2002-07/00241.html

<!DOCTYPE html>
<html>
  <head>
    <meta charset="utf-8">
    <title>00241.mbox</title>
  </head>
  <body>
  <div><strong>From</strong>: nisse@lysator.liu.se</div>
  <div><strong>Date</strong>: Tue, 16 Jul 2002 16:09:16 +0200</div>
  <hr/>
  <pre>rreale@iol.it writes:

&gt; But how much performance imprevement would we achieve if we created such
&gt; a mechanism, only for pure data?

The fastest way to do it that I can imagine, for the simplest case of
mutually trusting processes, is to have a (possibly) short message
queue, protected by a mutex and some condition variables, all in th
eshared memory page. It&#39;s not obvious to me if that will be slower or
faster than L4 ipc that can pass the data in registers.

Furthermore, I have never implemented mutexes and condition variables,
but I wouldn&#39;t be surprised if some syscall is needed for conditions
to propagate between threads or processes.

Regards,
/Niels
</pre>
</body>
</html>