> On Mon, Oct 27, 2014 at 11:05:43PM +0000, Robinson, Paul wrote: > > New patch attached that defines new builtins and uses them in > > lzcntintrin.h. > > Wouldn't the point of the new builtins to *require* the 0 case to be > defined? > > Joerg
Yes, that is what they do. The second parameter to llvm.ct*z specifies "the zero case is undef" so passing false means the zero case is defined. --paulr _______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
