[PATCH] OpenCL 1.2 allow explicit zero-to-event_t cast

[Tom Stellard]

On Mon, Sep 29, 2014 at 11:16:46PM +0300, Pekka Jääskeläinen wrote:
> Hi,
> 
> On 09/29/2014 07:06 PM, Sahasrabuddhe, Sameer wrote:
> > But if such a cast is already present in a conformance test, I suppose every
> > OpenCL vendor has a frontend (possibly Clang-based) that allows the explicit
> > cast. Can this not be considered a "de facto" refinement to the standard? Or do
> > we strictly need a clarification from Khronos?
> 
> OK. Seems to me the actual culprit is the conformance test that tests
> undefined behavior.
> 
> However, my opinion is that the best way forward would be to commit the
> patch to enable the current conformance test to pass with an upstream clang,
> and point this to the Khronos group so they can either remove the test or
> make it explicitly allowed in the specs. Hard to see why they would
> explicitly forbid it.
>

The conformance test is wrong here, so I think we should push for fixing
the test rather than adding a work-around to clang.

-Tom

> -- 
> --Pekka
> 
> _______________________________________________
> cfe-commits mailing list
> cfe-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits