search for: canevaluatesextd

...narrow the width of lshr. https://reviews.llvm.org/D47113 As Sanjay Patel notes in https://reviews.llvm.org/D47113#1106601, > <cut> but I doubt that we can extend that reasoning to all binops > on all targets. This will fight with transforms in instcombine > (see canEvaluateSExtd / canEvaluateZExtd). But, this is already sufficient to handle the [1]. Thus, as suggested, i'm sending this RFC. What do people think about "CVP should always narrow lshr" approach? Or is that likely to be too bad, and some more concrete approach, like the aggressive-instcombine...