This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Might a -native-semantics switch, forcing native target optimization semantics, be reasonable?


Paul Schlie wrote:

- Are there any particular formally "undefined" language semantics you
 perceive as being difficult associate with an alternatively well defined
 target specific implementation behavior?  As if not, I can only interpret
 your response as being itself both seemingly unfounded and meaningless.




IN every case where the standard specifies undefined behavior, it has a very good reason for doing so.
The burden is on you to describe what on earth you are talking about. And you need to be addressing
the semantics required in the language of the standard. You seem to have some vague informal notion
of semantics here which is quite unclear. The notion of native target semantics is also very vague,
since of course on any modern archictecture, any reasonable semantics can be implemented.


I think you should start with a quote from the standard regarding undefined behavior. Then for
some particular target, define an example of what you are proposing. To fill out the proposal
you then have to set up a giant matrix with all the undefined behaviors on one axis, and all
the possible targets on the other, and fill in the intersections. Not an easy task at all!




Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]