MET

  • Next message: Benoit Deboursetty: "Re: typing of a class"

    Julian Assange wrote:

    > What is the source of the linearity of thread context switches in
    > ocaml? Are there ways to eliminate it? If so I'd be happy to have a
    > look at doing so.

    One problem with this is that it only affects the bytecode interpreter.
    It would be nice to have lightweight threads available for compiled
    ocaml too.

    The system I need to implement uses a (heavyweight) event dispatcher
    thread and a few (heavyweight) worker threads. The events are
    'executed',
    in other words, the client code built as an object with callbacks.
    This is very fast: there's no need to 'check' for a blocked IO
    operation and restart the 'thread'. The thread is restarted directly
    by the event.

    This all works fine, except that it is very hard to code
    the logic of a 'logical' thread of control. So I need to
    'control invert' a client program, which is written in procedural style,
    to hide the ugly event driven implementation.

    -- 
    John (Max) Skaller at OTT [Open Telecommications Ltd]
    mailto:maxs@in.ot.com.au      -- at work
    mailto:skaller@maxtal.com.au  -- at home
    



    This archive was generated by hypermail 2b29 : Fri Mar 10 2000 - 09:11:05 MET