On Sat, 5 Jun 2010, Andrew Pinski wrote: > This is why OpenCL spec is not very good here really and not > consistent with the rest of the C/C++ standard. In fact I would say > we should treat it as the same as arrays out of bounds where it is > just undefined at runtime rather than an error. This is what the Cell
Where does the OpenCL spec actually define things in this level of detail anyway? It doesn't generally define things in anywhere near as much precision as the C standard - that is, clarifying whether something is a constraint violation (error or pedwarn needed), undefined behavior as a property of a program (error or pedwarn OK) or undefined behavior as a property of a particular execution of a program (error or pedwarn not OK, warning OK, trap on execution OK). Looking through chapter 6 of the OpenCL spec version 1.0.48, I don't actually see anything at all about using [] syntax to extract vector elements. -- Joseph S. Myers jos...@codesourcery.com
