In your previous email, you also suggest : <pre>-did not try to "multiplex" code modules with "path registration" or "handler registration"</pre>What do you mean exactly ? Jerome <div class="gmail_quote">2008/11/5 Havoc Pennington <<a href="mailto:hp@pobox.com">hp@pobox.com</a>> <blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> Hi, On Wed, Nov 5, 2008 at 8:41 AM, Jerome Philbert <div class="Ih2E3d"><<a href="mailto:jerome.philbert@gmail.com">jerome.philbert@gmail.com</a>> wrote: </div><div class="Ih2E3d">> I think I have also tried your suggestions number 3 and 4. > Is not it equivalent to: > - remove paranoid validation ? (by using > DBUS_VALIDATION_MODE_WE_TRUST_THIS_DATA_ABSOLUTELY in the code) > - build D-Bus with disable-checks option ? > In this way, I have measured that D-Bus was 7% faster. > </div>I think if you wrote a library from scratch, you could disable even more validation, basically - write code that would just crash and burn horribly on invalid data, that never checked anything at all, but was pretty fast. If you profile just plain raw sockets with no protocol parsing, exchanging messages in a blocking way, that will tell you what the absolute theoretical fastest library you could write would be like. Havoc </blockquote></div>