On Fri, 2003-03-21 at 09:29, Fergus Henderson wrote: > On 21-Mar-2003, Richard Torkar <[EMAIL PROTECTED]> wrote: > > On Fri, 2003-03-21 at 04:10, Fergus Henderson wrote: > > > Note that the exception handling scheme used on Windows requires some > > > overhead for "try" statements even in the case when no exception is thrown. > > > > What? ^^^^^^^^^^^^^^ How can that be the case? > > > > If the catch does not fire then the try should not be evaluated, right? > > This is something I've missed completely. > > The Win32 "structured exception handling" mechanism that Windows uses > keeps a linked list of active exception handlers. When entering a > "try" statement, the code generated by the compiler needs to insert > a new handler in the list, and to fill in a variety of fields that > record information about the handler. When exiting a "try" statement, > the generated code needs to remove the handler from the list.
Good explanation, thanks! /Richard -- Ph.D. Student Dept. of Informatics and Mathematics HTU _______________________________________________ Mono-list maillist - [EMAIL PROTECTED] http://lists.ximian.com/mailman/listinfo/mono-list
