From: dave.g4ugm@gmail.com   
      
   On 04/10/2015 14:15, Francois LE COAT wrote:   
   > Hi,   
   >   
   > Michael Schwingen writes:   
   >> Francois LE COAT wrote:   
   >>> Well, the starting demo with a spinning hypercube is not played. The   
   >>> GEM interface seems correct, but if I want to describe a curve, the   
   >>> curve is not drawn. If I want to draw a surface, the surface is   
   >>> not drawn. Nothing happens with the binary like it should. The   
   >>> binary is simply not corresponding to sources. The program is broken.   
   >>   
   >> Im my experience with optimizing compilers, in most such cases the   
   >> fault is   
   >> not with the compiler, but instead the source code is broken, doing   
   >> things   
   >> that are not allowed by the C standard and relying on undefined   
   >> behaviour.   
   >>   
   >> If it worked with the older compiler (that had a weaker optimizer)   
   >> that does   
   >> not mean anything for the correctness of the source code.   
   >>   
   >> Did you compile with all warning enabled, and look at the warnings? I   
   >> can't   
   >> believe old, misbehaving code would compile after a gcc3 -> gcc4 switch   
   >> without producing at least some new warnings!   
   >   
   > You'll agree that it's very peculiar ... How bizarre a warning will   
   > generate an error when building C program's sources ?   
      
   This is typical of "C" and is one of the problems with the language,   
   Most compilers do not warn about undefined behaviour. Try posting this   
   on comp.lang.c and you will be told exactly the same.   
      
      
   > This is not   
   > a warning, but should be alerted as an error, don't you think so ?   
   > I never seen before in the C standard definition that a warning   
   > should imperatively be taken into account, otherwise generating an   
   > error. The warnings must often, used to generate good code, not errors.   
   > Strong optimizations mean that warnings are now considered as errors ?   
   >   
      
   The "C" standard has changed substancially with the introduction of C11.   
   I work on the Hercules project which is an IBM Mainframe emulator and we   
   are having the same problems...   
      
      
   > Please take into account that I practice C language since 1986,   
   > first Kernighan and Ritchie, then ANSI C standard. Notice the   
   > C standard must have evoluted because my C sources are now obsolete.   
   >   
      
   Possibly need adjustments to cope with changes...   
      
   > Thanks for your answer.   
   >   
   > Best regards,   
   >   
      
   Dave Wade   
   G4UGM   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)