From: francis.glassborow@btinternet.com   
      
   On 07/04/2013 04:34, Balog Pal wrote:   
   > Err, it knows better what exactly? We were talking about cases where   
   > the object was just constructed, destructed, and in between only   
   > operations that the compiler can figure out the end result at   
   > compile time. That would be replaced with just result if no call to   
   > op new/delete pair was sitting there.   
      
   But the implementer already has a licence to use such an optimisation   
   providing that there are also no side-effects within the code for   
   operator new and and operator delete.   
      
   Actually I am always a little concerned when a compiler decides that   
   it can remove my code. At the very least it should warn me and provide   
   a mechanism by which I can insist that it does it my way.   
      
   Note that that implies that I should always be able to turn off link   
   time optimisations that are based on whole code analysis.   
      
   Also note that even copy ctor elision is strictly controlled and is   
   only permitted in certain well defined contexts.   
      
      
   Francis   
      
      
   --   
         [ See http://www.gotw.ca/resources/clcm.htm for info about ]   
         [ comp.lang.c++.moderated.    First time posters: Do this! ]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)