[CostModel][X86] Improve AVX1/AVX2 v16i32->v16i16/v16i8 truncation costs (PR51972)
Based off worst case btver2 (AVX1) and haswell (AVX2) llvm-mca reports
Loading
Please sign in to comment
Based off worst case btver2 (AVX1) and haswell (AVX2) llvm-mca reports