[X86] SimplifyDemandedBitsForTargetNode - fold MOVMSK(YMM) -> MOVMSK(XMM)
If we don't demand the upper elements of the 256-bit vector, then just perform as a 128-bit vector
Loading
Please register or sign in to comment
If we don't demand the upper elements of the 256-bit vector, then just perform as a 128-bit vector