; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
; RUN: opt < %s -mtriple=riscv32 -mattr=+v,+zfh,+zvfh -passes="print<cost-model>" -cost-kind=throughput -intrinsic-cost-strategy=type-based-intrinsic-cost 2>&1 -disable-output | FileCheck %s --check-prefixes=CHECK,ZVE64D
; RUN: opt < %s -mtriple=riscv64 -mattr=+v,+zfh,+zvfh -passes="print<cost-model>" -cost-kind=throughput -intrinsic-cost-strategy=type-based-intrinsic-cost 2>&1 -disable-output | FileCheck %s --check-prefixes=CHECK,ZVE64D
; RUN: opt < %s -mtriple=riscv64 -mattr=+zvl128b,+zve64x -passes="print<cost-model>" -cost-kind=throughput -intrinsic-cost-strategy=type-based-intrinsic-cost 2>&1 -disable-output | FileCheck %s --check-prefixes=CHECK,ZVE64X

define void @sext() {
; CHECK-LABEL: 'sext'
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i8_v2i16 = sext <2 x i8> undef to <2 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i8_v2i32 = sext <2 x i8> undef to <2 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i8_v2i64 = sext <2 x i8> undef to <2 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i16_v2i32 = sext <2 x i16> undef to <2 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i16_v2i64 = sext <2 x i16> undef to <2 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i32_v2i64 = sext <2 x i32> undef to <2 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2i1_v2i8 = sext <2 x i1> undef to <2 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2i1_v2i16 = sext <2 x i1> undef to <2 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2i1_v2i32 = sext <2 x i1> undef to <2 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2i1_v2i64 = sext <2 x i1> undef to <2 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2i8_v2i16 = call <2 x i16> @llvm.vp.sext.v2i16.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2i8_v2i32 = call <2 x i32> @llvm.vp.sext.v2i32.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2i8_v2i64 = call <2 x i64> @llvm.vp.sext.v2i64.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2i16_v2i32 = call <2 x i32> @llvm.vp.sext.v2i32.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2i16_v2i64 = call <2 x i64> @llvm.vp.sext.v2i64.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2i32_v2i64 = call <2 x i64> @llvm.vp.sext.v2i64.v2i32(<2 x i32> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2i1_v2i8 = call <2 x i8> @llvm.vp.sext.v2i8.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2i1_v2i16 = call <2 x i16> @llvm.vp.sext.v2i16.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2i1_v2i32 = call <2 x i32> @llvm.vp.sext.v2i32.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2i1_v2i64 = call <2 x i64> @llvm.vp.sext.v2i64.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4i8_v4i16 = sext <4 x i8> undef to <4 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4i8_v4i32 = sext <4 x i8> undef to <4 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i8_v4i64 = sext <4 x i8> undef to <4 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4i16_v4i32 = sext <4 x i16> undef to <4 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i16_v4i64 = sext <4 x i16> undef to <4 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i32_v4i64 = sext <4 x i32> undef to <4 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i1_v4i8 = sext <4 x i1> undef to <4 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i1_v4i16 = sext <4 x i1> undef to <4 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i1_v4i32 = sext <4 x i1> undef to <4 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v4i1_v4i64 = sext <4 x i1> undef to <4 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v4i8_v4i16 = call <4 x i16> @llvm.vp.sext.v4i16.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v4i8_v4i32 = call <4 x i32> @llvm.vp.sext.v4i32.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4i8_v4i64 = call <4 x i64> @llvm.vp.sext.v4i64.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v4i16_v4i32 = call <4 x i32> @llvm.vp.sext.v4i32.v4i16(<4 x i16> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4i16_v4i64 = call <4 x i64> @llvm.vp.sext.v4i64.v4i16(<4 x i16> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4i32_v4i64 = call <4 x i64> @llvm.vp.sext.v4i64.v4i32(<4 x i32> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4i1_v4i8 = call <4 x i8> @llvm.vp.sext.v4i8.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4i1_v4i16 = call <4 x i16> @llvm.vp.sext.v4i16.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4i1_v4i32 = call <4 x i32> @llvm.vp.sext.v4i32.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v4i1_v4i64 = call <4 x i64> @llvm.vp.sext.v4i64.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8i8_v8i16 = sext <8 x i8> undef to <8 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8i8_v8i32 = sext <8 x i8> undef to <8 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8i8_v8i64 = sext <8 x i8> undef to <8 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8i16_v8i32 = sext <8 x i16> undef to <8 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8i16_v8i64 = sext <8 x i16> undef to <8 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8i32_v8i64 = sext <8 x i32> undef to <8 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8i1_v8i8 = sext <8 x i1> undef to <8 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8i1_v8i16 = sext <8 x i1> undef to <8 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8i1_v8i32 = sext <8 x i1> undef to <8 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v8i1_v8i64 = sext <8 x i1> undef to <8 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v8i8_v8i16 = call <8 x i16> @llvm.vp.sext.v8i16.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8i8_v8i32 = call <8 x i32> @llvm.vp.sext.v8i32.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v8i8_v8i64 = call <8 x i64> @llvm.vp.sext.v8i64.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8i16_v8i32 = call <8 x i32> @llvm.vp.sext.v8i32.v8i16(<8 x i16> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v8i16_v8i64 = call <8 x i64> @llvm.vp.sext.v8i64.v8i16(<8 x i16> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v8i32_v8i64 = call <8 x i64> @llvm.vp.sext.v8i64.v8i32(<8 x i32> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8i1_v8i8 = call <8 x i8> @llvm.vp.sext.v8i8.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8i1_v8i16 = call <8 x i16> @llvm.vp.sext.v8i16.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v8i1_v8i32 = call <8 x i32> @llvm.vp.sext.v8i32.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v8i1_v8i64 = call <8 x i64> @llvm.vp.sext.v8i64.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v16i8_v16i16 = sext <16 x i8> undef to <16 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v16i8_v16i32 = sext <16 x i8> undef to <16 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v16i8_v16i64 = sext <16 x i8> undef to <16 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v16i16_v16i32 = sext <16 x i16> undef to <16 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v16i16_v16i64 = sext <16 x i16> undef to <16 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v16i32_v16i64 = sext <16 x i32> undef to <16 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v16i1_v16i8 = sext <16 x i1> undef to <16 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v16i1_v16i16 = sext <16 x i1> undef to <16 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v16i1_v16i32 = sext <16 x i1> undef to <16 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v16i1_v16i64 = sext <16 x i1> undef to <16 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v16i8_v16i16 = call <16 x i16> @llvm.vp.sext.v16i16.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v16i8_v16i32 = call <16 x i32> @llvm.vp.sext.v16i32.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v16i8_v16i64 = call <16 x i64> @llvm.vp.sext.v16i64.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v16i16_v16i32 = call <16 x i32> @llvm.vp.sext.v16i32.v16i16(<16 x i16> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v16i16_v16i64 = call <16 x i64> @llvm.vp.sext.v16i64.v16i16(<16 x i16> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v16i32_v16i64 = call <16 x i64> @llvm.vp.sext.v16i64.v16i32(<16 x i32> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v16i1_v16i8 = call <16 x i8> @llvm.vp.sext.v16i8.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v16i1_v16i16 = call <16 x i16> @llvm.vp.sext.v16i16.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v16i1_v16i32 = call <16 x i32> @llvm.vp.sext.v16i32.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v16i1_v16i64 = call <16 x i64> @llvm.vp.sext.v16i64.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v32i8_v32i16 = sext <32 x i8> undef to <32 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v32i8_v32i32 = sext <32 x i8> undef to <32 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %v32i8_v32i64 = sext <32 x i8> undef to <32 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v32i16_v32i32 = sext <32 x i16> undef to <32 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %v32i16_v32i64 = sext <32 x i16> undef to <32 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %v32i32_v32i64 = sext <32 x i32> undef to <32 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v32i1_v32i8 = sext <32 x i1> undef to <32 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v32i1_v32i16 = sext <32 x i1> undef to <32 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v32i1_v32i32 = sext <32 x i1> undef to <32 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %v32i1_v32i64 = sext <32 x i1> undef to <32 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v32i8_v32i16 = call <32 x i16> @llvm.vp.sext.v32i16.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v32i8_v32i32 = call <32 x i32> @llvm.vp.sext.v32i32.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_v32i8_v32i64 = call <32 x i64> @llvm.vp.sext.v32i64.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v32i16_v32i32 = call <32 x i32> @llvm.vp.sext.v32i32.v32i16(<32 x i16> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_v32i16_v32i64 = call <32 x i64> @llvm.vp.sext.v32i64.v32i16(<32 x i16> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_v32i32_v32i64 = call <32 x i64> @llvm.vp.sext.v32i64.v32i32(<32 x i32> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v32i1_v32i8 = call <32 x i8> @llvm.vp.sext.v32i8.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v32i1_v32i16 = call <32 x i16> @llvm.vp.sext.v32i16.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v32i1_v32i32 = call <32 x i32> @llvm.vp.sext.v32i32.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_v32i1_v32i64 = call <32 x i64> @llvm.vp.sext.v32i64.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v64i8_v64i16 = sext <64 x i8> undef to <64 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %v64i8_v64i32 = sext <64 x i8> undef to <64 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %v64i8_v64i64 = sext <64 x i8> undef to <64 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %v64i16_v64i32 = sext <64 x i16> undef to <64 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %v64i16_v64i64 = sext <64 x i16> undef to <64 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %v64i32_v64i64 = sext <64 x i32> undef to <64 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v64i1_v64i8 = sext <64 x i1> undef to <64 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v64i1_v64i16 = sext <64 x i1> undef to <64 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %v64i1_v64i32 = sext <64 x i1> undef to <64 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %v64i1_v64i64 = sext <64 x i1> undef to <64 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v64i8_v64i16 = call <64 x i16> @llvm.vp.sext.v64i16.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_v64i8_v64i32 = call <64 x i32> @llvm.vp.sext.v64i32.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %vp_v64i8_v64i64 = call <64 x i64> @llvm.vp.sext.v64i64.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_v64i16_v64i32 = call <64 x i32> @llvm.vp.sext.v64i32.v64i16(<64 x i16> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %vp_v64i16_v64i64 = call <64 x i64> @llvm.vp.sext.v64i64.v64i16(<64 x i16> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_v64i32_v64i64 = call <64 x i64> @llvm.vp.sext.v64i64.v64i32(<64 x i32> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v64i1_v64i8 = call <64 x i8> @llvm.vp.sext.v64i8.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v64i1_v64i16 = call <64 x i16> @llvm.vp.sext.v64i16.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_v64i1_v64i32 = call <64 x i32> @llvm.vp.sext.v64i32.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %vp_v64i1_v64i64 = call <64 x i64> @llvm.vp.sext.v64i64.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %v128i8_v128i16 = sext <128 x i8> undef to <128 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %v128i8_v128i32 = sext <128 x i8> undef to <128 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 71 for instruction: %v128i8_v128i64 = sext <128 x i8> undef to <128 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %v128i16_v128i32 = sext <128 x i16> undef to <128 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 70 for instruction: %v128i16_v128i64 = sext <128 x i16> undef to <128 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %v128i32_v128i64 = sext <128 x i32> undef to <128 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v128i1_v128i8 = sext <128 x i1> undef to <128 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %v128i1_v128i16 = sext <128 x i1> undef to <128 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %v128i1_v128i32 = sext <128 x i1> undef to <128 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 79 for instruction: %v128i1_v128i64 = sext <128 x i1> undef to <128 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_v128i8_v128i16 = call <128 x i16> @llvm.vp.sext.v128i16.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %vp_v128i8_v128i32 = call <128 x i32> @llvm.vp.sext.v128i32.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 71 for instruction: %vp_v128i8_v128i64 = call <128 x i64> @llvm.vp.sext.v128i64.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_v128i16_v128i32 = call <128 x i32> @llvm.vp.sext.v128i32.v128i16(<128 x i16> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 70 for instruction: %vp_v128i16_v128i64 = call <128 x i64> @llvm.vp.sext.v128i64.v128i16(<128 x i16> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %vp_v128i32_v128i64 = call <128 x i64> @llvm.vp.sext.v128i64.v128i32(<128 x i32> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v128i1_v128i8 = call <128 x i8> @llvm.vp.sext.v128i8.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_v128i1_v128i16 = call <128 x i16> @llvm.vp.sext.v128i16.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %vp_v128i1_v128i32 = call <128 x i32> @llvm.vp.sext.v128i32.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 79 for instruction: %vp_v128i1_v128i64 = call <128 x i64> @llvm.vp.sext.v128i64.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %v256i8_v256i16 = sext <256 x i8> undef to <256 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 70 for instruction: %v256i8_v256i32 = sext <256 x i8> undef to <256 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 142 for instruction: %v256i8_v256i64 = sext <256 x i8> undef to <256 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %v256i16_v256i32 = sext <256 x i16> undef to <256 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %v256i16_v256i64 = sext <256 x i16> undef to <256 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %v256i32_v256i64 = sext <256 x i32> undef to <256 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %v256i1_v256i8 = sext <256 x i1> undef to <256 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %v256i1_v256i16 = sext <256 x i1> undef to <256 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 79 for instruction: %v256i1_v256i32 = sext <256 x i1> undef to <256 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 151 for instruction: %v256i1_v256i64 = sext <256 x i1> undef to <256 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_v256i8_v256i16 = call <256 x i16> @llvm.vp.sext.v256i16.v256i8(<256 x i8> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 70 for instruction: %vp_v256i8_v256i32 = call <256 x i32> @llvm.vp.sext.v256i32.v256i8(<256 x i8> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 142 for instruction: %vp_v256i8_v256i64 = call <256 x i64> @llvm.vp.sext.v256i64.v256i8(<256 x i8> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %vp_v256i16_v256i32 = call <256 x i32> @llvm.vp.sext.v256i32.v256i16(<256 x i16> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %vp_v256i16_v256i64 = call <256 x i64> @llvm.vp.sext.v256i64.v256i16(<256 x i16> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %vp_v256i32_v256i64 = call <256 x i64> @llvm.vp.sext.v256i64.v256i32(<256 x i32> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_v256i1_v256i8 = call <256 x i8> @llvm.vp.sext.v256i8.v256i1(<256 x i1> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %vp_v256i1_v256i16 = call <256 x i16> @llvm.vp.sext.v256i16.v256i1(<256 x i1> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 79 for instruction: %vp_v256i1_v256i32 = call <256 x i32> @llvm.vp.sext.v256i32.v256i1(<256 x i1> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 151 for instruction: %vp_v256i1_v256i64 = call <256 x i64> @llvm.vp.sext.v256i64.v256i1(<256 x i1> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1i8_nxv1i16 = sext <vscale x 1 x i8> undef to <vscale x 1 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1i8_nxv1i32 = sext <vscale x 1 x i8> undef to <vscale x 1 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1i8_nxv1i64 = sext <vscale x 1 x i8> undef to <vscale x 1 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1i16_nxv1i32 = sext <vscale x 1 x i16> undef to <vscale x 1 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1i16_nxv1i64 = sext <vscale x 1 x i16> undef to <vscale x 1 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1i32_nxv1i64 = sext <vscale x 1 x i32> undef to <vscale x 1 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1i1_nxv1i8 = sext <vscale x 1 x i1> undef to <vscale x 1 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1i1_nxv1i16 = sext <vscale x 1 x i1> undef to <vscale x 1 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1i1_nxv1i32 = sext <vscale x 1 x i1> undef to <vscale x 1 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1i1_nxv1i64 = sext <vscale x 1 x i1> undef to <vscale x 1 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1i8_nxv1i16 = call <vscale x 1 x i16> @llvm.vp.sext.nxv1i16.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1i8_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.sext.nxv1i32.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1i8_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.sext.nxv1i64.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1i16_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.sext.nxv1i32.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1i16_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.sext.nxv1i64.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1i32_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.sext.nxv1i64.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1i1_nxv1i8 = call <vscale x 1 x i8> @llvm.vp.sext.nxv1i8.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1i1_nxv1i16 = call <vscale x 1 x i16> @llvm.vp.sext.nxv1i16.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1i1_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.sext.nxv1i32.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1i1_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.sext.nxv1i64.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv2i8_nxv2i16 = sext <vscale x 2 x i8> undef to <vscale x 2 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv2i8_nxv2i32 = sext <vscale x 2 x i8> undef to <vscale x 2 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2i8_nxv2i64 = sext <vscale x 2 x i8> undef to <vscale x 2 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv2i16_nxv2i32 = sext <vscale x 2 x i16> undef to <vscale x 2 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2i16_nxv2i64 = sext <vscale x 2 x i16> undef to <vscale x 2 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2i32_nxv2i64 = sext <vscale x 2 x i32> undef to <vscale x 2 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2i1_nxv2i8 = sext <vscale x 2 x i1> undef to <vscale x 2 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2i1_nxv2i16 = sext <vscale x 2 x i1> undef to <vscale x 2 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2i1_nxv2i32 = sext <vscale x 2 x i1> undef to <vscale x 2 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv2i1_nxv2i64 = sext <vscale x 2 x i1> undef to <vscale x 2 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv2i8_nxv2i16 = call <vscale x 2 x i16> @llvm.vp.sext.nxv2i16.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv2i8_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.sext.nxv2i32.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2i8_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.sext.nxv2i64.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv2i16_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.sext.nxv2i32.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2i16_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.sext.nxv2i64.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2i32_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.sext.nxv2i64.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2i1_nxv2i8 = call <vscale x 2 x i8> @llvm.vp.sext.nxv2i8.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2i1_nxv2i16 = call <vscale x 2 x i16> @llvm.vp.sext.nxv2i16.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2i1_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.sext.nxv2i32.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv2i1_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.sext.nxv2i64.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv4i8_nxv4i16 = sext <vscale x 4 x i8> undef to <vscale x 4 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4i8_nxv4i32 = sext <vscale x 4 x i8> undef to <vscale x 4 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv4i8_nxv4i64 = sext <vscale x 4 x i8> undef to <vscale x 4 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4i16_nxv4i32 = sext <vscale x 4 x i16> undef to <vscale x 4 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv4i16_nxv4i64 = sext <vscale x 4 x i16> undef to <vscale x 4 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv4i32_nxv4i64 = sext <vscale x 4 x i32> undef to <vscale x 4 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4i1_nxv4i8 = sext <vscale x 4 x i1> undef to <vscale x 4 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4i1_nxv4i16 = sext <vscale x 4 x i1> undef to <vscale x 4 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv4i1_nxv4i32 = sext <vscale x 4 x i1> undef to <vscale x 4 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv4i1_nxv4i64 = sext <vscale x 4 x i1> undef to <vscale x 4 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv4i8_nxv4i16 = call <vscale x 4 x i16> @llvm.vp.sext.nxv4i16.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4i8_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.sext.nxv4i32.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv4i8_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.sext.nxv4i64.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4i16_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.sext.nxv4i32.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv4i16_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.sext.nxv4i64.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv4i32_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.sext.nxv4i64.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4i1_nxv4i8 = call <vscale x 4 x i8> @llvm.vp.sext.nxv4i8.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4i1_nxv4i16 = call <vscale x 4 x i16> @llvm.vp.sext.nxv4i16.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv4i1_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.sext.nxv4i32.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv4i1_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.sext.nxv4i64.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv8i8_nxv8i16 = sext <vscale x 8 x i8> undef to <vscale x 8 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv8i8_nxv8i32 = sext <vscale x 8 x i8> undef to <vscale x 8 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv8i8_nxv8i64 = sext <vscale x 8 x i8> undef to <vscale x 8 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv8i16_nxv8i32 = sext <vscale x 8 x i16> undef to <vscale x 8 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv8i16_nxv8i64 = sext <vscale x 8 x i16> undef to <vscale x 8 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv8i32_nxv8i64 = sext <vscale x 8 x i32> undef to <vscale x 8 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv8i1_nxv8i8 = sext <vscale x 8 x i1> undef to <vscale x 8 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv8i1_nxv8i16 = sext <vscale x 8 x i1> undef to <vscale x 8 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv8i1_nxv8i32 = sext <vscale x 8 x i1> undef to <vscale x 8 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %nxv8i1_nxv8i64 = sext <vscale x 8 x i1> undef to <vscale x 8 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv8i8_nxv8i16 = call <vscale x 8 x i16> @llvm.vp.sext.nxv8i16.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv8i8_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.sext.nxv8i32.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv8i8_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.sext.nxv8i64.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv8i16_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.sext.nxv8i32.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv8i16_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.sext.nxv8i64.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv8i32_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.sext.nxv8i64.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv8i1_nxv8i8 = call <vscale x 8 x i8> @llvm.vp.sext.nxv8i8.nxv8i1(<vscale x 8 x i1> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv8i1_nxv8i16 = call <vscale x 8 x i16> @llvm.vp.sext.nxv8i16.nxv8i1(<vscale x 8 x i1> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv8i1_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.sext.nxv8i32.nxv8i1(<vscale x 8 x i1> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_nxv8i1_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.sext.nxv8i64.nxv8i1(<vscale x 8 x i1> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv16i8_nxv16i16 = sext <vscale x 16 x i8> undef to <vscale x 16 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv16i8_nxv16i32 = sext <vscale x 16 x i8> undef to <vscale x 16 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %nxv16i8_nxv16i64 = sext <vscale x 16 x i8> undef to <vscale x 16 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv16i16_nxv16i32 = sext <vscale x 16 x i16> undef to <vscale x 16 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %nxv16i16_nxv16i64 = sext <vscale x 16 x i16> undef to <vscale x 16 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %nxv16i32_nxv16i64 = sext <vscale x 16 x i32> undef to <vscale x 16 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv16i1_nxv16i8 = sext <vscale x 16 x i1> undef to <vscale x 16 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv16i1_nxv16i16 = sext <vscale x 16 x i1> undef to <vscale x 16 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %nxv16i1_nxv16i32 = sext <vscale x 16 x i1> undef to <vscale x 16 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %nxv16i1_nxv16i64 = sext <vscale x 16 x i1> undef to <vscale x 16 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv16i8_nxv16i16 = call <vscale x 16 x i16> @llvm.vp.sext.nxv16i16.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv16i8_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.sext.nxv16i32.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_nxv16i8_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.sext.nxv16i64.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv16i16_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.sext.nxv16i32.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_nxv16i16_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.sext.nxv16i64.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_nxv16i32_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.sext.nxv16i64.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv16i1_nxv16i8 = call <vscale x 16 x i8> @llvm.vp.sext.nxv16i8.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv16i1_nxv16i16 = call <vscale x 16 x i16> @llvm.vp.sext.nxv16i16.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_nxv16i1_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.sext.nxv16i32.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_nxv16i1_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.sext.nxv16i64.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv32i8_nxv32i16 = sext <vscale x 32 x i8> undef to <vscale x 32 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %nxv32i8_nxv32i32 = sext <vscale x 32 x i8> undef to <vscale x 32 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %nxv32i8_nxv32i64 = sext <vscale x 32 x i8> undef to <vscale x 32 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %nxv32i16_nxv32i32 = sext <vscale x 32 x i16> undef to <vscale x 32 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %nxv32i16_nxv32i64 = sext <vscale x 32 x i16> undef to <vscale x 32 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %nxv32i32_nxv32i64 = sext <vscale x 32 x i32> undef to <vscale x 32 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv32i1_nxv32i8 = sext <vscale x 32 x i1> undef to <vscale x 32 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %nxv32i1_nxv32i16 = sext <vscale x 32 x i1> undef to <vscale x 32 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %nxv32i1_nxv32i32 = sext <vscale x 32 x i1> undef to <vscale x 32 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %nxv32i1_nxv32i64 = sext <vscale x 32 x i1> undef to <vscale x 32 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv32i8_nxv32i16 = call <vscale x 32 x i16> @llvm.vp.sext.nxv32i16.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_nxv32i8_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.sext.nxv32i32.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %vp_nxv32i8_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.sext.nxv32i64.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_nxv32i16_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.sext.nxv32i32.nxv32i16(<vscale x 32 x i16> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %vp_nxv32i16_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.sext.nxv32i64.nxv32i16(<vscale x 32 x i16> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_nxv32i32_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.sext.nxv32i64.nxv32i32(<vscale x 32 x i32> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv32i1_nxv32i8 = call <vscale x 32 x i8> @llvm.vp.sext.nxv32i8.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_nxv32i1_nxv32i16 = call <vscale x 32 x i16> @llvm.vp.sext.nxv32i16.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_nxv32i1_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.sext.nxv32i32.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %vp_nxv32i1_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.sext.nxv32i64.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %nxv64i8_nxv64i16 = sext <vscale x 64 x i8> undef to <vscale x 64 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %nxv64i8_nxv64i32 = sext <vscale x 64 x i8> undef to <vscale x 64 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 71 for instruction: %nxv64i8_nxv64i64 = sext <vscale x 64 x i8> undef to <vscale x 64 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %nxv64i16_nxv64i32 = sext <vscale x 64 x i16> undef to <vscale x 64 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 70 for instruction: %nxv64i16_nxv64i64 = sext <vscale x 64 x i16> undef to <vscale x 64 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %nxv64i32_nxv64i64 = sext <vscale x 64 x i32> undef to <vscale x 64 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %nxv64i1_nxv64i8 = sext <vscale x 64 x i1> undef to <vscale x 64 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %nxv64i1_nxv64i16 = sext <vscale x 64 x i1> undef to <vscale x 64 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %nxv64i1_nxv64i32 = sext <vscale x 64 x i1> undef to <vscale x 64 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 79 for instruction: %nxv64i1_nxv64i64 = sext <vscale x 64 x i1> undef to <vscale x 64 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_nxv64i8_nxv64i16 = call <vscale x 64 x i16> @llvm.vp.sext.nxv64i16.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %vp_nxv64i8_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.sext.nxv64i32.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 71 for instruction: %vp_nxv64i8_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.sext.nxv64i64.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_nxv64i16_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.sext.nxv64i32.nxv64i16(<vscale x 64 x i16> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 70 for instruction: %vp_nxv64i16_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.sext.nxv64i64.nxv64i16(<vscale x 64 x i16> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %vp_nxv64i32_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.sext.nxv64i64.nxv64i32(<vscale x 64 x i32> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_nxv64i1_nxv64i8 = call <vscale x 64 x i8> @llvm.vp.sext.nxv64i8.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_nxv64i1_nxv64i16 = call <vscale x 64 x i16> @llvm.vp.sext.nxv64i16.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %vp_nxv64i1_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.sext.nxv64i32.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 79 for instruction: %vp_nxv64i1_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.sext.nxv64i64.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %nxv128i8_nxv128i16 = sext <vscale x 128 x i8> undef to <vscale x 128 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 70 for instruction: %nxv128i8_nxv128i32 = sext <vscale x 128 x i8> undef to <vscale x 128 x i32>
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %nxv128i8_nxv128i128 = sext <vscale x 128 x i8> undef to <vscale x 128 x i128>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %nxv128i16_nxv128i32 = sext <vscale x 128 x i16> undef to <vscale x 128 x i32>
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %nxv128i16_nxv128i128 = sext <vscale x 128 x i16> undef to <vscale x 128 x i128>
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %nxv128i32_nxv128i128 = sext <vscale x 128 x i32> undef to <vscale x 128 x i128>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %nxv128i1_nxv128i8 = sext <vscale x 128 x i1> undef to <vscale x 128 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %nxv128i1_nxv128i16 = sext <vscale x 128 x i1> undef to <vscale x 128 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 79 for instruction: %nxv128i1_nxv128i32 = sext <vscale x 128 x i1> undef to <vscale x 128 x i32>
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %nxv128i1_nxv128i128 = sext <vscale x 128 x i1> undef to <vscale x 128 x i128>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_nxv128i8_nxv128i16 = call <vscale x 128 x i16> @llvm.vp.sext.nxv128i16.nxv128i8(<vscale x 128 x i8> undef, <vscale x 128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 70 for instruction: %vp_nxv128i8_nxv128i32 = call <vscale x 128 x i32> @llvm.vp.sext.nxv128i32.nxv128i8(<vscale x 128 x i8> undef, <vscale x 128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv128i8_nxv128i128 = call <vscale x 128 x i128> @llvm.vp.sext.nxv128i128.nxv128i8(<vscale x 128 x i8> undef, <vscale x 128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %vp_nxv128i16_nxv128i32 = call <vscale x 128 x i32> @llvm.vp.sext.nxv128i32.nxv128i16(<vscale x 128 x i16> undef, <vscale x 128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv128i16_nxv128i128 = call <vscale x 128 x i128> @llvm.vp.sext.nxv128i128.nxv128i16(<vscale x 128 x i16> undef, <vscale x 128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv128i32_nxv128i128 = call <vscale x 128 x i128> @llvm.vp.sext.nxv128i128.nxv128i32(<vscale x 128 x i32> undef, <vscale x 128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_nxv128i1_nxv128i8 = call <vscale x 128 x i8> @llvm.vp.sext.nxv128i8.nxv128i1(<vscale x 128 x i1> undef, <vscale x 128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %vp_nxv128i1_nxv128i16 = call <vscale x 128 x i16> @llvm.vp.sext.nxv128i16.nxv128i1(<vscale x 128 x i1> undef, <vscale x 128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 79 for instruction: %vp_nxv128i1_nxv128i32 = call <vscale x 128 x i32> @llvm.vp.sext.nxv128i32.nxv128i1(<vscale x 128 x i1> undef, <vscale x 128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv128i1_nxv128i128 = call <vscale x 128 x i128> @llvm.vp.sext.nxv128i128.nxv128i1(<vscale x 128 x i1> undef, <vscale x 128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  %v2i8_v2i16 = sext <2 x i8> undef to <2 x i16>
  %v2i8_v2i32 = sext <2 x i8> undef to <2 x i32>
  %v2i8_v2i64 = sext <2 x i8> undef to <2 x i64>
  %v2i16_v2i32 = sext <2 x i16> undef to <2 x i32>
  %v2i16_v2i64 = sext <2 x i16> undef to <2 x i64>
  %v2i32_v2i64 = sext <2 x i32> undef to <2 x i64>
  %v2i1_v2i8 = sext <2 x i1> undef to <2 x i8>
  %v2i1_v2i16 = sext <2 x i1> undef to <2 x i16>
  %v2i1_v2i32 = sext <2 x i1> undef to <2 x i32>
  %v2i1_v2i64 = sext <2 x i1> undef to <2 x i64>

  %vp_v2i8_v2i16 = call <2 x i16> @llvm.vp.sext.v2i16.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
  %vp_v2i8_v2i32 = call <2 x i32> @llvm.vp.sext.v2i32.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
  %vp_v2i8_v2i64 = call <2 x i64> @llvm.vp.sext.v2i64.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
  %vp_v2i16_v2i32 = call <2 x i32> @llvm.vp.sext.v2i32.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
  %vp_v2i16_v2i64 = call <2 x i64> @llvm.vp.sext.v2i64.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
  %vp_v2i32_v2i64 = call <2 x i64> @llvm.vp.sext.v2i64.v2i32(<2 x i32> undef, <2 x i1> undef, i32 undef)
  %vp_v2i1_v2i8 = call <2 x i8> @llvm.vp.sext.v2i8.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)
  %vp_v2i1_v2i16 = call <2 x i16> @llvm.vp.sext.v2i16.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)
  %vp_v2i1_v2i32 = call <2 x i32> @llvm.vp.sext.v2i32.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)
  %vp_v2i1_v2i64 = call <2 x i64> @llvm.vp.sext.v2i64.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)

  %v4i8_v4i16 = sext <4 x i8> undef to <4 x i16>
  %v4i8_v4i32 = sext <4 x i8> undef to <4 x i32>
  %v4i8_v4i64 = sext <4 x i8> undef to <4 x i64>
  %v4i16_v4i32 = sext <4 x i16> undef to <4 x i32>
  %v4i16_v4i64 = sext <4 x i16> undef to <4 x i64>
  %v4i32_v4i64 = sext <4 x i32> undef to <4 x i64>
  %v4i1_v4i8 = sext <4 x i1> undef to <4 x i8>
  %v4i1_v4i16 = sext <4 x i1> undef to <4 x i16>
  %v4i1_v4i32 = sext <4 x i1> undef to <4 x i32>
  %v4i1_v4i64 = sext <4 x i1> undef to <4 x i64>

  %vp_v4i8_v4i16 = call <4 x i16> @llvm.vp.sext.v4i16.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
  %vp_v4i8_v4i32 = call <4 x i32> @llvm.vp.sext.v4i32.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
  %vp_v4i8_v4i64 = call <4 x i64> @llvm.vp.sext.v4i64.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
  %vp_v4i16_v4i32 = call <4 x i32> @llvm.vp.sext.v4i32.v4i16(<4 x i16> undef, <4 x i1> undef, i32 undef)
  %vp_v4i16_v4i64 = call <4 x i64> @llvm.vp.sext.v4i64.v4i16(<4 x i16> undef, <4 x i1> undef, i32 undef)
  %vp_v4i32_v4i64 = call <4 x i64> @llvm.vp.sext.v4i64.v4i32(<4 x i32> undef, <4 x i1> undef, i32 undef)
  %vp_v4i1_v4i8 = call <4 x i8> @llvm.vp.sext.v4i8.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)
  %vp_v4i1_v4i16 = call <4 x i16> @llvm.vp.sext.v4i16.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)
  %vp_v4i1_v4i32 = call <4 x i32> @llvm.vp.sext.v4i32.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)
  %vp_v4i1_v4i64 = call <4 x i64> @llvm.vp.sext.v4i64.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)

  %v8i8_v8i16 = sext <8 x i8> undef to <8 x i16>
  %v8i8_v8i32 = sext <8 x i8> undef to <8 x i32>
  %v8i8_v8i64 = sext <8 x i8> undef to <8 x i64>
  %v8i16_v8i32 = sext <8 x i16> undef to <8 x i32>
  %v8i16_v8i64 = sext <8 x i16> undef to <8 x i64>
  %v8i32_v8i64 = sext <8 x i32> undef to <8 x i64>
  %v8i1_v8i8 = sext <8 x i1> undef to <8 x i8>
  %v8i1_v8i16 = sext <8 x i1> undef to <8 x i16>
  %v8i1_v8i32 = sext <8 x i1> undef to <8 x i32>
  %v8i1_v8i64 = sext <8 x i1> undef to <8 x i64>

  %vp_v8i8_v8i16 = call <8 x i16> @llvm.vp.sext.v8i16.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
  %vp_v8i8_v8i32 = call <8 x i32> @llvm.vp.sext.v8i32.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
  %vp_v8i8_v8i64 = call <8 x i64> @llvm.vp.sext.v8i64.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
  %vp_v8i16_v8i32 = call <8 x i32> @llvm.vp.sext.v8i32.v8i16(<8 x i16> undef, <8 x i1> undef, i32 undef)
  %vp_v8i16_v8i64 = call <8 x i64> @llvm.vp.sext.v8i64.v8i16(<8 x i16> undef, <8 x i1> undef, i32 undef)
  %vp_v8i32_v8i64 = call <8 x i64> @llvm.vp.sext.v8i64.v8i32(<8 x i32> undef, <8 x i1> undef, i32 undef)
  %vp_v8i1_v8i8 = call <8 x i8> @llvm.vp.sext.v8i8.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)
  %vp_v8i1_v8i16 = call <8 x i16> @llvm.vp.sext.v8i16.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)
  %vp_v8i1_v8i32 = call <8 x i32> @llvm.vp.sext.v8i32.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)
  %vp_v8i1_v8i64 = call <8 x i64> @llvm.vp.sext.v8i64.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)

  %v16i8_v16i16 = sext <16 x i8> undef to <16 x i16>
  %v16i8_v16i32 = sext <16 x i8> undef to <16 x i32>
  %v16i8_v16i64 = sext <16 x i8> undef to <16 x i64>
  %v16i16_v16i32 = sext <16 x i16> undef to <16 x i32>
  %v16i16_v16i64 = sext <16 x i16> undef to <16 x i64>
  %v16i32_v16i64 = sext <16 x i32> undef to <16 x i64>
  %v16i1_v16i8 = sext <16 x i1> undef to <16 x i8>
  %v16i1_v16i16 = sext <16 x i1> undef to <16 x i16>
  %v16i1_v16i32 = sext <16 x i1> undef to <16 x i32>
  %v16i1_v16i64 = sext <16 x i1> undef to <16 x i64>

  %vp_v16i8_v16i16 = call <16 x i16> @llvm.vp.sext.v16i16.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
  %vp_v16i8_v16i32 = call <16 x i32> @llvm.vp.sext.v16i32.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
  %vp_v16i8_v16i64 = call <16 x i64> @llvm.vp.sext.v16i64.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
  %vp_v16i16_v16i32 = call <16 x i32> @llvm.vp.sext.v16i32.v16i16(<16 x i16> undef, <16 x i1> undef, i32 undef)
  %vp_v16i16_v16i64 = call <16 x i64> @llvm.vp.sext.v16i64.v16i16(<16 x i16> undef, <16 x i1> undef, i32 undef)
  %vp_v16i32_v16i64 = call <16 x i64> @llvm.vp.sext.v16i64.v16i32(<16 x i32> undef, <16 x i1> undef, i32 undef)
  %vp_v16i1_v16i8 = call <16 x i8> @llvm.vp.sext.v16i8.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)
  %vp_v16i1_v16i16 = call <16 x i16> @llvm.vp.sext.v16i16.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)
  %vp_v16i1_v16i32 = call <16 x i32> @llvm.vp.sext.v16i32.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)
  %vp_v16i1_v16i64 = call <16 x i64> @llvm.vp.sext.v16i64.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)

  %v32i8_v32i16 = sext <32 x i8> undef to <32 x i16>
  %v32i8_v32i32 = sext <32 x i8> undef to <32 x i32>
  %v32i8_v32i64 = sext <32 x i8> undef to <32 x i64>
  %v32i16_v32i32 = sext <32 x i16> undef to <32 x i32>
  %v32i16_v32i64 = sext <32 x i16> undef to <32 x i64>
  %v32i32_v32i64 = sext <32 x i32> undef to <32 x i64>
  %v32i1_v32i8 = sext <32 x i1> undef to <32 x i8>
  %v32i1_v32i16 = sext <32 x i1> undef to <32 x i16>
  %v32i1_v32i32 = sext <32 x i1> undef to <32 x i32>
  %v32i1_v32i64 = sext <32 x i1> undef to <32 x i64>

  %vp_v32i8_v32i16 = call <32 x i16> @llvm.vp.sext.v32i16.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
  %vp_v32i8_v32i32 = call <32 x i32> @llvm.vp.sext.v32i32.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
  %vp_v32i8_v32i64 = call <32 x i64> @llvm.vp.sext.v32i64.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
  %vp_v32i16_v32i32 = call <32 x i32> @llvm.vp.sext.v32i32.v32i16(<32 x i16> undef, <32 x i1> undef, i32 undef)
  %vp_v32i16_v32i64 = call <32 x i64> @llvm.vp.sext.v32i64.v32i16(<32 x i16> undef, <32 x i1> undef, i32 undef)
  %vp_v32i32_v32i64 = call <32 x i64> @llvm.vp.sext.v32i64.v32i32(<32 x i32> undef, <32 x i1> undef, i32 undef)
  %vp_v32i1_v32i8 = call <32 x i8> @llvm.vp.sext.v32i8.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)
  %vp_v32i1_v32i16 = call <32 x i16> @llvm.vp.sext.v32i16.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)
  %vp_v32i1_v32i32 = call <32 x i32> @llvm.vp.sext.v32i32.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)
  %vp_v32i1_v32i64 = call <32 x i64> @llvm.vp.sext.v32i64.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)

  %v64i8_v64i16 = sext <64 x i8> undef to <64 x i16>
  %v64i8_v64i32 = sext <64 x i8> undef to <64 x i32>
  %v64i8_v64i64 = sext <64 x i8> undef to <64 x i64>
  %v64i16_v64i32 = sext <64 x i16> undef to <64 x i32>
  %v64i16_v64i64 = sext <64 x i16> undef to <64 x i64>
  %v64i32_v64i64 = sext <64 x i32> undef to <64 x i64>
  %v64i1_v64i8 = sext <64 x i1> undef to <64 x i8>
  %v64i1_v64i16 = sext <64 x i1> undef to <64 x i16>
  %v64i1_v64i32 = sext <64 x i1> undef to <64 x i32>
  %v64i1_v64i64 = sext <64 x i1> undef to <64 x i64>

  %vp_v64i8_v64i16 = call <64 x i16> @llvm.vp.sext.v64i16.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
  %vp_v64i8_v64i32 = call <64 x i32> @llvm.vp.sext.v64i32.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
  %vp_v64i8_v64i64 = call <64 x i64> @llvm.vp.sext.v64i64.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
  %vp_v64i16_v64i32 = call <64 x i32> @llvm.vp.sext.v64i32.v64i16(<64 x i16> undef, <64 x i1> undef, i32 undef)
  %vp_v64i16_v64i64 = call <64 x i64> @llvm.vp.sext.v64i64.v64i16(<64 x i16> undef, <64 x i1> undef, i32 undef)
  %vp_v64i32_v64i64 = call <64 x i64> @llvm.vp.sext.v64i64.v64i32(<64 x i32> undef, <64 x i1> undef, i32 undef)
  %vp_v64i1_v64i8 = call <64 x i8> @llvm.vp.sext.v64i8.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)
  %vp_v64i1_v64i16 = call <64 x i16> @llvm.vp.sext.v64i16.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)
  %vp_v64i1_v64i32 = call <64 x i32> @llvm.vp.sext.v64i32.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)
  %vp_v64i1_v64i64 = call <64 x i64> @llvm.vp.sext.v64i64.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)

  %v128i8_v128i16 = sext <128 x i8> undef to <128 x i16>
  %v128i8_v128i32 = sext <128 x i8> undef to <128 x i32>
  %v128i8_v128i64 = sext <128 x i8> undef to <128 x i64>
  %v128i16_v128i32 = sext <128 x i16> undef to <128 x i32>
  %v128i16_v128i64 = sext <128 x i16> undef to <128 x i64>
  %v128i32_v128i64 = sext <128 x i32> undef to <128 x i64>
  %v128i1_v128i8 = sext <128 x i1> undef to <128 x i8>
  %v128i1_v128i16 = sext <128 x i1> undef to <128 x i16>
  %v128i1_v128i32 = sext <128 x i1> undef to <128 x i32>
  %v128i1_v128i64 = sext <128 x i1> undef to <128 x i64>

  %vp_v128i8_v128i16 = call <128 x i16> @llvm.vp.sext.v128i16.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
  %vp_v128i8_v128i32 = call <128 x i32> @llvm.vp.sext.v128i32.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
  %vp_v128i8_v128i64 = call <128 x i64> @llvm.vp.sext.v128i64.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
  %vp_v128i16_v128i32 = call <128 x i32> @llvm.vp.sext.v128i32.v128i16(<128 x i16> undef, <128 x i1> undef, i32 undef)
  %vp_v128i16_v128i64 = call <128 x i64> @llvm.vp.sext.v128i64.v128i16(<128 x i16> undef, <128 x i1> undef, i32 undef)
  %vp_v128i32_v128i64 = call <128 x i64> @llvm.vp.sext.v128i64.v128i32(<128 x i32> undef, <128 x i1> undef, i32 undef)
  %vp_v128i1_v128i8 = call <128 x i8> @llvm.vp.sext.v128i8.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)
  %vp_v128i1_v128i16 = call <128 x i16> @llvm.vp.sext.v128i16.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)
  %vp_v128i1_v128i32 = call <128 x i32> @llvm.vp.sext.v128i32.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)
  %vp_v128i1_v128i64 = call <128 x i64> @llvm.vp.sext.v128i64.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)

  %v256i8_v256i16 = sext <256 x i8> undef to <256 x i16>
  %v256i8_v256i32 = sext <256 x i8> undef to <256 x i32>
  %v256i8_v256i64 = sext <256 x i8> undef to <256 x i64>
  %v256i16_v256i32 = sext <256 x i16> undef to <256 x i32>
  %v256i16_v256i64 = sext <256 x i16> undef to <256 x i64>
  %v256i32_v256i64 = sext <256 x i32> undef to <256 x i64>
  %v256i1_v256i8 = sext <256 x i1> undef to <256 x i8>
  %v256i1_v256i16 = sext <256 x i1> undef to <256 x i16>
  %v256i1_v256i32 = sext <256 x i1> undef to <256 x i32>
  %v256i1_v256i64 = sext <256 x i1> undef to <256 x i64>

  %vp_v256i8_v256i16 = call <256 x i16> @llvm.vp.sext.v256i16.v256i8(<256 x i8> undef, <256 x i1> undef, i32 undef)
  %vp_v256i8_v256i32 = call <256 x i32> @llvm.vp.sext.v256i32.v256i8(<256 x i8> undef, <256 x i1> undef, i32 undef)
  %vp_v256i8_v256i64 = call <256 x i64> @llvm.vp.sext.v256i64.v256i8(<256 x i8> undef, <256 x i1> undef, i32 undef)
  %vp_v256i16_v256i32 = call <256 x i32> @llvm.vp.sext.v256i32.v256i16(<256 x i16> undef, <256 x i1> undef, i32 undef)
  %vp_v256i16_v256i64 = call <256 x i64> @llvm.vp.sext.v256i64.v256i16(<256 x i16> undef, <256 x i1> undef, i32 undef)
  %vp_v256i32_v256i64 = call <256 x i64> @llvm.vp.sext.v256i64.v256i32(<256 x i32> undef, <256 x i1> undef, i32 undef)
  %vp_v256i1_v256i8 = call <256 x i8> @llvm.vp.sext.v256i8.v256i1(<256 x i1> undef, <256 x i1> undef, i32 undef)
  %vp_v256i1_v256i16 = call <256 x i16> @llvm.vp.sext.v256i16.v256i1(<256 x i1> undef, <256 x i1> undef, i32 undef)
  %vp_v256i1_v256i32 = call <256 x i32> @llvm.vp.sext.v256i32.v256i1(<256 x i1> undef, <256 x i1> undef, i32 undef)
  %vp_v256i1_v256i64 = call <256 x i64> @llvm.vp.sext.v256i64.v256i1(<256 x i1> undef, <256 x i1> undef, i32 undef)

  %nxv1i8_nxv1i16 = sext <vscale x 1 x i8> undef to <vscale x 1 x i16>
  %nxv1i8_nxv1i32 = sext <vscale x 1 x i8> undef to <vscale x 1 x i32>
  %nxv1i8_nxv1i64 = sext <vscale x 1 x i8> undef to <vscale x 1 x i64>
  %nxv1i16_nxv1i32 = sext <vscale x 1 x i16> undef to <vscale x 1 x i32>
  %nxv1i16_nxv1i64 = sext <vscale x 1 x i16> undef to <vscale x 1 x i64>
  %nxv1i32_nxv1i64 = sext <vscale x 1 x i32> undef to <vscale x 1 x i64>
  %nxv1i1_nxv1i8 = sext <vscale x 1 x i1> undef to <vscale x 1 x i8>
  %nxv1i1_nxv1i16 = sext <vscale x 1 x i1> undef to <vscale x 1 x i16>
  %nxv1i1_nxv1i32 = sext <vscale x 1 x i1> undef to <vscale x 1 x i32>
  %nxv1i1_nxv1i64 = sext <vscale x 1 x i1> undef to <vscale x 1 x i64>

  %vp_nxv1i8_nxv1i16 = call <vscale x 1 x i16> @llvm.vp.sext.nxv1i16.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i8_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.sext.nxv1i32.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i8_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.sext.nxv1i64.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i16_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.sext.nxv1i32.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i16_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.sext.nxv1i64.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i32_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.sext.nxv1i64.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i1_nxv1i8 = call <vscale x 1 x i8> @llvm.vp.sext.nxv1i8.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i1_nxv1i16 = call <vscale x 1 x i16> @llvm.vp.sext.nxv1i16.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i1_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.sext.nxv1i32.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i1_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.sext.nxv1i64.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)

  %nxv2i8_nxv2i16 = sext <vscale x 2 x i8> undef to <vscale x 2 x i16>
  %nxv2i8_nxv2i32 = sext <vscale x 2 x i8> undef to <vscale x 2 x i32>
  %nxv2i8_nxv2i64 = sext <vscale x 2 x i8> undef to <vscale x 2 x i64>
  %nxv2i16_nxv2i32 = sext <vscale x 2 x i16> undef to <vscale x 2 x i32>
  %nxv2i16_nxv2i64 = sext <vscale x 2 x i16> undef to <vscale x 2 x i64>
  %nxv2i32_nxv2i64 = sext <vscale x 2 x i32> undef to <vscale x 2 x i64>
  %nxv2i1_nxv2i8 = sext <vscale x 2 x i1> undef to <vscale x 2 x i8>
  %nxv2i1_nxv2i16 = sext <vscale x 2 x i1> undef to <vscale x 2 x i16>
  %nxv2i1_nxv2i32 = sext <vscale x 2 x i1> undef to <vscale x 2 x i32>
  %nxv2i1_nxv2i64 = sext <vscale x 2 x i1> undef to <vscale x 2 x i64>

  %vp_nxv2i8_nxv2i16 = call <vscale x 2 x i16> @llvm.vp.sext.nxv2i16.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i8_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.sext.nxv2i32.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i8_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.sext.nxv2i64.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i16_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.sext.nxv2i32.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i16_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.sext.nxv2i64.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i32_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.sext.nxv2i64.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i1_nxv2i8 = call <vscale x 2 x i8> @llvm.vp.sext.nxv2i8.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i1_nxv2i16 = call <vscale x 2 x i16> @llvm.vp.sext.nxv2i16.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i1_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.sext.nxv2i32.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i1_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.sext.nxv2i64.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)

  %nxv4i8_nxv4i16 = sext <vscale x 4 x i8> undef to <vscale x 4 x i16>
  %nxv4i8_nxv4i32 = sext <vscale x 4 x i8> undef to <vscale x 4 x i32>
  %nxv4i8_nxv4i64 = sext <vscale x 4 x i8> undef to <vscale x 4 x i64>
  %nxv4i16_nxv4i32 = sext <vscale x 4 x i16> undef to <vscale x 4 x i32>
  %nxv4i16_nxv4i64 = sext <vscale x 4 x i16> undef to <vscale x 4 x i64>
  %nxv4i32_nxv4i64 = sext <vscale x 4 x i32> undef to <vscale x 4 x i64>
  %nxv4i1_nxv4i8 = sext <vscale x 4 x i1> undef to <vscale x 4 x i8>
  %nxv4i1_nxv4i16 = sext <vscale x 4 x i1> undef to <vscale x 4 x i16>
  %nxv4i1_nxv4i32 = sext <vscale x 4 x i1> undef to <vscale x 4 x i32>
  %nxv4i1_nxv4i64 = sext <vscale x 4 x i1> undef to <vscale x 4 x i64>

  %vp_nxv4i8_nxv4i16 = call <vscale x 4 x i16> @llvm.vp.sext.nxv4i16.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i8_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.sext.nxv4i32.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i8_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.sext.nxv4i64.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i16_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.sext.nxv4i32.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i16_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.sext.nxv4i64.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i32_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.sext.nxv4i64.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i1_nxv4i8 = call <vscale x 4 x i8> @llvm.vp.sext.nxv4i8.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i1_nxv4i16 = call <vscale x 4 x i16> @llvm.vp.sext.nxv4i16.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i1_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.sext.nxv4i32.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i1_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.sext.nxv4i64.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)

  %nxv8i8_nxv8i16 = sext <vscale x 8 x i8> undef to <vscale x 8 x i16>
  %nxv8i8_nxv8i32 = sext <vscale x 8 x i8> undef to <vscale x 8 x i32>
  %nxv8i8_nxv8i64 = sext <vscale x 8 x i8> undef to <vscale x 8 x i64>
  %nxv8i16_nxv8i32 = sext <vscale x 8 x i16> undef to <vscale x 8 x i32>
  %nxv8i16_nxv8i64 = sext <vscale x 8 x i16> undef to <vscale x 8 x i64>
  %nxv8i32_nxv8i64 = sext <vscale x 8 x i32> undef to <vscale x 8 x i64>
  %nxv8i1_nxv8i8 = sext <vscale x 8 x i1> undef to <vscale x 8 x i8>
  %nxv8i1_nxv8i16 = sext <vscale x 8 x i1> undef to <vscale x 8 x i16>
  %nxv8i1_nxv8i32 = sext <vscale x 8 x i1> undef to <vscale x 8 x i32>
  %nxv8i1_nxv8i64 = sext <vscale x 8 x i1> undef to <vscale x 8 x i64>

  %vp_nxv8i8_nxv8i16 = call <vscale x 8 x i16> @llvm.vp.sext.nxv8i16.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i8_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.sext.nxv8i32.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i8_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.sext.nxv8i64.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i16_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.sext.nxv8i32.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i16_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.sext.nxv8i64.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i32_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.sext.nxv8i64.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i1_nxv8i8 = call <vscale x 8 x i8> @llvm.vp.sext.nxv8i8.nxv8i1(<vscale x 8 x i1> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i1_nxv8i16 = call <vscale x 8 x i16> @llvm.vp.sext.nxv8i16.nxv8i1(<vscale x 8 x i1> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i1_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.sext.nxv8i32.nxv8i1(<vscale x 8 x i1> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i1_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.sext.nxv8i64.nxv8i1(<vscale x 8 x i1> undef, <vscale x 8 x i1> undef, i32 undef)

  %nxv16i8_nxv16i16 = sext <vscale x 16 x i8> undef to <vscale x 16 x i16>
  %nxv16i8_nxv16i32 = sext <vscale x 16 x i8> undef to <vscale x 16 x i32>
  %nxv16i8_nxv16i64 = sext <vscale x 16 x i8> undef to <vscale x 16 x i64>
  %nxv16i16_nxv16i32 = sext <vscale x 16 x i16> undef to <vscale x 16 x i32>
  %nxv16i16_nxv16i64 = sext <vscale x 16 x i16> undef to <vscale x 16 x i64>
  %nxv16i32_nxv16i64 = sext <vscale x 16 x i32> undef to <vscale x 16 x i64>
  %nxv16i1_nxv16i8 = sext <vscale x 16 x i1> undef to <vscale x 16 x i8>
  %nxv16i1_nxv16i16 = sext <vscale x 16 x i1> undef to <vscale x 16 x i16>
  %nxv16i1_nxv16i32 = sext <vscale x 16 x i1> undef to <vscale x 16 x i32>
  %nxv16i1_nxv16i64 = sext <vscale x 16 x i1> undef to <vscale x 16 x i64>

  %vp_nxv16i8_nxv16i16 = call <vscale x 16 x i16> @llvm.vp.sext.nxv16i16.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i8_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.sext.nxv16i32.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i8_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.sext.nxv16i64.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i16_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.sext.nxv16i32.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i16_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.sext.nxv16i64.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i32_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.sext.nxv16i64.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i1_nxv16i8 = call <vscale x 16 x i8> @llvm.vp.sext.nxv16i8.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i1_nxv16i16 = call <vscale x 16 x i16> @llvm.vp.sext.nxv16i16.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i1_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.sext.nxv16i32.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i1_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.sext.nxv16i64.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)

  %nxv32i8_nxv32i16 = sext <vscale x 32 x i8> undef to <vscale x 32 x i16>
  %nxv32i8_nxv32i32 = sext <vscale x 32 x i8> undef to <vscale x 32 x i32>
  %nxv32i8_nxv32i64 = sext <vscale x 32 x i8> undef to <vscale x 32 x i64>
  %nxv32i16_nxv32i32 = sext <vscale x 32 x i16> undef to <vscale x 32 x i32>
  %nxv32i16_nxv32i64 = sext <vscale x 32 x i16> undef to <vscale x 32 x i64>
  %nxv32i32_nxv32i64 = sext <vscale x 32 x i32> undef to <vscale x 32 x i64>
  %nxv32i1_nxv32i8 = sext <vscale x 32 x i1> undef to <vscale x 32 x i8>
  %nxv32i1_nxv32i16 = sext <vscale x 32 x i1> undef to <vscale x 32 x i16>
  %nxv32i1_nxv32i32 = sext <vscale x 32 x i1> undef to <vscale x 32 x i32>
  %nxv32i1_nxv32i64 = sext <vscale x 32 x i1> undef to <vscale x 32 x i64>

  %vp_nxv32i8_nxv32i16 = call <vscale x 32 x i16> @llvm.vp.sext.nxv32i16.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i8_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.sext.nxv32i32.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i8_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.sext.nxv32i64.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i16_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.sext.nxv32i32.nxv32i16(<vscale x 32 x i16> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i16_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.sext.nxv32i64.nxv32i16(<vscale x 32 x i16> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i32_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.sext.nxv32i64.nxv32i32(<vscale x 32 x i32> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i1_nxv32i8 = call <vscale x 32 x i8> @llvm.vp.sext.nxv32i8.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i1_nxv32i16 = call <vscale x 32 x i16> @llvm.vp.sext.nxv32i16.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i1_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.sext.nxv32i32.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i1_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.sext.nxv32i64.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)

  %nxv64i8_nxv64i16 = sext <vscale x 64 x i8> undef to <vscale x 64 x i16>
  %nxv64i8_nxv64i32 = sext <vscale x 64 x i8> undef to <vscale x 64 x i32>
  %nxv64i8_nxv64i64 = sext <vscale x 64 x i8> undef to <vscale x 64 x i64>
  %nxv64i16_nxv64i32 = sext <vscale x 64 x i16> undef to <vscale x 64 x i32>
  %nxv64i16_nxv64i64 = sext <vscale x 64 x i16> undef to <vscale x 64 x i64>
  %nxv64i32_nxv64i64 = sext <vscale x 64 x i32> undef to <vscale x 64 x i64>
  %nxv64i1_nxv64i8 = sext <vscale x 64 x i1> undef to <vscale x 64 x i8>
  %nxv64i1_nxv64i16 = sext <vscale x 64 x i1> undef to <vscale x 64 x i16>
  %nxv64i1_nxv64i32 = sext <vscale x 64 x i1> undef to <vscale x 64 x i32>
  %nxv64i1_nxv64i64 = sext <vscale x 64 x i1> undef to <vscale x 64 x i64>

  %vp_nxv64i8_nxv64i16 = call <vscale x 64 x i16> @llvm.vp.sext.nxv64i16.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i8_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.sext.nxv64i32.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i8_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.sext.nxv64i64.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i16_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.sext.nxv64i32.nxv64i16(<vscale x 64 x i16> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i16_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.sext.nxv64i64.nxv64i16(<vscale x 64 x i16> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i32_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.sext.nxv64i64.nxv64i32(<vscale x 64 x i32> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i1_nxv64i8 = call <vscale x 64 x i8> @llvm.vp.sext.nxv64i8.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i1_nxv64i16 = call <vscale x 64 x i16> @llvm.vp.sext.nxv64i16.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i1_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.sext.nxv64i32.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i1_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.sext.nxv64i64.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)

  %nxv128i8_nxv128i16 = sext <vscale x 128 x i8> undef to <vscale x 128 x i16>
  %nxv128i8_nxv128i32 = sext <vscale x 128 x i8> undef to <vscale x 128 x i32>
  %nxv128i8_nxv128i128 = sext <vscale x 128 x i8> undef to <vscale x 128 x i128>
  %nxv128i16_nxv128i32 = sext <vscale x 128 x i16> undef to <vscale x 128 x i32>
  %nxv128i16_nxv128i128 = sext <vscale x 128 x i16> undef to <vscale x 128 x i128>
  %nxv128i32_nxv128i128 = sext <vscale x 128 x i32> undef to <vscale x 128 x i128>
  %nxv128i1_nxv128i8 = sext <vscale x 128 x i1> undef to <vscale x 128 x i8>
  %nxv128i1_nxv128i16 = sext <vscale x 128 x i1> undef to <vscale x 128 x i16>
  %nxv128i1_nxv128i32 = sext <vscale x 128 x i1> undef to <vscale x 128 x i32>
  %nxv128i1_nxv128i128 = sext <vscale x 128 x i1> undef to <vscale x 128 x i128>

  %vp_nxv128i8_nxv128i16 = call <vscale x 128 x i16> @llvm.vp.sext.nxv128i16.nxv128i8(<vscale x 128 x i8> undef, <vscale x 128 x i1> undef, i32 undef)
  %vp_nxv128i8_nxv128i32 = call <vscale x 128 x i32> @llvm.vp.sext.nxv128i32.nxv128i8(<vscale x 128 x i8> undef, <vscale x 128 x i1> undef, i32 undef)
  %vp_nxv128i8_nxv128i128 = call <vscale x 128 x i128> @llvm.vp.sext.nxv128i128.nxv128i8(<vscale x 128 x i8> undef, <vscale x 128 x i1> undef, i32 undef)
  %vp_nxv128i16_nxv128i32 = call <vscale x 128 x i32> @llvm.vp.sext.nxv128i32.nxv128i16(<vscale x 128 x i16> undef, <vscale x 128 x i1> undef, i32 undef)
  %vp_nxv128i16_nxv128i128 = call <vscale x 128 x i128> @llvm.vp.sext.nxv128i128.nxv128i16(<vscale x 128 x i16> undef, <vscale x 128 x i1> undef, i32 undef)
  %vp_nxv128i32_nxv128i128 = call <vscale x 128 x i128> @llvm.vp.sext.nxv128i128.nxv128i32(<vscale x 128 x i32> undef, <vscale x 128 x i1> undef, i32 undef)
  %vp_nxv128i1_nxv128i8 = call <vscale x 128 x i8> @llvm.vp.sext.nxv128i8.nxv128i1(<vscale x 128 x i1> undef, <vscale x 128 x i1> undef, i32 undef)
  %vp_nxv128i1_nxv128i16 = call <vscale x 128 x i16> @llvm.vp.sext.nxv128i16.nxv128i1(<vscale x 128 x i1> undef, <vscale x 128 x i1> undef, i32 undef)
  %vp_nxv128i1_nxv128i32 = call <vscale x 128 x i32> @llvm.vp.sext.nxv128i32.nxv128i1(<vscale x 128 x i1> undef, <vscale x 128 x i1> undef, i32 undef)
  %vp_nxv128i1_nxv128i128 = call <vscale x 128 x i128> @llvm.vp.sext.nxv128i128.nxv128i1(<vscale x 128 x i1> undef, <vscale x 128 x i1> undef, i32 undef)

  ret void
}

define void @zext() {
; CHECK-LABEL: 'zext'
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i8_v2i16 = zext <2 x i8> undef to <2 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i8_v2i32 = zext <2 x i8> undef to <2 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i8_v2i64 = zext <2 x i8> undef to <2 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i16_v2i32 = zext <2 x i16> undef to <2 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i16_v2i64 = zext <2 x i16> undef to <2 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i32_v2i64 = zext <2 x i32> undef to <2 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2i1_v2i8 = zext <2 x i1> undef to <2 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2i1_v2i16 = zext <2 x i1> undef to <2 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2i1_v2i32 = zext <2 x i1> undef to <2 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2i1_v2i64 = zext <2 x i1> undef to <2 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2i8_v2i16 = call <2 x i16> @llvm.vp.zext.v2i16.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2i8_v2i32 = call <2 x i32> @llvm.vp.zext.v2i32.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2i8_v2i64 = call <2 x i64> @llvm.vp.zext.v2i64.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2i16_v2i32 = call <2 x i32> @llvm.vp.zext.v2i32.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2i16_v2i64 = call <2 x i64> @llvm.vp.zext.v2i64.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2i32_v2i64 = call <2 x i64> @llvm.vp.zext.v2i64.v2i32(<2 x i32> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2i1_v2i8 = call <2 x i8> @llvm.vp.zext.v2i8.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2i1_v2i16 = call <2 x i16> @llvm.vp.zext.v2i16.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2i1_v2i32 = call <2 x i32> @llvm.vp.zext.v2i32.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2i1_v2i64 = call <2 x i64> @llvm.vp.zext.v2i64.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4i8_v4i16 = zext <4 x i8> undef to <4 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4i8_v4i32 = zext <4 x i8> undef to <4 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i8_v4i64 = zext <4 x i8> undef to <4 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4i16_v4i32 = zext <4 x i16> undef to <4 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i16_v4i64 = zext <4 x i16> undef to <4 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i32_v4i64 = zext <4 x i32> undef to <4 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i1_v4i8 = zext <4 x i1> undef to <4 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i1_v4i16 = zext <4 x i1> undef to <4 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i1_v4i32 = zext <4 x i1> undef to <4 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v4i1_v4i64 = zext <4 x i1> undef to <4 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v4i8_v4i16 = call <4 x i16> @llvm.vp.zext.v4i16.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v4i8_v4i32 = call <4 x i32> @llvm.vp.zext.v4i32.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4i8_v4i64 = call <4 x i64> @llvm.vp.zext.v4i64.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v4i16_v4i32 = call <4 x i32> @llvm.vp.zext.v4i32.v4i16(<4 x i16> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4i16_v4i64 = call <4 x i64> @llvm.vp.zext.v4i64.v4i16(<4 x i16> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4i32_v4i64 = call <4 x i64> @llvm.vp.zext.v4i64.v4i32(<4 x i32> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4i1_v4i8 = call <4 x i8> @llvm.vp.zext.v4i8.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4i1_v4i16 = call <4 x i16> @llvm.vp.zext.v4i16.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4i1_v4i32 = call <4 x i32> @llvm.vp.zext.v4i32.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v4i1_v4i64 = call <4 x i64> @llvm.vp.zext.v4i64.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8i8_v8i16 = zext <8 x i8> undef to <8 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8i8_v8i32 = zext <8 x i8> undef to <8 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8i8_v8i64 = zext <8 x i8> undef to <8 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8i16_v8i32 = zext <8 x i16> undef to <8 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8i16_v8i64 = zext <8 x i16> undef to <8 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8i32_v8i64 = zext <8 x i32> undef to <8 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8i1_v8i8 = zext <8 x i1> undef to <8 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8i1_v8i16 = zext <8 x i1> undef to <8 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8i1_v8i32 = zext <8 x i1> undef to <8 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v8i1_v8i64 = zext <8 x i1> undef to <8 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v8i8_v8i16 = call <8 x i16> @llvm.vp.zext.v8i16.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8i8_v8i32 = call <8 x i32> @llvm.vp.zext.v8i32.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v8i8_v8i64 = call <8 x i64> @llvm.vp.zext.v8i64.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8i16_v8i32 = call <8 x i32> @llvm.vp.zext.v8i32.v8i16(<8 x i16> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v8i16_v8i64 = call <8 x i64> @llvm.vp.zext.v8i64.v8i16(<8 x i16> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v8i32_v8i64 = call <8 x i64> @llvm.vp.zext.v8i64.v8i32(<8 x i32> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8i1_v8i8 = call <8 x i8> @llvm.vp.zext.v8i8.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8i1_v8i16 = call <8 x i16> @llvm.vp.zext.v8i16.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v8i1_v8i32 = call <8 x i32> @llvm.vp.zext.v8i32.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v8i1_v8i64 = call <8 x i64> @llvm.vp.zext.v8i64.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v16i8_v16i16 = zext <16 x i8> undef to <16 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v16i8_v16i32 = zext <16 x i8> undef to <16 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v16i8_v16i64 = zext <16 x i8> undef to <16 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v16i16_v16i32 = zext <16 x i16> undef to <16 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v16i16_v16i64 = zext <16 x i16> undef to <16 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v16i32_v16i64 = zext <16 x i32> undef to <16 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v16i1_v16i8 = zext <16 x i1> undef to <16 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v16i1_v16i16 = zext <16 x i1> undef to <16 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v16i1_v16i32 = zext <16 x i1> undef to <16 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v16i1_v16i64 = zext <16 x i1> undef to <16 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v16i8_v16i16 = call <16 x i16> @llvm.vp.zext.v16i16.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v16i8_v16i32 = call <16 x i32> @llvm.vp.zext.v16i32.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v16i8_v16i64 = call <16 x i64> @llvm.vp.zext.v16i64.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v16i16_v16i32 = call <16 x i32> @llvm.vp.zext.v16i32.v16i16(<16 x i16> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v16i16_v16i64 = call <16 x i64> @llvm.vp.zext.v16i64.v16i16(<16 x i16> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v16i32_v16i64 = call <16 x i64> @llvm.vp.zext.v16i64.v16i32(<16 x i32> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v16i1_v16i8 = call <16 x i8> @llvm.vp.zext.v16i8.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v16i1_v16i16 = call <16 x i16> @llvm.vp.zext.v16i16.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v16i1_v16i32 = call <16 x i32> @llvm.vp.zext.v16i32.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v16i1_v16i64 = call <16 x i64> @llvm.vp.zext.v16i64.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v32i8_v32i16 = zext <32 x i8> undef to <32 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v32i8_v32i32 = zext <32 x i8> undef to <32 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %v32i8_v32i64 = zext <32 x i8> undef to <32 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v32i16_v32i32 = zext <32 x i16> undef to <32 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %v32i16_v32i64 = zext <32 x i16> undef to <32 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %v32i32_v32i64 = zext <32 x i32> undef to <32 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v32i1_v32i8 = zext <32 x i1> undef to <32 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v32i1_v32i16 = zext <32 x i1> undef to <32 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v32i1_v32i32 = zext <32 x i1> undef to <32 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %v32i1_v32i64 = zext <32 x i1> undef to <32 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v32i8_v32i16 = call <32 x i16> @llvm.vp.zext.v32i16.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v32i8_v32i32 = call <32 x i32> @llvm.vp.zext.v32i32.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_v32i8_v32i64 = call <32 x i64> @llvm.vp.zext.v32i64.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v32i16_v32i32 = call <32 x i32> @llvm.vp.zext.v32i32.v32i16(<32 x i16> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_v32i16_v32i64 = call <32 x i64> @llvm.vp.zext.v32i64.v32i16(<32 x i16> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_v32i32_v32i64 = call <32 x i64> @llvm.vp.zext.v32i64.v32i32(<32 x i32> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v32i1_v32i8 = call <32 x i8> @llvm.vp.zext.v32i8.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v32i1_v32i16 = call <32 x i16> @llvm.vp.zext.v32i16.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v32i1_v32i32 = call <32 x i32> @llvm.vp.zext.v32i32.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_v32i1_v32i64 = call <32 x i64> @llvm.vp.zext.v32i64.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v64i8_v64i16 = zext <64 x i8> undef to <64 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %v64i8_v64i32 = zext <64 x i8> undef to <64 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %v64i8_v64i64 = zext <64 x i8> undef to <64 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %v64i16_v64i32 = zext <64 x i16> undef to <64 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %v64i16_v64i64 = zext <64 x i16> undef to <64 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %v64i32_v64i64 = zext <64 x i32> undef to <64 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v64i1_v64i8 = zext <64 x i1> undef to <64 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v64i1_v64i16 = zext <64 x i1> undef to <64 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %v64i1_v64i32 = zext <64 x i1> undef to <64 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %v64i1_v64i64 = zext <64 x i1> undef to <64 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v64i8_v64i16 = call <64 x i16> @llvm.vp.zext.v64i16.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_v64i8_v64i32 = call <64 x i32> @llvm.vp.zext.v64i32.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %vp_v64i8_v64i64 = call <64 x i64> @llvm.vp.zext.v64i64.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_v64i16_v64i32 = call <64 x i32> @llvm.vp.zext.v64i32.v64i16(<64 x i16> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %vp_v64i16_v64i64 = call <64 x i64> @llvm.vp.zext.v64i64.v64i16(<64 x i16> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_v64i32_v64i64 = call <64 x i64> @llvm.vp.zext.v64i64.v64i32(<64 x i32> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v64i1_v64i8 = call <64 x i8> @llvm.vp.zext.v64i8.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v64i1_v64i16 = call <64 x i16> @llvm.vp.zext.v64i16.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_v64i1_v64i32 = call <64 x i32> @llvm.vp.zext.v64i32.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %vp_v64i1_v64i64 = call <64 x i64> @llvm.vp.zext.v64i64.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %v128i8_v128i16 = zext <128 x i8> undef to <128 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %v128i8_v128i32 = zext <128 x i8> undef to <128 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 71 for instruction: %v128i8_v128i64 = zext <128 x i8> undef to <128 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %v128i16_v128i32 = zext <128 x i16> undef to <128 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 70 for instruction: %v128i16_v128i64 = zext <128 x i16> undef to <128 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %v128i32_v128i64 = zext <128 x i32> undef to <128 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v128i1_v128i8 = zext <128 x i1> undef to <128 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %v128i1_v128i16 = zext <128 x i1> undef to <128 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %v128i1_v128i32 = zext <128 x i1> undef to <128 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 79 for instruction: %v128i1_v128i64 = zext <128 x i1> undef to <128 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_v128i8_v128i16 = call <128 x i16> @llvm.vp.zext.v128i16.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %vp_v128i8_v128i32 = call <128 x i32> @llvm.vp.zext.v128i32.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 71 for instruction: %vp_v128i8_v128i64 = call <128 x i64> @llvm.vp.zext.v128i64.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_v128i16_v128i32 = call <128 x i32> @llvm.vp.zext.v128i32.v128i16(<128 x i16> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 70 for instruction: %vp_v128i16_v128i64 = call <128 x i64> @llvm.vp.zext.v128i64.v128i16(<128 x i16> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %vp_v128i32_v128i64 = call <128 x i64> @llvm.vp.zext.v128i64.v128i32(<128 x i32> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v128i1_v128i8 = call <128 x i8> @llvm.vp.zext.v128i8.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_v128i1_v128i16 = call <128 x i16> @llvm.vp.zext.v128i16.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %vp_v128i1_v128i32 = call <128 x i32> @llvm.vp.zext.v128i32.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 79 for instruction: %vp_v128i1_v128i64 = call <128 x i64> @llvm.vp.zext.v128i64.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %v256i8_v256i16 = zext <256 x i8> undef to <256 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 70 for instruction: %v256i8_v256i32 = zext <256 x i8> undef to <256 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 142 for instruction: %v256i8_v256i64 = zext <256 x i8> undef to <256 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %v256i16_v256i32 = zext <256 x i16> undef to <256 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %v256i16_v256i64 = zext <256 x i16> undef to <256 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %v256i32_v256i64 = zext <256 x i32> undef to <256 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %v256i1_v256i8 = zext <256 x i1> undef to <256 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %v256i1_v256i16 = zext <256 x i1> undef to <256 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 79 for instruction: %v256i1_v256i32 = zext <256 x i1> undef to <256 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 151 for instruction: %v256i1_v256i64 = zext <256 x i1> undef to <256 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_v256i8_v256i16 = call <256 x i16> @llvm.vp.zext.v256i16.v256i8(<256 x i8> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 70 for instruction: %vp_v256i8_v256i32 = call <256 x i32> @llvm.vp.zext.v256i32.v256i8(<256 x i8> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 142 for instruction: %vp_v256i8_v256i64 = call <256 x i64> @llvm.vp.zext.v256i64.v256i8(<256 x i8> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %vp_v256i16_v256i32 = call <256 x i32> @llvm.vp.zext.v256i32.v256i16(<256 x i16> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 140 for instruction: %vp_v256i16_v256i64 = call <256 x i64> @llvm.vp.zext.v256i64.v256i16(<256 x i16> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %vp_v256i32_v256i64 = call <256 x i64> @llvm.vp.zext.v256i64.v256i32(<256 x i32> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_v256i1_v256i8 = call <256 x i8> @llvm.vp.zext.v256i8.v256i1(<256 x i1> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %vp_v256i1_v256i16 = call <256 x i16> @llvm.vp.zext.v256i16.v256i1(<256 x i1> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 79 for instruction: %vp_v256i1_v256i32 = call <256 x i32> @llvm.vp.zext.v256i32.v256i1(<256 x i1> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 151 for instruction: %vp_v256i1_v256i64 = call <256 x i64> @llvm.vp.zext.v256i64.v256i1(<256 x i1> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1i8_nxv1i16 = zext <vscale x 1 x i8> undef to <vscale x 1 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1i8_nxv1i32 = zext <vscale x 1 x i8> undef to <vscale x 1 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1i8_nxv1i64 = zext <vscale x 1 x i8> undef to <vscale x 1 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1i16_nxv1i32 = zext <vscale x 1 x i16> undef to <vscale x 1 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1i16_nxv1i64 = zext <vscale x 1 x i16> undef to <vscale x 1 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1i32_nxv1i64 = zext <vscale x 1 x i32> undef to <vscale x 1 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1i1_nxv1i8 = zext <vscale x 1 x i1> undef to <vscale x 1 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1i1_nxv1i16 = zext <vscale x 1 x i1> undef to <vscale x 1 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1i1_nxv1i32 = zext <vscale x 1 x i1> undef to <vscale x 1 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1i1_nxv1i64 = zext <vscale x 1 x i1> undef to <vscale x 1 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1i8_nxv1i16 = call <vscale x 1 x i16> @llvm.vp.zext.nxv1i16.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1i8_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.zext.nxv1i32.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1i8_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.zext.nxv1i64.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1i16_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.zext.nxv1i32.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1i16_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.zext.nxv1i64.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1i32_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.zext.nxv1i64.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1i1_nxv1i8 = call <vscale x 1 x i8> @llvm.vp.zext.nxv1i8.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1i1_nxv1i16 = call <vscale x 1 x i16> @llvm.vp.zext.nxv1i16.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1i1_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.zext.nxv1i32.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1i1_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.zext.nxv1i64.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv2i8_nxv2i16 = zext <vscale x 2 x i8> undef to <vscale x 2 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv2i8_nxv2i32 = zext <vscale x 2 x i8> undef to <vscale x 2 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2i8_nxv2i64 = zext <vscale x 2 x i8> undef to <vscale x 2 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv2i16_nxv2i32 = zext <vscale x 2 x i16> undef to <vscale x 2 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2i16_nxv2i64 = zext <vscale x 2 x i16> undef to <vscale x 2 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2i32_nxv2i64 = zext <vscale x 2 x i32> undef to <vscale x 2 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2i1_nxv2i8 = zext <vscale x 2 x i1> undef to <vscale x 2 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2i1_nxv2i16 = zext <vscale x 2 x i1> undef to <vscale x 2 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2i1_nxv2i32 = zext <vscale x 2 x i1> undef to <vscale x 2 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv2i1_nxv2i64 = zext <vscale x 2 x i1> undef to <vscale x 2 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv2i8_nxv2i16 = call <vscale x 2 x i16> @llvm.vp.zext.nxv2i16.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv2i8_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.zext.nxv2i32.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2i8_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.zext.nxv2i64.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv2i16_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.zext.nxv2i32.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2i16_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.zext.nxv2i64.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2i32_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.zext.nxv2i64.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2i1_nxv2i8 = call <vscale x 2 x i8> @llvm.vp.zext.nxv2i8.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2i1_nxv2i16 = call <vscale x 2 x i16> @llvm.vp.zext.nxv2i16.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2i1_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.zext.nxv2i32.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv2i1_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.zext.nxv2i64.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv4i8_nxv4i16 = zext <vscale x 4 x i8> undef to <vscale x 4 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4i8_nxv4i32 = zext <vscale x 4 x i8> undef to <vscale x 4 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv4i8_nxv4i64 = zext <vscale x 4 x i8> undef to <vscale x 4 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4i16_nxv4i32 = zext <vscale x 4 x i16> undef to <vscale x 4 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv4i16_nxv4i64 = zext <vscale x 4 x i16> undef to <vscale x 4 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv4i32_nxv4i64 = zext <vscale x 4 x i32> undef to <vscale x 4 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4i1_nxv4i8 = zext <vscale x 4 x i1> undef to <vscale x 4 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4i1_nxv4i16 = zext <vscale x 4 x i1> undef to <vscale x 4 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv4i1_nxv4i32 = zext <vscale x 4 x i1> undef to <vscale x 4 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv4i1_nxv4i64 = zext <vscale x 4 x i1> undef to <vscale x 4 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv4i8_nxv4i16 = call <vscale x 4 x i16> @llvm.vp.zext.nxv4i16.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4i8_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.zext.nxv4i32.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv4i8_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.zext.nxv4i64.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4i16_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.zext.nxv4i32.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv4i16_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.zext.nxv4i64.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv4i32_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.zext.nxv4i64.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4i1_nxv4i8 = call <vscale x 4 x i8> @llvm.vp.zext.nxv4i8.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4i1_nxv4i16 = call <vscale x 4 x i16> @llvm.vp.zext.nxv4i16.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv4i1_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.zext.nxv4i32.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv4i1_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.zext.nxv4i64.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv8i8_nxv8i16 = zext <vscale x 8 x i8> undef to <vscale x 8 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv8i8_nxv8i32 = zext <vscale x 8 x i8> undef to <vscale x 8 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv8i8_nxv8i64 = zext <vscale x 8 x i8> undef to <vscale x 8 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv8i16_nxv8i32 = zext <vscale x 8 x i16> undef to <vscale x 8 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv8i16_nxv8i64 = zext <vscale x 8 x i16> undef to <vscale x 8 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv8i32_nxv8i64 = zext <vscale x 8 x i32> undef to <vscale x 8 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv8i1_nxv8i8 = zext <vscale x 8 x i1> undef to <vscale x 8 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv8i1_nxv8i16 = zext <vscale x 8 x i1> undef to <vscale x 8 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv8i1_nxv8i32 = zext <vscale x 8 x i1> undef to <vscale x 8 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %nxv8i1_nxv8i64 = zext <vscale x 8 x i1> undef to <vscale x 8 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv8i8_nxv8i16 = call <vscale x 8 x i16> @llvm.vp.zext.nxv8i16.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv8i8_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.zext.nxv8i32.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv8i8_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.zext.nxv8i64.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv8i16_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.zext.nxv8i32.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv8i16_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.zext.nxv8i64.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv8i32_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.zext.nxv8i64.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv8i1_nxv8i8 = call <vscale x 8 x i8> @llvm.vp.zext.nxv8i8.nxv8i1(<vscale x 8 x i1> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv8i1_nxv8i16 = call <vscale x 8 x i16> @llvm.vp.zext.nxv8i16.nxv8i1(<vscale x 8 x i1> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv8i1_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.zext.nxv8i32.nxv8i1(<vscale x 8 x i1> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_nxv8i1_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.zext.nxv8i64.nxv8i1(<vscale x 8 x i1> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv16i8_nxv16i16 = zext <vscale x 16 x i8> undef to <vscale x 16 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv16i8_nxv16i32 = zext <vscale x 16 x i8> undef to <vscale x 16 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %nxv16i8_nxv16i64 = zext <vscale x 16 x i8> undef to <vscale x 16 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv16i16_nxv16i32 = zext <vscale x 16 x i16> undef to <vscale x 16 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %nxv16i16_nxv16i64 = zext <vscale x 16 x i16> undef to <vscale x 16 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %nxv16i32_nxv16i64 = zext <vscale x 16 x i32> undef to <vscale x 16 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv16i1_nxv16i8 = zext <vscale x 16 x i1> undef to <vscale x 16 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv16i1_nxv16i16 = zext <vscale x 16 x i1> undef to <vscale x 16 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %nxv16i1_nxv16i32 = zext <vscale x 16 x i1> undef to <vscale x 16 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %nxv16i1_nxv16i64 = zext <vscale x 16 x i1> undef to <vscale x 16 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv16i8_nxv16i16 = call <vscale x 16 x i16> @llvm.vp.zext.nxv16i16.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv16i8_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.zext.nxv16i32.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_nxv16i8_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.zext.nxv16i64.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv16i16_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.zext.nxv16i32.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_nxv16i16_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.zext.nxv16i64.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_nxv16i32_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.zext.nxv16i64.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv16i1_nxv16i8 = call <vscale x 16 x i8> @llvm.vp.zext.nxv16i8.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv16i1_nxv16i16 = call <vscale x 16 x i16> @llvm.vp.zext.nxv16i16.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_nxv16i1_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.zext.nxv16i32.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_nxv16i1_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.zext.nxv16i64.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv32i8_nxv32i16 = zext <vscale x 32 x i8> undef to <vscale x 32 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %nxv32i8_nxv32i32 = zext <vscale x 32 x i8> undef to <vscale x 32 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %nxv32i8_nxv32i64 = zext <vscale x 32 x i8> undef to <vscale x 32 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %nxv32i16_nxv32i32 = zext <vscale x 32 x i16> undef to <vscale x 32 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %nxv32i16_nxv32i64 = zext <vscale x 32 x i16> undef to <vscale x 32 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %nxv32i32_nxv32i64 = zext <vscale x 32 x i32> undef to <vscale x 32 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv32i1_nxv32i8 = zext <vscale x 32 x i1> undef to <vscale x 32 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %nxv32i1_nxv32i16 = zext <vscale x 32 x i1> undef to <vscale x 32 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %nxv32i1_nxv32i32 = zext <vscale x 32 x i1> undef to <vscale x 32 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %nxv32i1_nxv32i64 = zext <vscale x 32 x i1> undef to <vscale x 32 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv32i8_nxv32i16 = call <vscale x 32 x i16> @llvm.vp.zext.nxv32i16.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_nxv32i8_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.zext.nxv32i32.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %vp_nxv32i8_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.zext.nxv32i64.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_nxv32i16_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.zext.nxv32i32.nxv32i16(<vscale x 32 x i16> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %vp_nxv32i16_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.zext.nxv32i64.nxv32i16(<vscale x 32 x i16> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_nxv32i32_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.zext.nxv32i64.nxv32i32(<vscale x 32 x i32> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv32i1_nxv32i8 = call <vscale x 32 x i8> @llvm.vp.zext.nxv32i8.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_nxv32i1_nxv32i16 = call <vscale x 32 x i16> @llvm.vp.zext.nxv32i16.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_nxv32i1_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.zext.nxv32i32.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %vp_nxv32i1_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.zext.nxv32i64.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %nxv64i8_nxv64i16 = zext <vscale x 64 x i8> undef to <vscale x 64 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %nxv64i8_nxv64i32 = zext <vscale x 64 x i8> undef to <vscale x 64 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 71 for instruction: %nxv64i8_nxv64i64 = zext <vscale x 64 x i8> undef to <vscale x 64 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %nxv64i16_nxv64i32 = zext <vscale x 64 x i16> undef to <vscale x 64 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 70 for instruction: %nxv64i16_nxv64i64 = zext <vscale x 64 x i16> undef to <vscale x 64 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %nxv64i32_nxv64i64 = zext <vscale x 64 x i32> undef to <vscale x 64 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %nxv64i1_nxv64i8 = zext <vscale x 64 x i1> undef to <vscale x 64 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %nxv64i1_nxv64i16 = zext <vscale x 64 x i1> undef to <vscale x 64 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %nxv64i1_nxv64i32 = zext <vscale x 64 x i1> undef to <vscale x 64 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 79 for instruction: %nxv64i1_nxv64i64 = zext <vscale x 64 x i1> undef to <vscale x 64 x i64>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_nxv64i8_nxv64i16 = call <vscale x 64 x i16> @llvm.vp.zext.nxv64i16.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %vp_nxv64i8_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.zext.nxv64i32.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 71 for instruction: %vp_nxv64i8_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.zext.nxv64i64.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_nxv64i16_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.zext.nxv64i32.nxv64i16(<vscale x 64 x i16> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 70 for instruction: %vp_nxv64i16_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.zext.nxv64i64.nxv64i16(<vscale x 64 x i16> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %vp_nxv64i32_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.zext.nxv64i64.nxv64i32(<vscale x 64 x i32> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_nxv64i1_nxv64i8 = call <vscale x 64 x i8> @llvm.vp.zext.nxv64i8.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_nxv64i1_nxv64i16 = call <vscale x 64 x i16> @llvm.vp.zext.nxv64i16.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %vp_nxv64i1_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.zext.nxv64i32.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 79 for instruction: %vp_nxv64i1_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.zext.nxv64i64.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %nxv128i8_nxv128i16 = zext <vscale x 128 x i8> undef to <vscale x 128 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 70 for instruction: %nxv128i8_nxv128i32 = zext <vscale x 128 x i8> undef to <vscale x 128 x i32>
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %nxv128i8_nxv128i128 = zext <vscale x 128 x i8> undef to <vscale x 128 x i128>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %nxv128i16_nxv128i32 = zext <vscale x 128 x i16> undef to <vscale x 128 x i32>
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %nxv128i16_nxv128i128 = zext <vscale x 128 x i16> undef to <vscale x 128 x i128>
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %nxv128i32_nxv128i128 = zext <vscale x 128 x i32> undef to <vscale x 128 x i128>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %nxv128i1_nxv128i8 = zext <vscale x 128 x i1> undef to <vscale x 128 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %nxv128i1_nxv128i16 = zext <vscale x 128 x i1> undef to <vscale x 128 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 79 for instruction: %nxv128i1_nxv128i32 = zext <vscale x 128 x i1> undef to <vscale x 128 x i32>
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %nxv128i1_nxv128i128 = zext <vscale x 128 x i1> undef to <vscale x 128 x i128>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_nxv128i8_nxv128i16 = call <vscale x 128 x i16> @llvm.vp.zext.nxv128i16.nxv128i8(<vscale x 128 x i8> undef, <vscale x 128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 70 for instruction: %vp_nxv128i8_nxv128i32 = call <vscale x 128 x i32> @llvm.vp.zext.nxv128i32.nxv128i8(<vscale x 128 x i8> undef, <vscale x 128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv128i8_nxv128i128 = call <vscale x 128 x i128> @llvm.vp.zext.nxv128i128.nxv128i8(<vscale x 128 x i8> undef, <vscale x 128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %vp_nxv128i16_nxv128i32 = call <vscale x 128 x i32> @llvm.vp.zext.nxv128i32.nxv128i16(<vscale x 128 x i16> undef, <vscale x 128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv128i16_nxv128i128 = call <vscale x 128 x i128> @llvm.vp.zext.nxv128i128.nxv128i16(<vscale x 128 x i16> undef, <vscale x 128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv128i32_nxv128i128 = call <vscale x 128 x i128> @llvm.vp.zext.nxv128i128.nxv128i32(<vscale x 128 x i32> undef, <vscale x 128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_nxv128i1_nxv128i8 = call <vscale x 128 x i8> @llvm.vp.zext.nxv128i8.nxv128i1(<vscale x 128 x i1> undef, <vscale x 128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 43 for instruction: %vp_nxv128i1_nxv128i16 = call <vscale x 128 x i16> @llvm.vp.zext.nxv128i16.nxv128i1(<vscale x 128 x i1> undef, <vscale x 128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 79 for instruction: %vp_nxv128i1_nxv128i32 = call <vscale x 128 x i32> @llvm.vp.zext.nxv128i32.nxv128i1(<vscale x 128 x i1> undef, <vscale x 128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv128i1_nxv128i128 = call <vscale x 128 x i128> @llvm.vp.zext.nxv128i128.nxv128i1(<vscale x 128 x i1> undef, <vscale x 128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  %v2i8_v2i16 = zext <2 x i8> undef to <2 x i16>
  %v2i8_v2i32 = zext <2 x i8> undef to <2 x i32>
  %v2i8_v2i64 = zext <2 x i8> undef to <2 x i64>
  %v2i16_v2i32 = zext <2 x i16> undef to <2 x i32>
  %v2i16_v2i64 = zext <2 x i16> undef to <2 x i64>
  %v2i32_v2i64 = zext <2 x i32> undef to <2 x i64>
  %v2i1_v2i8 = zext <2 x i1> undef to <2 x i8>
  %v2i1_v2i16 = zext <2 x i1> undef to <2 x i16>
  %v2i1_v2i32 = zext <2 x i1> undef to <2 x i32>
  %v2i1_v2i64 = zext <2 x i1> undef to <2 x i64>

  %vp_v2i8_v2i16 = call <2 x i16> @llvm.vp.zext.v2i16.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
  %vp_v2i8_v2i32 = call <2 x i32> @llvm.vp.zext.v2i32.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
  %vp_v2i8_v2i64 = call <2 x i64> @llvm.vp.zext.v2i64.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
  %vp_v2i16_v2i32 = call <2 x i32> @llvm.vp.zext.v2i32.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
  %vp_v2i16_v2i64 = call <2 x i64> @llvm.vp.zext.v2i64.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
  %vp_v2i32_v2i64 = call <2 x i64> @llvm.vp.zext.v2i64.v2i32(<2 x i32> undef, <2 x i1> undef, i32 undef)
  %vp_v2i1_v2i8 = call <2 x i8> @llvm.vp.zext.v2i8.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)
  %vp_v2i1_v2i16 = call <2 x i16> @llvm.vp.zext.v2i16.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)
  %vp_v2i1_v2i32 = call <2 x i32> @llvm.vp.zext.v2i32.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)
  %vp_v2i1_v2i64 = call <2 x i64> @llvm.vp.zext.v2i64.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)

  %v4i8_v4i16 = zext <4 x i8> undef to <4 x i16>
  %v4i8_v4i32 = zext <4 x i8> undef to <4 x i32>
  %v4i8_v4i64 = zext <4 x i8> undef to <4 x i64>
  %v4i16_v4i32 = zext <4 x i16> undef to <4 x i32>
  %v4i16_v4i64 = zext <4 x i16> undef to <4 x i64>
  %v4i32_v4i64 = zext <4 x i32> undef to <4 x i64>
  %v4i1_v4i8 = zext <4 x i1> undef to <4 x i8>
  %v4i1_v4i16 = zext <4 x i1> undef to <4 x i16>
  %v4i1_v4i32 = zext <4 x i1> undef to <4 x i32>
  %v4i1_v4i64 = zext <4 x i1> undef to <4 x i64>

  %vp_v4i8_v4i16 = call <4 x i16> @llvm.vp.zext.v4i16.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
  %vp_v4i8_v4i32 = call <4 x i32> @llvm.vp.zext.v4i32.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
  %vp_v4i8_v4i64 = call <4 x i64> @llvm.vp.zext.v4i64.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
  %vp_v4i16_v4i32 = call <4 x i32> @llvm.vp.zext.v4i32.v4i16(<4 x i16> undef, <4 x i1> undef, i32 undef)
  %vp_v4i16_v4i64 = call <4 x i64> @llvm.vp.zext.v4i64.v4i16(<4 x i16> undef, <4 x i1> undef, i32 undef)
  %vp_v4i32_v4i64 = call <4 x i64> @llvm.vp.zext.v4i64.v4i32(<4 x i32> undef, <4 x i1> undef, i32 undef)
  %vp_v4i1_v4i8 = call <4 x i8> @llvm.vp.zext.v4i8.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)
  %vp_v4i1_v4i16 = call <4 x i16> @llvm.vp.zext.v4i16.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)
  %vp_v4i1_v4i32 = call <4 x i32> @llvm.vp.zext.v4i32.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)
  %vp_v4i1_v4i64 = call <4 x i64> @llvm.vp.zext.v4i64.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)

  %v8i8_v8i16 = zext <8 x i8> undef to <8 x i16>
  %v8i8_v8i32 = zext <8 x i8> undef to <8 x i32>
  %v8i8_v8i64 = zext <8 x i8> undef to <8 x i64>
  %v8i16_v8i32 = zext <8 x i16> undef to <8 x i32>
  %v8i16_v8i64 = zext <8 x i16> undef to <8 x i64>
  %v8i32_v8i64 = zext <8 x i32> undef to <8 x i64>
  %v8i1_v8i8 = zext <8 x i1> undef to <8 x i8>
  %v8i1_v8i16 = zext <8 x i1> undef to <8 x i16>
  %v8i1_v8i32 = zext <8 x i1> undef to <8 x i32>
  %v8i1_v8i64 = zext <8 x i1> undef to <8 x i64>

  %vp_v8i8_v8i16 = call <8 x i16> @llvm.vp.zext.v8i16.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
  %vp_v8i8_v8i32 = call <8 x i32> @llvm.vp.zext.v8i32.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
  %vp_v8i8_v8i64 = call <8 x i64> @llvm.vp.zext.v8i64.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
  %vp_v8i16_v8i32 = call <8 x i32> @llvm.vp.zext.v8i32.v8i16(<8 x i16> undef, <8 x i1> undef, i32 undef)
  %vp_v8i16_v8i64 = call <8 x i64> @llvm.vp.zext.v8i64.v8i16(<8 x i16> undef, <8 x i1> undef, i32 undef)
  %vp_v8i32_v8i64 = call <8 x i64> @llvm.vp.zext.v8i64.v8i32(<8 x i32> undef, <8 x i1> undef, i32 undef)
  %vp_v8i1_v8i8 = call <8 x i8> @llvm.vp.zext.v8i8.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)
  %vp_v8i1_v8i16 = call <8 x i16> @llvm.vp.zext.v8i16.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)
  %vp_v8i1_v8i32 = call <8 x i32> @llvm.vp.zext.v8i32.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)
  %vp_v8i1_v8i64 = call <8 x i64> @llvm.vp.zext.v8i64.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)

  %v16i8_v16i16 = zext <16 x i8> undef to <16 x i16>
  %v16i8_v16i32 = zext <16 x i8> undef to <16 x i32>
  %v16i8_v16i64 = zext <16 x i8> undef to <16 x i64>
  %v16i16_v16i32 = zext <16 x i16> undef to <16 x i32>
  %v16i16_v16i64 = zext <16 x i16> undef to <16 x i64>
  %v16i32_v16i64 = zext <16 x i32> undef to <16 x i64>
  %v16i1_v16i8 = zext <16 x i1> undef to <16 x i8>
  %v16i1_v16i16 = zext <16 x i1> undef to <16 x i16>
  %v16i1_v16i32 = zext <16 x i1> undef to <16 x i32>
  %v16i1_v16i64 = zext <16 x i1> undef to <16 x i64>

  %vp_v16i8_v16i16 = call <16 x i16> @llvm.vp.zext.v16i16.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
  %vp_v16i8_v16i32 = call <16 x i32> @llvm.vp.zext.v16i32.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
  %vp_v16i8_v16i64 = call <16 x i64> @llvm.vp.zext.v16i64.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
  %vp_v16i16_v16i32 = call <16 x i32> @llvm.vp.zext.v16i32.v16i16(<16 x i16> undef, <16 x i1> undef, i32 undef)
  %vp_v16i16_v16i64 = call <16 x i64> @llvm.vp.zext.v16i64.v16i16(<16 x i16> undef, <16 x i1> undef, i32 undef)
  %vp_v16i32_v16i64 = call <16 x i64> @llvm.vp.zext.v16i64.v16i32(<16 x i32> undef, <16 x i1> undef, i32 undef)
  %vp_v16i1_v16i8 = call <16 x i8> @llvm.vp.zext.v16i8.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)
  %vp_v16i1_v16i16 = call <16 x i16> @llvm.vp.zext.v16i16.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)
  %vp_v16i1_v16i32 = call <16 x i32> @llvm.vp.zext.v16i32.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)
  %vp_v16i1_v16i64 = call <16 x i64> @llvm.vp.zext.v16i64.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)

  %v32i8_v32i16 = zext <32 x i8> undef to <32 x i16>
  %v32i8_v32i32 = zext <32 x i8> undef to <32 x i32>
  %v32i8_v32i64 = zext <32 x i8> undef to <32 x i64>
  %v32i16_v32i32 = zext <32 x i16> undef to <32 x i32>
  %v32i16_v32i64 = zext <32 x i16> undef to <32 x i64>
  %v32i32_v32i64 = zext <32 x i32> undef to <32 x i64>
  %v32i1_v32i8 = zext <32 x i1> undef to <32 x i8>
  %v32i1_v32i16 = zext <32 x i1> undef to <32 x i16>
  %v32i1_v32i32 = zext <32 x i1> undef to <32 x i32>
  %v32i1_v32i64 = zext <32 x i1> undef to <32 x i64>

  %vp_v32i8_v32i16 = call <32 x i16> @llvm.vp.zext.v32i16.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
  %vp_v32i8_v32i32 = call <32 x i32> @llvm.vp.zext.v32i32.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
  %vp_v32i8_v32i64 = call <32 x i64> @llvm.vp.zext.v32i64.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
  %vp_v32i16_v32i32 = call <32 x i32> @llvm.vp.zext.v32i32.v32i16(<32 x i16> undef, <32 x i1> undef, i32 undef)
  %vp_v32i16_v32i64 = call <32 x i64> @llvm.vp.zext.v32i64.v32i16(<32 x i16> undef, <32 x i1> undef, i32 undef)
  %vp_v32i32_v32i64 = call <32 x i64> @llvm.vp.zext.v32i64.v32i32(<32 x i32> undef, <32 x i1> undef, i32 undef)
  %vp_v32i1_v32i8 = call <32 x i8> @llvm.vp.zext.v32i8.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)
  %vp_v32i1_v32i16 = call <32 x i16> @llvm.vp.zext.v32i16.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)
  %vp_v32i1_v32i32 = call <32 x i32> @llvm.vp.zext.v32i32.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)
  %vp_v32i1_v32i64 = call <32 x i64> @llvm.vp.zext.v32i64.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)

  %v64i8_v64i16 = zext <64 x i8> undef to <64 x i16>
  %v64i8_v64i32 = zext <64 x i8> undef to <64 x i32>
  %v64i8_v64i64 = zext <64 x i8> undef to <64 x i64>
  %v64i16_v64i32 = zext <64 x i16> undef to <64 x i32>
  %v64i16_v64i64 = zext <64 x i16> undef to <64 x i64>
  %v64i32_v64i64 = zext <64 x i32> undef to <64 x i64>
  %v64i1_v64i8 = zext <64 x i1> undef to <64 x i8>
  %v64i1_v64i16 = zext <64 x i1> undef to <64 x i16>
  %v64i1_v64i32 = zext <64 x i1> undef to <64 x i32>
  %v64i1_v64i64 = zext <64 x i1> undef to <64 x i64>

  %vp_v64i8_v64i16 = call <64 x i16> @llvm.vp.zext.v64i16.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
  %vp_v64i8_v64i32 = call <64 x i32> @llvm.vp.zext.v64i32.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
  %vp_v64i8_v64i64 = call <64 x i64> @llvm.vp.zext.v64i64.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
  %vp_v64i16_v64i32 = call <64 x i32> @llvm.vp.zext.v64i32.v64i16(<64 x i16> undef, <64 x i1> undef, i32 undef)
  %vp_v64i16_v64i64 = call <64 x i64> @llvm.vp.zext.v64i64.v64i16(<64 x i16> undef, <64 x i1> undef, i32 undef)
  %vp_v64i32_v64i64 = call <64 x i64> @llvm.vp.zext.v64i64.v64i32(<64 x i32> undef, <64 x i1> undef, i32 undef)
  %vp_v64i1_v64i8 = call <64 x i8> @llvm.vp.zext.v64i8.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)
  %vp_v64i1_v64i16 = call <64 x i16> @llvm.vp.zext.v64i16.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)
  %vp_v64i1_v64i32 = call <64 x i32> @llvm.vp.zext.v64i32.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)
  %vp_v64i1_v64i64 = call <64 x i64> @llvm.vp.zext.v64i64.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)

  %v128i8_v128i16 = zext <128 x i8> undef to <128 x i16>
  %v128i8_v128i32 = zext <128 x i8> undef to <128 x i32>
  %v128i8_v128i64 = zext <128 x i8> undef to <128 x i64>
  %v128i16_v128i32 = zext <128 x i16> undef to <128 x i32>
  %v128i16_v128i64 = zext <128 x i16> undef to <128 x i64>
  %v128i32_v128i64 = zext <128 x i32> undef to <128 x i64>
  %v128i1_v128i8 = zext <128 x i1> undef to <128 x i8>
  %v128i1_v128i16 = zext <128 x i1> undef to <128 x i16>
  %v128i1_v128i32 = zext <128 x i1> undef to <128 x i32>
  %v128i1_v128i64 = zext <128 x i1> undef to <128 x i64>

  %vp_v128i8_v128i16 = call <128 x i16> @llvm.vp.zext.v128i16.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
  %vp_v128i8_v128i32 = call <128 x i32> @llvm.vp.zext.v128i32.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
  %vp_v128i8_v128i64 = call <128 x i64> @llvm.vp.zext.v128i64.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
  %vp_v128i16_v128i32 = call <128 x i32> @llvm.vp.zext.v128i32.v128i16(<128 x i16> undef, <128 x i1> undef, i32 undef)
  %vp_v128i16_v128i64 = call <128 x i64> @llvm.vp.zext.v128i64.v128i16(<128 x i16> undef, <128 x i1> undef, i32 undef)
  %vp_v128i32_v128i64 = call <128 x i64> @llvm.vp.zext.v128i64.v128i32(<128 x i32> undef, <128 x i1> undef, i32 undef)
  %vp_v128i1_v128i8 = call <128 x i8> @llvm.vp.zext.v128i8.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)
  %vp_v128i1_v128i16 = call <128 x i16> @llvm.vp.zext.v128i16.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)
  %vp_v128i1_v128i32 = call <128 x i32> @llvm.vp.zext.v128i32.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)
  %vp_v128i1_v128i64 = call <128 x i64> @llvm.vp.zext.v128i64.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)

  %v256i8_v256i16 = zext <256 x i8> undef to <256 x i16>
  %v256i8_v256i32 = zext <256 x i8> undef to <256 x i32>
  %v256i8_v256i64 = zext <256 x i8> undef to <256 x i64>
  %v256i16_v256i32 = zext <256 x i16> undef to <256 x i32>
  %v256i16_v256i64 = zext <256 x i16> undef to <256 x i64>
  %v256i32_v256i64 = zext <256 x i32> undef to <256 x i64>
  %v256i1_v256i8 = zext <256 x i1> undef to <256 x i8>
  %v256i1_v256i16 = zext <256 x i1> undef to <256 x i16>
  %v256i1_v256i32 = zext <256 x i1> undef to <256 x i32>
  %v256i1_v256i64 = zext <256 x i1> undef to <256 x i64>

  %vp_v256i8_v256i16 = call <256 x i16> @llvm.vp.zext.v256i16.v256i8(<256 x i8> undef, <256 x i1> undef, i32 undef)
  %vp_v256i8_v256i32 = call <256 x i32> @llvm.vp.zext.v256i32.v256i8(<256 x i8> undef, <256 x i1> undef, i32 undef)
  %vp_v256i8_v256i64 = call <256 x i64> @llvm.vp.zext.v256i64.v256i8(<256 x i8> undef, <256 x i1> undef, i32 undef)
  %vp_v256i16_v256i32 = call <256 x i32> @llvm.vp.zext.v256i32.v256i16(<256 x i16> undef, <256 x i1> undef, i32 undef)
  %vp_v256i16_v256i64 = call <256 x i64> @llvm.vp.zext.v256i64.v256i16(<256 x i16> undef, <256 x i1> undef, i32 undef)
  %vp_v256i32_v256i64 = call <256 x i64> @llvm.vp.zext.v256i64.v256i32(<256 x i32> undef, <256 x i1> undef, i32 undef)
  %vp_v256i1_v256i8 = call <256 x i8> @llvm.vp.zext.v256i8.v256i1(<256 x i1> undef, <256 x i1> undef, i32 undef)
  %vp_v256i1_v256i16 = call <256 x i16> @llvm.vp.zext.v256i16.v256i1(<256 x i1> undef, <256 x i1> undef, i32 undef)
  %vp_v256i1_v256i32 = call <256 x i32> @llvm.vp.zext.v256i32.v256i1(<256 x i1> undef, <256 x i1> undef, i32 undef)
  %vp_v256i1_v256i64 = call <256 x i64> @llvm.vp.zext.v256i64.v256i1(<256 x i1> undef, <256 x i1> undef, i32 undef)

  %nxv1i8_nxv1i16 = zext <vscale x 1 x i8> undef to <vscale x 1 x i16>
  %nxv1i8_nxv1i32 = zext <vscale x 1 x i8> undef to <vscale x 1 x i32>
  %nxv1i8_nxv1i64 = zext <vscale x 1 x i8> undef to <vscale x 1 x i64>
  %nxv1i16_nxv1i32 = zext <vscale x 1 x i16> undef to <vscale x 1 x i32>
  %nxv1i16_nxv1i64 = zext <vscale x 1 x i16> undef to <vscale x 1 x i64>
  %nxv1i32_nxv1i64 = zext <vscale x 1 x i32> undef to <vscale x 1 x i64>
  %nxv1i1_nxv1i8 = zext <vscale x 1 x i1> undef to <vscale x 1 x i8>
  %nxv1i1_nxv1i16 = zext <vscale x 1 x i1> undef to <vscale x 1 x i16>
  %nxv1i1_nxv1i32 = zext <vscale x 1 x i1> undef to <vscale x 1 x i32>
  %nxv1i1_nxv1i64 = zext <vscale x 1 x i1> undef to <vscale x 1 x i64>

  %vp_nxv1i8_nxv1i16 = call <vscale x 1 x i16> @llvm.vp.zext.nxv1i16.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i8_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.zext.nxv1i32.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i8_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.zext.nxv1i64.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i16_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.zext.nxv1i32.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i16_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.zext.nxv1i64.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i32_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.zext.nxv1i64.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i1_nxv1i8 = call <vscale x 1 x i8> @llvm.vp.zext.nxv1i8.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i1_nxv1i16 = call <vscale x 1 x i16> @llvm.vp.zext.nxv1i16.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i1_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.zext.nxv1i32.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i1_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.zext.nxv1i64.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)

  %nxv2i8_nxv2i16 = zext <vscale x 2 x i8> undef to <vscale x 2 x i16>
  %nxv2i8_nxv2i32 = zext <vscale x 2 x i8> undef to <vscale x 2 x i32>
  %nxv2i8_nxv2i64 = zext <vscale x 2 x i8> undef to <vscale x 2 x i64>
  %nxv2i16_nxv2i32 = zext <vscale x 2 x i16> undef to <vscale x 2 x i32>
  %nxv2i16_nxv2i64 = zext <vscale x 2 x i16> undef to <vscale x 2 x i64>
  %nxv2i32_nxv2i64 = zext <vscale x 2 x i32> undef to <vscale x 2 x i64>
  %nxv2i1_nxv2i8 = zext <vscale x 2 x i1> undef to <vscale x 2 x i8>
  %nxv2i1_nxv2i16 = zext <vscale x 2 x i1> undef to <vscale x 2 x i16>
  %nxv2i1_nxv2i32 = zext <vscale x 2 x i1> undef to <vscale x 2 x i32>
  %nxv2i1_nxv2i64 = zext <vscale x 2 x i1> undef to <vscale x 2 x i64>

  %vp_nxv2i8_nxv2i16 = call <vscale x 2 x i16> @llvm.vp.zext.nxv2i16.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i8_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.zext.nxv2i32.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i8_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.zext.nxv2i64.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i16_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.zext.nxv2i32.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i16_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.zext.nxv2i64.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i32_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.zext.nxv2i64.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i1_nxv2i8 = call <vscale x 2 x i8> @llvm.vp.zext.nxv2i8.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i1_nxv2i16 = call <vscale x 2 x i16> @llvm.vp.zext.nxv2i16.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i1_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.zext.nxv2i32.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i1_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.zext.nxv2i64.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)

  %nxv4i8_nxv4i16 = zext <vscale x 4 x i8> undef to <vscale x 4 x i16>
  %nxv4i8_nxv4i32 = zext <vscale x 4 x i8> undef to <vscale x 4 x i32>
  %nxv4i8_nxv4i64 = zext <vscale x 4 x i8> undef to <vscale x 4 x i64>
  %nxv4i16_nxv4i32 = zext <vscale x 4 x i16> undef to <vscale x 4 x i32>
  %nxv4i16_nxv4i64 = zext <vscale x 4 x i16> undef to <vscale x 4 x i64>
  %nxv4i32_nxv4i64 = zext <vscale x 4 x i32> undef to <vscale x 4 x i64>
  %nxv4i1_nxv4i8 = zext <vscale x 4 x i1> undef to <vscale x 4 x i8>
  %nxv4i1_nxv4i16 = zext <vscale x 4 x i1> undef to <vscale x 4 x i16>
  %nxv4i1_nxv4i32 = zext <vscale x 4 x i1> undef to <vscale x 4 x i32>
  %nxv4i1_nxv4i64 = zext <vscale x 4 x i1> undef to <vscale x 4 x i64>

  %vp_nxv4i8_nxv4i16 = call <vscale x 4 x i16> @llvm.vp.zext.nxv4i16.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i8_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.zext.nxv4i32.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i8_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.zext.nxv4i64.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i16_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.zext.nxv4i32.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i16_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.zext.nxv4i64.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i32_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.zext.nxv4i64.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i1_nxv4i8 = call <vscale x 4 x i8> @llvm.vp.zext.nxv4i8.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i1_nxv4i16 = call <vscale x 4 x i16> @llvm.vp.zext.nxv4i16.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i1_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.zext.nxv4i32.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i1_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.zext.nxv4i64.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)

  %nxv8i8_nxv8i16 = zext <vscale x 8 x i8> undef to <vscale x 8 x i16>
  %nxv8i8_nxv8i32 = zext <vscale x 8 x i8> undef to <vscale x 8 x i32>
  %nxv8i8_nxv8i64 = zext <vscale x 8 x i8> undef to <vscale x 8 x i64>
  %nxv8i16_nxv8i32 = zext <vscale x 8 x i16> undef to <vscale x 8 x i32>
  %nxv8i16_nxv8i64 = zext <vscale x 8 x i16> undef to <vscale x 8 x i64>
  %nxv8i32_nxv8i64 = zext <vscale x 8 x i32> undef to <vscale x 8 x i64>
  %nxv8i1_nxv8i8 = zext <vscale x 8 x i1> undef to <vscale x 8 x i8>
  %nxv8i1_nxv8i16 = zext <vscale x 8 x i1> undef to <vscale x 8 x i16>
  %nxv8i1_nxv8i32 = zext <vscale x 8 x i1> undef to <vscale x 8 x i32>
  %nxv8i1_nxv8i64 = zext <vscale x 8 x i1> undef to <vscale x 8 x i64>

  %vp_nxv8i8_nxv8i16 = call <vscale x 8 x i16> @llvm.vp.zext.nxv8i16.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i8_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.zext.nxv8i32.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i8_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.zext.nxv8i64.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i16_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.zext.nxv8i32.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i16_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.zext.nxv8i64.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i32_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.zext.nxv8i64.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i1_nxv8i8 = call <vscale x 8 x i8> @llvm.vp.zext.nxv8i8.nxv8i1(<vscale x 8 x i1> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i1_nxv8i16 = call <vscale x 8 x i16> @llvm.vp.zext.nxv8i16.nxv8i1(<vscale x 8 x i1> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i1_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.zext.nxv8i32.nxv8i1(<vscale x 8 x i1> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i1_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.zext.nxv8i64.nxv8i1(<vscale x 8 x i1> undef, <vscale x 8 x i1> undef, i32 undef)

  %nxv16i8_nxv16i16 = zext <vscale x 16 x i8> undef to <vscale x 16 x i16>
  %nxv16i8_nxv16i32 = zext <vscale x 16 x i8> undef to <vscale x 16 x i32>
  %nxv16i8_nxv16i64 = zext <vscale x 16 x i8> undef to <vscale x 16 x i64>
  %nxv16i16_nxv16i32 = zext <vscale x 16 x i16> undef to <vscale x 16 x i32>
  %nxv16i16_nxv16i64 = zext <vscale x 16 x i16> undef to <vscale x 16 x i64>
  %nxv16i32_nxv16i64 = zext <vscale x 16 x i32> undef to <vscale x 16 x i64>
  %nxv16i1_nxv16i8 = zext <vscale x 16 x i1> undef to <vscale x 16 x i8>
  %nxv16i1_nxv16i16 = zext <vscale x 16 x i1> undef to <vscale x 16 x i16>
  %nxv16i1_nxv16i32 = zext <vscale x 16 x i1> undef to <vscale x 16 x i32>
  %nxv16i1_nxv16i64 = zext <vscale x 16 x i1> undef to <vscale x 16 x i64>

  %vp_nxv16i8_nxv16i16 = call <vscale x 16 x i16> @llvm.vp.zext.nxv16i16.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i8_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.zext.nxv16i32.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i8_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.zext.nxv16i64.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i16_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.zext.nxv16i32.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i16_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.zext.nxv16i64.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i32_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.zext.nxv16i64.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i1_nxv16i8 = call <vscale x 16 x i8> @llvm.vp.zext.nxv16i8.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i1_nxv16i16 = call <vscale x 16 x i16> @llvm.vp.zext.nxv16i16.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i1_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.zext.nxv16i32.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i1_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.zext.nxv16i64.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)

  %nxv32i8_nxv32i16 = zext <vscale x 32 x i8> undef to <vscale x 32 x i16>
  %nxv32i8_nxv32i32 = zext <vscale x 32 x i8> undef to <vscale x 32 x i32>
  %nxv32i8_nxv32i64 = zext <vscale x 32 x i8> undef to <vscale x 32 x i64>
  %nxv32i16_nxv32i32 = zext <vscale x 32 x i16> undef to <vscale x 32 x i32>
  %nxv32i16_nxv32i64 = zext <vscale x 32 x i16> undef to <vscale x 32 x i64>
  %nxv32i32_nxv32i64 = zext <vscale x 32 x i32> undef to <vscale x 32 x i64>
  %nxv32i1_nxv32i8 = zext <vscale x 32 x i1> undef to <vscale x 32 x i8>
  %nxv32i1_nxv32i16 = zext <vscale x 32 x i1> undef to <vscale x 32 x i16>
  %nxv32i1_nxv32i32 = zext <vscale x 32 x i1> undef to <vscale x 32 x i32>
  %nxv32i1_nxv32i64 = zext <vscale x 32 x i1> undef to <vscale x 32 x i64>

  %vp_nxv32i8_nxv32i16 = call <vscale x 32 x i16> @llvm.vp.zext.nxv32i16.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i8_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.zext.nxv32i32.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i8_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.zext.nxv32i64.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i16_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.zext.nxv32i32.nxv32i16(<vscale x 32 x i16> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i16_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.zext.nxv32i64.nxv32i16(<vscale x 32 x i16> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i32_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.zext.nxv32i64.nxv32i32(<vscale x 32 x i32> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i1_nxv32i8 = call <vscale x 32 x i8> @llvm.vp.zext.nxv32i8.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i1_nxv32i16 = call <vscale x 32 x i16> @llvm.vp.zext.nxv32i16.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i1_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.zext.nxv32i32.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i1_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.zext.nxv32i64.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)

  %nxv64i8_nxv64i16 = zext <vscale x 64 x i8> undef to <vscale x 64 x i16>
  %nxv64i8_nxv64i32 = zext <vscale x 64 x i8> undef to <vscale x 64 x i32>
  %nxv64i8_nxv64i64 = zext <vscale x 64 x i8> undef to <vscale x 64 x i64>
  %nxv64i16_nxv64i32 = zext <vscale x 64 x i16> undef to <vscale x 64 x i32>
  %nxv64i16_nxv64i64 = zext <vscale x 64 x i16> undef to <vscale x 64 x i64>
  %nxv64i32_nxv64i64 = zext <vscale x 64 x i32> undef to <vscale x 64 x i64>
  %nxv64i1_nxv64i8 = zext <vscale x 64 x i1> undef to <vscale x 64 x i8>
  %nxv64i1_nxv64i16 = zext <vscale x 64 x i1> undef to <vscale x 64 x i16>
  %nxv64i1_nxv64i32 = zext <vscale x 64 x i1> undef to <vscale x 64 x i32>
  %nxv64i1_nxv64i64 = zext <vscale x 64 x i1> undef to <vscale x 64 x i64>

  %vp_nxv64i8_nxv64i16 = call <vscale x 64 x i16> @llvm.vp.zext.nxv64i16.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i8_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.zext.nxv64i32.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i8_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.zext.nxv64i64.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i16_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.zext.nxv64i32.nxv64i16(<vscale x 64 x i16> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i16_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.zext.nxv64i64.nxv64i16(<vscale x 64 x i16> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i32_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.zext.nxv64i64.nxv64i32(<vscale x 64 x i32> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i1_nxv64i8 = call <vscale x 64 x i8> @llvm.vp.zext.nxv64i8.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i1_nxv64i16 = call <vscale x 64 x i16> @llvm.vp.zext.nxv64i16.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i1_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.zext.nxv64i32.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i1_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.zext.nxv64i64.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)

  %nxv128i8_nxv128i16 = zext <vscale x 128 x i8> undef to <vscale x 128 x i16>
  %nxv128i8_nxv128i32 = zext <vscale x 128 x i8> undef to <vscale x 128 x i32>
  %nxv128i8_nxv128i128 = zext <vscale x 128 x i8> undef to <vscale x 128 x i128>
  %nxv128i16_nxv128i32 = zext <vscale x 128 x i16> undef to <vscale x 128 x i32>
  %nxv128i16_nxv128i128 = zext <vscale x 128 x i16> undef to <vscale x 128 x i128>
  %nxv128i32_nxv128i128 = zext <vscale x 128 x i32> undef to <vscale x 128 x i128>
  %nxv128i1_nxv128i8 = zext <vscale x 128 x i1> undef to <vscale x 128 x i8>
  %nxv128i1_nxv128i16 = zext <vscale x 128 x i1> undef to <vscale x 128 x i16>
  %nxv128i1_nxv128i32 = zext <vscale x 128 x i1> undef to <vscale x 128 x i32>
  %nxv128i1_nxv128i128 = zext <vscale x 128 x i1> undef to <vscale x 128 x i128>

  %vp_nxv128i8_nxv128i16 = call <vscale x 128 x i16> @llvm.vp.zext.nxv128i16.nxv128i8(<vscale x 128 x i8> undef, <vscale x 128 x i1> undef, i32 undef)
  %vp_nxv128i8_nxv128i32 = call <vscale x 128 x i32> @llvm.vp.zext.nxv128i32.nxv128i8(<vscale x 128 x i8> undef, <vscale x 128 x i1> undef, i32 undef)
  %vp_nxv128i8_nxv128i128 = call <vscale x 128 x i128> @llvm.vp.zext.nxv128i128.nxv128i8(<vscale x 128 x i8> undef, <vscale x 128 x i1> undef, i32 undef)
  %vp_nxv128i16_nxv128i32 = call <vscale x 128 x i32> @llvm.vp.zext.nxv128i32.nxv128i16(<vscale x 128 x i16> undef, <vscale x 128 x i1> undef, i32 undef)
  %vp_nxv128i16_nxv128i128 = call <vscale x 128 x i128> @llvm.vp.zext.nxv128i128.nxv128i16(<vscale x 128 x i16> undef, <vscale x 128 x i1> undef, i32 undef)
  %vp_nxv128i32_nxv128i128 = call <vscale x 128 x i128> @llvm.vp.zext.nxv128i128.nxv128i32(<vscale x 128 x i32> undef, <vscale x 128 x i1> undef, i32 undef)
  %vp_nxv128i1_nxv128i8 = call <vscale x 128 x i8> @llvm.vp.zext.nxv128i8.nxv128i1(<vscale x 128 x i1> undef, <vscale x 128 x i1> undef, i32 undef)
  %vp_nxv128i1_nxv128i16 = call <vscale x 128 x i16> @llvm.vp.zext.nxv128i16.nxv128i1(<vscale x 128 x i1> undef, <vscale x 128 x i1> undef, i32 undef)
  %vp_nxv128i1_nxv128i32 = call <vscale x 128 x i32> @llvm.vp.zext.nxv128i32.nxv128i1(<vscale x 128 x i1> undef, <vscale x 128 x i1> undef, i32 undef)
  %vp_nxv128i1_nxv128i128 = call <vscale x 128 x i128> @llvm.vp.zext.nxv128i128.nxv128i1(<vscale x 128 x i1> undef, <vscale x 128 x i1> undef, i32 undef)

  ret void
}

define void @trunc() {
; CHECK-LABEL: 'trunc'
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i16_v2i2 = trunc <2 x i16> undef to <2 x i2>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i16_v2i4 = trunc <2 x i16> undef to <2 x i4>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i16_v2i6 = trunc <2 x i16> undef to <2 x i6>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v2i4_v2i2 = trunc <2 x i4> undef to <2 x i2>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v2i6_v2i4 = trunc <2 x i6> undef to <2 x i4>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2i16_v2i2 = call <2 x i2> @llvm.vp.trunc.v2i2.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2i16_v2i4 = call <2 x i4> @llvm.vp.trunc.v2i4.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2i16_v2i6 = call <2 x i6> @llvm.vp.trunc.v2i6.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %vp_v2i4_v2i2 = call <2 x i2> @llvm.vp.trunc.v2i2.v2i4(<2 x i4> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %vp_v2i6_v2i4 = call <2 x i4> @llvm.vp.trunc.v2i4.v2i6(<2 x i6> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i16_v2i8 = trunc <2 x i16> undef to <2 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2i32_v2i8 = trunc <2 x i32> undef to <2 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2i64_v2i8 = trunc <2 x i64> undef to <2 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i32_v2i16 = trunc <2 x i32> undef to <2 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2i64_v2i16 = trunc <2 x i64> undef to <2 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i64_v2i32 = trunc <2 x i64> undef to <2 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2i8_v2i1 = trunc <2 x i8> undef to <2 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2i16_v2i1 = trunc <2 x i16> undef to <2 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2i32_v2i1 = trunc <2 x i32> undef to <2 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2i64_v2i1 = trunc <2 x i64> undef to <2 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2i16_v2i8 = call <2 x i8> @llvm.vp.trunc.v2i8.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2i32_v2i8 = call <2 x i8> @llvm.vp.trunc.v2i8.v2i32(<2 x i32> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2i64_v2i8 = call <2 x i8> @llvm.vp.trunc.v2i8.v2i64(<2 x i64> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2i32_v2i16 = call <2 x i16> @llvm.vp.trunc.v2i16.v2i32(<2 x i32> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2i64_v2i16 = call <2 x i16> @llvm.vp.trunc.v2i16.v2i64(<2 x i64> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2i64_v2i32 = call <2 x i32> @llvm.vp.trunc.v2i32.v2i64(<2 x i64> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2i8_v2i1 = call <2 x i1> @llvm.vp.trunc.v2i1.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2i16_v2i1 = call <2 x i1> @llvm.vp.trunc.v2i1.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2i32_v2i1 = call <2 x i1> @llvm.vp.trunc.v2i1.v2i32(<2 x i32> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2i64_v2i1 = call <2 x i1> @llvm.vp.trunc.v2i1.v2i64(<2 x i64> undef, <2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4i16_v4i8 = trunc <4 x i16> undef to <4 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i32_v4i8 = trunc <4 x i32> undef to <4 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v4i64_v4i8 = trunc <4 x i64> undef to <4 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4i32_v4i16 = trunc <4 x i32> undef to <4 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i64_v4i16 = trunc <4 x i64> undef to <4 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4i64_v4i32 = trunc <4 x i64> undef to <4 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i8_v4i1 = trunc <4 x i8> undef to <4 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i16_v4i1 = trunc <4 x i16> undef to <4 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i32_v4i1 = trunc <4 x i32> undef to <4 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v4i64_v4i1 = trunc <4 x i64> undef to <4 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v4i16_v4i8 = call <4 x i8> @llvm.vp.trunc.v4i8.v4i16(<4 x i16> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4i32_v4i8 = call <4 x i8> @llvm.vp.trunc.v4i8.v4i32(<4 x i32> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v4i64_v4i8 = call <4 x i8> @llvm.vp.trunc.v4i8.v4i64(<4 x i64> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v4i32_v4i16 = call <4 x i16> @llvm.vp.trunc.v4i16.v4i32(<4 x i32> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4i64_v4i16 = call <4 x i16> @llvm.vp.trunc.v4i16.v4i64(<4 x i64> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v4i64_v4i32 = call <4 x i32> @llvm.vp.trunc.v4i32.v4i64(<4 x i64> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4i8_v4i1 = call <4 x i1> @llvm.vp.trunc.v4i1.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4i16_v4i1 = call <4 x i1> @llvm.vp.trunc.v4i1.v4i16(<4 x i16> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4i32_v4i1 = call <4 x i1> @llvm.vp.trunc.v4i1.v4i32(<4 x i32> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v4i64_v4i1 = call <4 x i1> @llvm.vp.trunc.v4i1.v4i64(<4 x i64> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8i16_v8i8 = trunc <8 x i16> undef to <8 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8i32_v8i8 = trunc <8 x i32> undef to <8 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8i64_v8i8 = trunc <8 x i64> undef to <8 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8i32_v8i16 = trunc <8 x i32> undef to <8 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v8i64_v8i16 = trunc <8 x i64> undef to <8 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8i64_v8i32 = trunc <8 x i64> undef to <8 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8i8_v8i1 = trunc <8 x i8> undef to <8 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8i16_v8i1 = trunc <8 x i16> undef to <8 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8i32_v8i1 = trunc <8 x i32> undef to <8 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v8i64_v8i1 = trunc <8 x i64> undef to <8 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v8i16_v8i8 = call <8 x i8> @llvm.vp.trunc.v8i8.v8i16(<8 x i16> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8i32_v8i8 = call <8 x i8> @llvm.vp.trunc.v8i8.v8i32(<8 x i32> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v8i64_v8i8 = call <8 x i8> @llvm.vp.trunc.v8i8.v8i64(<8 x i64> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v8i32_v8i16 = call <8 x i16> @llvm.vp.trunc.v8i16.v8i32(<8 x i32> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v8i64_v8i16 = call <8 x i16> @llvm.vp.trunc.v8i16.v8i64(<8 x i64> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8i64_v8i32 = call <8 x i32> @llvm.vp.trunc.v8i32.v8i64(<8 x i64> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8i8_v8i1 = call <8 x i1> @llvm.vp.trunc.v8i1.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8i16_v8i1 = call <8 x i1> @llvm.vp.trunc.v8i1.v8i16(<8 x i16> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v8i32_v8i1 = call <8 x i1> @llvm.vp.trunc.v8i1.v8i32(<8 x i32> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v8i64_v8i1 = call <8 x i1> @llvm.vp.trunc.v8i1.v8i64(<8 x i64> undef, <8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v16i16_v16i8 = trunc <2 x i16> undef to <2 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v16i32_v16i8 = trunc <2 x i32> undef to <2 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v16i64_v16i8 = trunc <2 x i64> undef to <2 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v16i32_v16i16 = trunc <2 x i32> undef to <2 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v16i64_v16i16 = trunc <2 x i64> undef to <2 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v16i64_v16i32 = trunc <2 x i64> undef to <2 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v16i8_v16i1 = trunc <2 x i8> undef to <2 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v16i16_v16i1 = trunc <2 x i16> undef to <2 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v16i32_v16i1 = trunc <2 x i32> undef to <2 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v16i64_v16i1 = trunc <2 x i64> undef to <2 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v16i16_v16i8 = call <16 x i8> @llvm.vp.trunc.v16i8.v16i16(<16 x i16> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v16i32_v16i8 = call <16 x i8> @llvm.vp.trunc.v16i8.v16i32(<16 x i32> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v16i64_v16i8 = call <16 x i8> @llvm.vp.trunc.v16i8.v16i64(<16 x i64> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v16i32_v16i16 = call <16 x i16> @llvm.vp.trunc.v16i16.v16i32(<16 x i32> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_v16i64_v16i16 = call <16 x i16> @llvm.vp.trunc.v16i16.v16i64(<16 x i64> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v16i64_v16i32 = call <16 x i32> @llvm.vp.trunc.v16i32.v16i64(<16 x i64> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v16i8_v16i1 = call <16 x i1> @llvm.vp.trunc.v16i1.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v16i16_v16i1 = call <16 x i1> @llvm.vp.trunc.v16i1.v16i16(<16 x i16> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v16i32_v16i1 = call <16 x i1> @llvm.vp.trunc.v16i1.v16i32(<16 x i32> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v16i64_v16i1 = call <16 x i1> @llvm.vp.trunc.v16i1.v16i64(<16 x i64> undef, <16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v32i16_v32i8 = trunc <16 x i16> undef to <16 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v32i32_v32i8 = trunc <16 x i32> undef to <16 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v32i64_v32i8 = trunc <16 x i64> undef to <16 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v32i32_v32i16 = trunc <16 x i32> undef to <16 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %v32i64_v32i16 = trunc <16 x i64> undef to <16 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v32i64_v32i32 = trunc <16 x i64> undef to <16 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v32i8_v32i1 = trunc <16 x i8> undef to <16 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v32i16_v32i1 = trunc <16 x i16> undef to <16 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v32i32_v32i1 = trunc <16 x i32> undef to <16 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v32i64_v32i1 = trunc <16 x i64> undef to <16 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v32i16_v32i8 = call <32 x i8> @llvm.vp.trunc.v32i8.v32i16(<32 x i16> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_v32i32_v32i8 = call <32 x i8> @llvm.vp.trunc.v32i8.v32i32(<32 x i32> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v32i64_v32i8 = call <32 x i8> @llvm.vp.trunc.v32i8.v32i64(<32 x i64> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v32i32_v32i16 = call <32 x i16> @llvm.vp.trunc.v32i16.v32i32(<32 x i32> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %vp_v32i64_v32i16 = call <32 x i16> @llvm.vp.trunc.v32i16.v32i64(<32 x i64> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %vp_v32i64_v32i32 = call <32 x i32> @llvm.vp.trunc.v32i32.v32i64(<32 x i64> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v32i8_v32i1 = call <32 x i1> @llvm.vp.trunc.v32i1.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v32i16_v32i1 = call <32 x i1> @llvm.vp.trunc.v32i1.v32i16(<32 x i16> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v32i32_v32i1 = call <32 x i1> @llvm.vp.trunc.v32i1.v32i32(<32 x i32> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %vp_v32i64_v32i1 = call <32 x i1> @llvm.vp.trunc.v32i1.v32i64(<32 x i64> undef, <32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v64i16_v64i8 = trunc <64 x i16> undef to <64 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %v64i32_v64i8 = trunc <64 x i32> undef to <64 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v64i64_v64i8 = trunc <64 x i64> undef to <64 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %v64i32_v64i16 = trunc <64 x i32> undef to <64 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %v64i64_v64i16 = trunc <64 x i64> undef to <64 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %v64i64_v64i32 = trunc <64 x i64> undef to <64 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v64i8_v64i1 = trunc <64 x i8> undef to <64 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v64i16_v64i1 = trunc <64 x i16> undef to <64 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %v64i32_v64i1 = trunc <64 x i32> undef to <64 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %v64i64_v64i1 = trunc <64 x i64> undef to <64 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v64i16_v64i8 = call <64 x i8> @llvm.vp.trunc.v64i8.v64i16(<64 x i16> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %vp_v64i32_v64i8 = call <64 x i8> @llvm.vp.trunc.v64i8.v64i32(<64 x i32> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v64i64_v64i8 = call <64 x i8> @llvm.vp.trunc.v64i8.v64i64(<64 x i64> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %vp_v64i32_v64i16 = call <64 x i16> @llvm.vp.trunc.v64i16.v64i32(<64 x i32> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %vp_v64i64_v64i16 = call <64 x i16> @llvm.vp.trunc.v64i16.v64i64(<64 x i64> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %vp_v64i64_v64i32 = call <64 x i32> @llvm.vp.trunc.v64i32.v64i64(<64 x i64> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v64i8_v64i1 = call <64 x i1> @llvm.vp.trunc.v64i1.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v64i16_v64i1 = call <64 x i1> @llvm.vp.trunc.v64i1.v64i16(<64 x i16> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %vp_v64i32_v64i1 = call <64 x i1> @llvm.vp.trunc.v64i1.v64i32(<64 x i32> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %vp_v64i64_v64i1 = call <64 x i1> @llvm.vp.trunc.v64i1.v64i64(<64 x i64> undef, <64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %v128i16_v128i8 = trunc <128 x i16> undef to <128 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %v128i32_v128i8 = trunc <128 x i32> undef to <128 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v128i64_v128i8 = trunc <128 x i64> undef to <128 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %v128i32_v128i16 = trunc <128 x i32> undef to <128 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %v128i64_v128i16 = trunc <128 x i64> undef to <128 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %v128i64_v128i32 = trunc <128 x i64> undef to <128 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v128i8_v128i1 = trunc <128 x i8> undef to <128 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %v128i16_v128i1 = trunc <128 x i16> undef to <128 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %v128i32_v128i1 = trunc <128 x i32> undef to <128 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 135 for instruction: %v128i64_v128i1 = trunc <128 x i64> undef to <128 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %vp_v128i16_v128i8 = call <128 x i8> @llvm.vp.trunc.v128i8.v128i16(<128 x i16> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %vp_v128i32_v128i8 = call <128 x i8> @llvm.vp.trunc.v128i8.v128i32(<128 x i32> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v128i64_v128i8 = call <128 x i8> @llvm.vp.trunc.v128i8.v128i64(<128 x i64> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %vp_v128i32_v128i16 = call <128 x i16> @llvm.vp.trunc.v128i16.v128i32(<128 x i32> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %vp_v128i64_v128i16 = call <128 x i16> @llvm.vp.trunc.v128i16.v128i64(<128 x i64> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %vp_v128i64_v128i32 = call <128 x i32> @llvm.vp.trunc.v128i32.v128i64(<128 x i64> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v128i8_v128i1 = call <128 x i1> @llvm.vp.trunc.v128i1.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %vp_v128i16_v128i1 = call <128 x i1> @llvm.vp.trunc.v128i1.v128i16(<128 x i16> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %vp_v128i32_v128i1 = call <128 x i1> @llvm.vp.trunc.v128i1.v128i32(<128 x i32> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 135 for instruction: %vp_v128i64_v128i1 = call <128 x i1> @llvm.vp.trunc.v128i1.v128i64(<128 x i64> undef, <128 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %v256i16_v256i8 = trunc <256 x i16> undef to <256 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %v256i32_v256i8 = trunc <256 x i32> undef to <256 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 126 for instruction: %v256i64_v256i8 = trunc <256 x i64> undef to <256 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %v256i32_v256i16 = trunc <256 x i32> undef to <256 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 108 for instruction: %v256i64_v256i16 = trunc <256 x i64> undef to <256 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 72 for instruction: %v256i64_v256i32 = trunc <256 x i64> undef to <256 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %v256i8_v256i1 = trunc <256 x i8> undef to <256 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %v256i16_v256i1 = trunc <256 x i16> undef to <256 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 135 for instruction: %v256i32_v256i1 = trunc <256 x i32> undef to <256 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 271 for instruction: %v256i64_v256i1 = trunc <256 x i64> undef to <256 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %vp_v256i16_v256i8 = call <256 x i8> @llvm.vp.trunc.v256i8.v256i16(<256 x i16> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %vp_v256i32_v256i8 = call <256 x i8> @llvm.vp.trunc.v256i8.v256i32(<256 x i32> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 126 for instruction: %vp_v256i64_v256i8 = call <256 x i8> @llvm.vp.trunc.v256i8.v256i64(<256 x i64> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %vp_v256i32_v256i16 = call <256 x i16> @llvm.vp.trunc.v256i16.v256i32(<256 x i32> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 108 for instruction: %vp_v256i64_v256i16 = call <256 x i16> @llvm.vp.trunc.v256i16.v256i64(<256 x i64> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 72 for instruction: %vp_v256i64_v256i32 = call <256 x i32> @llvm.vp.trunc.v256i32.v256i64(<256 x i64> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %vp_v256i8_v256i1 = call <256 x i1> @llvm.vp.trunc.v256i1.v256i8(<256 x i8> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %vp_v256i16_v256i1 = call <256 x i1> @llvm.vp.trunc.v256i1.v256i16(<256 x i16> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 135 for instruction: %vp_v256i32_v256i1 = call <256 x i1> @llvm.vp.trunc.v256i1.v256i32(<256 x i32> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 271 for instruction: %vp_v256i64_v256i1 = call <256 x i1> @llvm.vp.trunc.v256i1.v256i64(<256 x i64> undef, <256 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1i16_nxv1i8 = trunc <vscale x 1 x i16> undef to <vscale x 1 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1i32_nxv1i8 = trunc <vscale x 1 x i32> undef to <vscale x 1 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv1i64_nxv1i8 = trunc <vscale x 1 x i64> undef to <vscale x 1 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1i32_nxv1i16 = trunc <vscale x 1 x i32> undef to <vscale x 1 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1i64_nxv1i16 = trunc <vscale x 1 x i64> undef to <vscale x 1 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1i64_nxv1i32 = trunc <vscale x 1 x i64> undef to <vscale x 1 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1i8_nxv1i1 = trunc <vscale x 1 x i8> undef to <vscale x 1 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1i16_nxv1i1 = trunc <vscale x 1 x i16> undef to <vscale x 1 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1i32_nxv1i1 = trunc <vscale x 1 x i32> undef to <vscale x 1 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1i64_nxv1i1 = trunc <vscale x 1 x i64> undef to <vscale x 1 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1i16_nxv1i8 = call <vscale x 1 x i8> @llvm.vp.trunc.nxv1i8.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1i32_nxv1i8 = call <vscale x 1 x i8> @llvm.vp.trunc.nxv1i8.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv1i64_nxv1i8 = call <vscale x 1 x i8> @llvm.vp.trunc.nxv1i8.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1i32_nxv1i16 = call <vscale x 1 x i16> @llvm.vp.trunc.nxv1i16.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1i64_nxv1i16 = call <vscale x 1 x i16> @llvm.vp.trunc.nxv1i16.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1i64_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.trunc.nxv1i32.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1i8_nxv1i1 = call <vscale x 1 x i1> @llvm.vp.trunc.nxv1i1.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1i16_nxv1i1 = call <vscale x 1 x i1> @llvm.vp.trunc.nxv1i1.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1i32_nxv1i1 = call <vscale x 1 x i1> @llvm.vp.trunc.nxv1i1.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1i64_nxv1i1 = call <vscale x 1 x i1> @llvm.vp.trunc.nxv1i1.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv2i16_nxv2i8 = trunc <vscale x 2 x i16> undef to <vscale x 2 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2i32_nxv2i8 = trunc <vscale x 2 x i32> undef to <vscale x 2 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv2i64_nxv2i8 = trunc <vscale x 2 x i64> undef to <vscale x 2 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv2i32_nxv2i16 = trunc <vscale x 2 x i32> undef to <vscale x 2 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2i64_nxv2i16 = trunc <vscale x 2 x i64> undef to <vscale x 2 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv2i64_nxv2i32 = trunc <vscale x 2 x i64> undef to <vscale x 2 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2i8_nxv2i1 = trunc <vscale x 2 x i8> undef to <vscale x 2 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2i16_nxv2i1 = trunc <vscale x 2 x i16> undef to <vscale x 2 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2i32_nxv2i1 = trunc <vscale x 2 x i32> undef to <vscale x 2 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv2i64_nxv2i1 = trunc <vscale x 2 x i64> undef to <vscale x 2 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv2i16_nxv2i8 = call <vscale x 2 x i8> @llvm.vp.trunc.nxv2i8.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2i32_nxv2i8 = call <vscale x 2 x i8> @llvm.vp.trunc.nxv2i8.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv2i64_nxv2i8 = call <vscale x 2 x i8> @llvm.vp.trunc.nxv2i8.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv2i32_nxv2i16 = call <vscale x 2 x i16> @llvm.vp.trunc.nxv2i16.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2i64_nxv2i16 = call <vscale x 2 x i16> @llvm.vp.trunc.nxv2i16.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv2i64_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.trunc.nxv2i32.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2i8_nxv2i1 = call <vscale x 2 x i1> @llvm.vp.trunc.nxv2i1.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2i16_nxv2i1 = call <vscale x 2 x i1> @llvm.vp.trunc.nxv2i1.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2i32_nxv2i1 = call <vscale x 2 x i1> @llvm.vp.trunc.nxv2i1.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv2i64_nxv2i1 = call <vscale x 2 x i1> @llvm.vp.trunc.nxv2i1.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv4i16_nxv4i8 = trunc <vscale x 4 x i16> undef to <vscale x 4 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4i32_nxv4i8 = trunc <vscale x 4 x i32> undef to <vscale x 4 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv4i64_nxv4i8 = trunc <vscale x 4 x i64> undef to <vscale x 4 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv4i32_nxv4i16 = trunc <vscale x 4 x i32> undef to <vscale x 4 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv4i64_nxv4i16 = trunc <vscale x 4 x i64> undef to <vscale x 4 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4i64_nxv4i32 = trunc <vscale x 4 x i64> undef to <vscale x 4 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4i8_nxv4i1 = trunc <vscale x 4 x i8> undef to <vscale x 4 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4i16_nxv4i1 = trunc <vscale x 4 x i16> undef to <vscale x 4 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv4i32_nxv4i1 = trunc <vscale x 4 x i32> undef to <vscale x 4 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv4i64_nxv4i1 = trunc <vscale x 4 x i64> undef to <vscale x 4 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv4i16_nxv4i8 = call <vscale x 4 x i8> @llvm.vp.trunc.nxv4i8.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4i32_nxv4i8 = call <vscale x 4 x i8> @llvm.vp.trunc.nxv4i8.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv4i64_nxv4i8 = call <vscale x 4 x i8> @llvm.vp.trunc.nxv4i8.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv4i32_nxv4i16 = call <vscale x 4 x i16> @llvm.vp.trunc.nxv4i16.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv4i64_nxv4i16 = call <vscale x 4 x i16> @llvm.vp.trunc.nxv4i16.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4i64_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.trunc.nxv4i32.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4i8_nxv4i1 = call <vscale x 4 x i1> @llvm.vp.trunc.nxv4i1.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4i16_nxv4i1 = call <vscale x 4 x i1> @llvm.vp.trunc.nxv4i1.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv4i32_nxv4i1 = call <vscale x 4 x i1> @llvm.vp.trunc.nxv4i1.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv4i64_nxv4i1 = call <vscale x 4 x i1> @llvm.vp.trunc.nxv4i1.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv8i16_nxv8i8 = trunc <vscale x 8 x i16> undef to <vscale x 8 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv8i32_nxv8i8 = trunc <vscale x 8 x i32> undef to <vscale x 8 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %nxv8i64_nxv8i8 = trunc <vscale x 8 x i64> undef to <vscale x 8 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv8i32_nxv8i16 = trunc <vscale x 8 x i32> undef to <vscale x 8 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %nxv8i64_nxv8i16 = trunc <vscale x 8 x i64> undef to <vscale x 8 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv8i64_nxv8i32 = trunc <vscale x 8 x i64> undef to <vscale x 8 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv8i8_nxv8i1 = trunc <vscale x 8 x i8> undef to <vscale x 8 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv8i16_nxv8i1 = trunc <vscale x 8 x i16> undef to <vscale x 8 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv8i32_nxv8i1 = trunc <vscale x 8 x i32> undef to <vscale x 8 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %nxv8i64_nxv8i1 = trunc <vscale x 8 x i64> undef to <vscale x 8 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv8i16_nxv8i8 = call <vscale x 8 x i8> @llvm.vp.trunc.nxv8i8.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv8i32_nxv8i8 = call <vscale x 8 x i8> @llvm.vp.trunc.nxv8i8.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_nxv8i64_nxv8i8 = call <vscale x 8 x i8> @llvm.vp.trunc.nxv8i8.nxv8i64(<vscale x 8 x i64> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv8i32_nxv8i16 = call <vscale x 8 x i16> @llvm.vp.trunc.nxv8i16.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_nxv8i64_nxv8i16 = call <vscale x 8 x i16> @llvm.vp.trunc.nxv8i16.nxv8i64(<vscale x 8 x i64> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv8i64_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.trunc.nxv8i32.nxv8i64(<vscale x 8 x i64> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv8i8_nxv8i1 = call <vscale x 8 x i1> @llvm.vp.trunc.nxv8i1.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv8i16_nxv8i1 = call <vscale x 8 x i1> @llvm.vp.trunc.nxv8i1.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv8i32_nxv8i1 = call <vscale x 8 x i1> @llvm.vp.trunc.nxv8i1.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_nxv8i64_nxv8i1 = call <vscale x 8 x i1> @llvm.vp.trunc.nxv8i1.nxv8i64(<vscale x 8 x i64> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv16i16_nxv16i8 = trunc <vscale x 16 x i16> undef to <vscale x 16 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %nxv16i32_nxv16i8 = trunc <vscale x 16 x i32> undef to <vscale x 16 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %nxv16i64_nxv16i8 = trunc <vscale x 16 x i64> undef to <vscale x 16 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv16i32_nxv16i16 = trunc <vscale x 16 x i32> undef to <vscale x 16 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %nxv16i64_nxv16i16 = trunc <vscale x 16 x i64> undef to <vscale x 16 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %nxv16i64_nxv16i32 = trunc <vscale x 16 x i64> undef to <vscale x 16 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv16i8_nxv16i1 = trunc <vscale x 16 x i8> undef to <vscale x 16 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv16i16_nxv16i1 = trunc <vscale x 16 x i16> undef to <vscale x 16 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %nxv16i32_nxv16i1 = trunc <vscale x 16 x i32> undef to <vscale x 16 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %nxv16i64_nxv16i1 = trunc <vscale x 16 x i64> undef to <vscale x 16 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv16i16_nxv16i8 = call <vscale x 16 x i8> @llvm.vp.trunc.nxv16i8.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_nxv16i32_nxv16i8 = call <vscale x 16 x i8> @llvm.vp.trunc.nxv16i8.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_nxv16i64_nxv16i8 = call <vscale x 16 x i8> @llvm.vp.trunc.nxv16i8.nxv16i64(<vscale x 16 x i64> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv16i32_nxv16i16 = call <vscale x 16 x i16> @llvm.vp.trunc.nxv16i16.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %vp_nxv16i64_nxv16i16 = call <vscale x 16 x i16> @llvm.vp.trunc.nxv16i16.nxv16i64(<vscale x 16 x i64> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %vp_nxv16i64_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.trunc.nxv16i32.nxv16i64(<vscale x 16 x i64> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv16i8_nxv16i1 = call <vscale x 16 x i1> @llvm.vp.trunc.nxv16i1.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv16i16_nxv16i1 = call <vscale x 16 x i1> @llvm.vp.trunc.nxv16i1.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_nxv16i32_nxv16i1 = call <vscale x 16 x i1> @llvm.vp.trunc.nxv16i1.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %vp_nxv16i64_nxv16i1 = call <vscale x 16 x i1> @llvm.vp.trunc.nxv16i1.nxv16i64(<vscale x 16 x i64> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv32i16_nxv32i8 = trunc <vscale x 32 x i16> undef to <vscale x 32 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %nxv32i32_nxv32i8 = trunc <vscale x 32 x i32> undef to <vscale x 32 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %nxv32i64_nxv32i8 = trunc <vscale x 32 x i64> undef to <vscale x 32 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %nxv32i32_nxv32i16 = trunc <vscale x 32 x i32> undef to <vscale x 32 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %nxv32i64_nxv32i16 = trunc <vscale x 32 x i64> undef to <vscale x 32 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %nxv32i64_nxv32i32 = trunc <vscale x 32 x i64> undef to <vscale x 32 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv32i8_nxv32i1 = trunc <vscale x 32 x i8> undef to <vscale x 32 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %nxv32i16_nxv32i1 = trunc <vscale x 32 x i16> undef to <vscale x 32 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %nxv32i32_nxv32i1 = trunc <vscale x 32 x i32> undef to <vscale x 32 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %nxv32i64_nxv32i1 = trunc <vscale x 32 x i64> undef to <vscale x 32 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv32i16_nxv32i8 = call <vscale x 32 x i8> @llvm.vp.trunc.nxv32i8.nxv32i16(<vscale x 32 x i16> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %vp_nxv32i32_nxv32i8 = call <vscale x 32 x i8> @llvm.vp.trunc.nxv32i8.nxv32i32(<vscale x 32 x i32> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_nxv32i64_nxv32i8 = call <vscale x 32 x i8> @llvm.vp.trunc.nxv32i8.nxv32i64(<vscale x 32 x i64> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %vp_nxv32i32_nxv32i16 = call <vscale x 32 x i16> @llvm.vp.trunc.nxv32i16.nxv32i32(<vscale x 32 x i32> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %vp_nxv32i64_nxv32i16 = call <vscale x 32 x i16> @llvm.vp.trunc.nxv32i16.nxv32i64(<vscale x 32 x i64> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %vp_nxv32i64_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.trunc.nxv32i32.nxv32i64(<vscale x 32 x i64> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv32i8_nxv32i1 = call <vscale x 32 x i1> @llvm.vp.trunc.nxv32i1.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_nxv32i16_nxv32i1 = call <vscale x 32 x i1> @llvm.vp.trunc.nxv32i1.nxv32i16(<vscale x 32 x i16> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %vp_nxv32i32_nxv32i1 = call <vscale x 32 x i1> @llvm.vp.trunc.nxv32i1.nxv32i32(<vscale x 32 x i32> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %vp_nxv32i64_nxv32i1 = call <vscale x 32 x i1> @llvm.vp.trunc.nxv32i1.nxv32i64(<vscale x 32 x i64> undef, <vscale x 32 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %nxv64i16_nxv64i8 = trunc <vscale x 64 x i16> undef to <vscale x 64 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %nxv64i32_nxv64i8 = trunc <vscale x 64 x i32> undef to <vscale x 64 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %nxv64i64_nxv64i8 = trunc <vscale x 64 x i64> undef to <vscale x 64 x i8>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %nxv64i32_nxv64i16 = trunc <vscale x 64 x i32> undef to <vscale x 64 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %nxv64i64_nxv64i16 = trunc <vscale x 64 x i64> undef to <vscale x 64 x i16>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %nxv64i64_nxv64i32 = trunc <vscale x 64 x i64> undef to <vscale x 64 x i32>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %nxv64i8_nxv64i1 = trunc <vscale x 64 x i8> undef to <vscale x 64 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %nxv64i16_nxv64i1 = trunc <vscale x 64 x i16> undef to <vscale x 64 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %nxv64i32_nxv64i1 = trunc <vscale x 64 x i32> undef to <vscale x 64 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 135 for instruction: %nxv64i64_nxv64i1 = trunc <vscale x 64 x i64> undef to <vscale x 64 x i1>
; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %vp_nxv64i16_nxv64i8 = call <vscale x 64 x i8> @llvm.vp.trunc.nxv64i8.nxv64i16(<vscale x 64 x i16> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %vp_nxv64i32_nxv64i8 = call <vscale x 64 x i8> @llvm.vp.trunc.nxv64i8.nxv64i32(<vscale x 64 x i32> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_nxv64i64_nxv64i8 = call <vscale x 64 x i8> @llvm.vp.trunc.nxv64i8.nxv64i64(<vscale x 64 x i64> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %vp_nxv64i32_nxv64i16 = call <vscale x 64 x i16> @llvm.vp.trunc.nxv64i16.nxv64i32(<vscale x 64 x i32> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %vp_nxv64i64_nxv64i16 = call <vscale x 64 x i16> @llvm.vp.trunc.nxv64i16.nxv64i64(<vscale x 64 x i64> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %vp_nxv64i64_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.trunc.nxv64i32.nxv64i64(<vscale x 64 x i64> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_nxv64i8_nxv64i1 = call <vscale x 64 x i1> @llvm.vp.trunc.nxv64i1.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %vp_nxv64i16_nxv64i1 = call <vscale x 64 x i1> @llvm.vp.trunc.nxv64i1.nxv64i16(<vscale x 64 x i16> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %vp_nxv64i32_nxv64i1 = call <vscale x 64 x i1> @llvm.vp.trunc.nxv64i1.nxv64i32(<vscale x 64 x i32> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 135 for instruction: %vp_nxv64i64_nxv64i1 = call <vscale x 64 x i1> @llvm.vp.trunc.nxv64i1.nxv64i64(<vscale x 64 x i64> undef, <vscale x 64 x i1> undef, i32 undef)
; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;

  %v2i16_v2i2 = trunc <2 x i16> undef to <2 x i2>
  %v2i16_v2i4 = trunc <2 x i16> undef to <2 x i4>
  %v2i16_v2i6 = trunc <2 x i16> undef to <2 x i6>
  %v2i4_v2i2 = trunc <2 x i4> undef to <2 x i2>
  %v2i6_v2i4 = trunc <2 x i6> undef to <2 x i4>

  %vp_v2i16_v2i2 = call <2 x i2> @llvm.vp.trunc.v2i2.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
  %vp_v2i16_v2i4 = call <2 x i4> @llvm.vp.trunc.v2i4.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
  %vp_v2i16_v2i6 = call <2 x i6> @llvm.vp.trunc.v2i6.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
  %vp_v2i4_v2i2 = call <2 x i2> @llvm.vp.trunc.v2i2.v2i4(<2 x i4> undef, <2 x i1> undef, i32 undef)
  %vp_v2i6_v2i4 = call <2 x i4> @llvm.vp.trunc.v2i4.v2i6(<2 x i6> undef, <2 x i1> undef, i32 undef)

  %v2i16_v2i8 = trunc <2 x i16> undef to <2 x i8>
  %v2i32_v2i8 = trunc <2 x i32> undef to <2 x i8>
  %v2i64_v2i8 = trunc <2 x i64> undef to <2 x i8>
  %v2i32_v2i16 = trunc <2 x i32> undef to <2 x i16>
  %v2i64_v2i16 = trunc <2 x i64> undef to <2 x i16>
  %v2i64_v2i32 = trunc <2 x i64> undef to <2 x i32>
  %v2i8_v2i1 = trunc <2 x i8> undef to <2 x i1>
  %v2i16_v2i1 = trunc <2 x i16> undef to <2 x i1>
  %v2i32_v2i1 = trunc <2 x i32> undef to <2 x i1>
  %v2i64_v2i1 = trunc <2 x i64> undef to <2 x i1>

  %vp_v2i16_v2i8 = call <2 x i8> @llvm.vp.trunc.v2i8.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
  %vp_v2i32_v2i8 = call <2 x i8> @llvm.vp.trunc.v2i8.v2i32(<2 x i32> undef, <2 x i1> undef, i32 undef)
  %vp_v2i64_v2i8 = call <2 x i8> @llvm.vp.trunc.v2i8.v2i64(<2 x i64> undef, <2 x i1> undef, i32 undef)
  %vp_v2i32_v2i16 = call <2 x i16> @llvm.vp.trunc.v2i16.v2i32(<2 x i32> undef, <2 x i1> undef, i32 undef)
  %vp_v2i64_v2i16 = call <2 x i16> @llvm.vp.trunc.v2i16.v2i64(<2 x i64> undef, <2 x i1> undef, i32 undef)
  %vp_v2i64_v2i32 = call <2 x i32> @llvm.vp.trunc.v2i32.v2i64(<2 x i64> undef, <2 x i1> undef, i32 undef)
  %vp_v2i8_v2i1 = call <2 x i1> @llvm.vp.trunc.v2i1.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
  %vp_v2i16_v2i1 = call <2 x i1> @llvm.vp.trunc.v2i1.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
  %vp_v2i32_v2i1 = call <2 x i1> @llvm.vp.trunc.v2i1.v2i32(<2 x i32> undef, <2 x i1> undef, i32 undef)
  %vp_v2i64_v2i1 = call <2 x i1> @llvm.vp.trunc.v2i1.v2i64(<2 x i64> undef, <2 x i1> undef, i32 undef)

  %v4i16_v4i8 = trunc <4 x i16> undef to <4 x i8>
  %v4i32_v4i8 = trunc <4 x i32> undef to <4 x i8>
  %v4i64_v4i8 = trunc <4 x i64> undef to <4 x i8>
  %v4i32_v4i16 = trunc <4 x i32> undef to <4 x i16>
  %v4i64_v4i16 = trunc <4 x i64> undef to <4 x i16>
  %v4i64_v4i32 = trunc <4 x i64> undef to <4 x i32>
  %v4i8_v4i1 = trunc <4 x i8> undef to <4 x i1>
  %v4i16_v4i1 = trunc <4 x i16> undef to <4 x i1>
  %v4i32_v4i1 = trunc <4 x i32> undef to <4 x i1>
  %v4i64_v4i1 = trunc <4 x i64> undef to <4 x i1>

  %vp_v4i16_v4i8 = call <4 x i8> @llvm.vp.trunc.v4i8.v4i16(<4 x i16> undef, <4 x i1> undef, i32 undef)
  %vp_v4i32_v4i8 = call <4 x i8> @llvm.vp.trunc.v4i8.v4i32(<4 x i32> undef, <4 x i1> undef, i32 undef)
  %vp_v4i64_v4i8 = call <4 x i8> @llvm.vp.trunc.v4i8.v4i64(<4 x i64> undef, <4 x i1> undef, i32 undef)
  %vp_v4i32_v4i16 = call <4 x i16> @llvm.vp.trunc.v4i16.v4i32(<4 x i32> undef, <4 x i1> undef, i32 undef)
  %vp_v4i64_v4i16 = call <4 x i16> @llvm.vp.trunc.v4i16.v4i64(<4 x i64> undef, <4 x i1> undef, i32 undef)
  %vp_v4i64_v4i32 = call <4 x i32> @llvm.vp.trunc.v4i32.v4i64(<4 x i64> undef, <4 x i1> undef, i32 undef)
  %vp_v4i8_v4i1 = call <4 x i1> @llvm.vp.trunc.v4i1.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
  %vp_v4i16_v4i1 = call <4 x i1> @llvm.vp.trunc.v4i1.v4i16(<4 x i16> undef, <4 x i1> undef, i32 undef)
  %vp_v4i32_v4i1 = call <4 x i1> @llvm.vp.trunc.v4i1.v4i32(<4 x i32> undef, <4 x i1> undef, i32 undef)
  %vp_v4i64_v4i1 = call <4 x i1> @llvm.vp.trunc.v4i1.v4i64(<4 x i64> undef, <4 x i1> undef, i32 undef)

  %v8i16_v8i8 = trunc <8 x i16> undef to <8 x i8>
  %v8i32_v8i8 = trunc <8 x i32> undef to <8 x i8>
  %v8i64_v8i8 = trunc <8 x i64> undef to <8 x i8>
  %v8i32_v8i16 = trunc <8 x i32> undef to <8 x i16>
  %v8i64_v8i16 = trunc <8 x i64> undef to <8 x i16>
  %v8i64_v8i32 = trunc <8 x i64> undef to <8 x i32>
  %v8i8_v8i1 = trunc <8 x i8> undef to <8 x i1>
  %v8i16_v8i1 = trunc <8 x i16> undef to <8 x i1>
  %v8i32_v8i1 = trunc <8 x i32> undef to <8 x i1>
  %v8i64_v8i1 = trunc <8 x i64> undef to <8 x i1>

  %vp_v8i16_v8i8 = call <8 x i8> @llvm.vp.trunc.v8i8.v8i16(<8 x i16> undef, <8 x i1> undef, i32 undef)
  %vp_v8i32_v8i8 = call <8 x i8> @llvm.vp.trunc.v8i8.v8i32(<8 x i32> undef, <8 x i1> undef, i32 undef)
  %vp_v8i64_v8i8 = call <8 x i8> @llvm.vp.trunc.v8i8.v8i64(<8 x i64> undef, <8 x i1> undef, i32 undef)
  %vp_v8i32_v8i16 = call <8 x i16> @llvm.vp.trunc.v8i16.v8i32(<8 x i32> undef, <8 x i1> undef, i32 undef)
  %vp_v8i64_v8i16 = call <8 x i16> @llvm.vp.trunc.v8i16.v8i64(<8 x i64> undef, <8 x i1> undef, i32 undef)
  %vp_v8i64_v8i32 = call <8 x i32> @llvm.vp.trunc.v8i32.v8i64(<8 x i64> undef, <8 x i1> undef, i32 undef)
  %vp_v8i8_v8i1 = call <8 x i1> @llvm.vp.trunc.v8i1.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
  %vp_v8i16_v8i1 = call <8 x i1> @llvm.vp.trunc.v8i1.v8i16(<8 x i16> undef, <8 x i1> undef, i32 undef)
  %vp_v8i32_v8i1 = call <8 x i1> @llvm.vp.trunc.v8i1.v8i32(<8 x i32> undef, <8 x i1> undef, i32 undef)
  %vp_v8i64_v8i1 = call <8 x i1> @llvm.vp.trunc.v8i1.v8i64(<8 x i64> undef, <8 x i1> undef, i32 undef)

  %v16i16_v16i8 = trunc <2 x i16> undef to <2 x i8>
  %v16i32_v16i8 = trunc <2 x i32> undef to <2 x i8>
  %v16i64_v16i8 = trunc <2 x i64> undef to <2 x i8>
  %v16i32_v16i16 = trunc <2 x i32> undef to <2 x i16>
  %v16i64_v16i16 = trunc <2 x i64> undef to <2 x i16>
  %v16i64_v16i32 = trunc <2 x i64> undef to <2 x i32>
  %v16i8_v16i1 = trunc <2 x i8> undef to <2 x i1>
  %v16i16_v16i1 = trunc <2 x i16> undef to <2 x i1>
  %v16i32_v16i1 = trunc <2 x i32> undef to <2 x i1>
  %v16i64_v16i1 = trunc <2 x i64> undef to <2 x i1>

  %vp_v16i16_v16i8 = call <16 x i8> @llvm.vp.trunc.v16i8.v16i16(<16 x i16> undef, <16 x i1> undef, i32 undef)
  %vp_v16i32_v16i8 = call <16 x i8> @llvm.vp.trunc.v16i8.v16i32(<16 x i32> undef, <16 x i1> undef, i32 undef)
  %vp_v16i64_v16i8 = call <16 x i8> @llvm.vp.trunc.v16i8.v16i64(<16 x i64> undef, <16 x i1> undef, i32 undef)
  %vp_v16i32_v16i16 = call <16 x i16> @llvm.vp.trunc.v16i16.v16i32(<16 x i32> undef, <16 x i1> undef, i32 undef)
  %vp_v16i64_v16i16 = call <16 x i16> @llvm.vp.trunc.v16i16.v16i64(<16 x i64> undef, <16 x i1> undef, i32 undef)
  %vp_v16i64_v16i32 = call <16 x i32> @llvm.vp.trunc.v16i32.v16i64(<16 x i64> undef, <16 x i1> undef, i32 undef)
  %vp_v16i8_v16i1 = call <16 x i1> @llvm.vp.trunc.v16i1.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
  %vp_v16i16_v16i1 = call <16 x i1> @llvm.vp.trunc.v16i1.v16i16(<16 x i16> undef, <16 x i1> undef, i32 undef)
  %vp_v16i32_v16i1 = call <16 x i1> @llvm.vp.trunc.v16i1.v16i32(<16 x i32> undef, <16 x i1> undef, i32 undef)
  %vp_v16i64_v16i1 = call <16 x i1> @llvm.vp.trunc.v16i1.v16i64(<16 x i64> undef, <16 x i1> undef, i32 undef)

  %v32i16_v32i8 = trunc <16 x i16> undef to <16 x i8>
  %v32i32_v32i8 = trunc <16 x i32> undef to <16 x i8>
  %v32i64_v32i8 = trunc <16 x i64> undef to <16 x i8>
  %v32i32_v32i16 = trunc <16 x i32> undef to <16 x i16>
  %v32i64_v32i16 = trunc <16 x i64> undef to <16 x i16>
  %v32i64_v32i32 = trunc <16 x i64> undef to <16 x i32>
  %v32i8_v32i1 = trunc <16 x i8> undef to <16 x i1>
  %v32i16_v32i1 = trunc <16 x i16> undef to <16 x i1>
  %v32i32_v32i1 = trunc <16 x i32> undef to <16 x i1>
  %v32i64_v32i1 = trunc <16 x i64> undef to <16 x i1>

  %vp_v32i16_v32i8 = call <32 x i8> @llvm.vp.trunc.v32i8.v32i16(<32 x i16> undef, <32 x i1> undef, i32 undef)
  %vp_v32i32_v32i8 = call <32 x i8> @llvm.vp.trunc.v32i8.v32i32(<32 x i32> undef, <32 x i1> undef, i32 undef)
  %vp_v32i64_v32i8 = call <32 x i8> @llvm.vp.trunc.v32i8.v32i64(<32 x i64> undef, <32 x i1> undef, i32 undef)
  %vp_v32i32_v32i16 = call <32 x i16> @llvm.vp.trunc.v32i16.v32i32(<32 x i32> undef, <32 x i1> undef, i32 undef)
  %vp_v32i64_v32i16 = call <32 x i16> @llvm.vp.trunc.v32i16.v32i64(<32 x i64> undef, <32 x i1> undef, i32 undef)
  %vp_v32i64_v32i32 = call <32 x i32> @llvm.vp.trunc.v32i32.v32i64(<32 x i64> undef, <32 x i1> undef, i32 undef)
  %vp_v32i8_v32i1 = call <32 x i1> @llvm.vp.trunc.v32i1.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
  %vp_v32i16_v32i1 = call <32 x i1> @llvm.vp.trunc.v32i1.v32i16(<32 x i16> undef, <32 x i1> undef, i32 undef)
  %vp_v32i32_v32i1 = call <32 x i1> @llvm.vp.trunc.v32i1.v32i32(<32 x i32> undef, <32 x i1> undef, i32 undef)
  %vp_v32i64_v32i1 = call <32 x i1> @llvm.vp.trunc.v32i1.v32i64(<32 x i64> undef, <32 x i1> undef, i32 undef)

  %v64i16_v64i8 = trunc <64 x i16> undef to <64 x i8>
  %v64i32_v64i8 = trunc <64 x i32> undef to <64 x i8>
  %v64i64_v64i8 = trunc <64 x i64> undef to <64 x i8>
  %v64i32_v64i16 = trunc <64 x i32> undef to <64 x i16>
  %v64i64_v64i16 = trunc <64 x i64> undef to <64 x i16>
  %v64i64_v64i32 = trunc <64 x i64> undef to <64 x i32>
  %v64i8_v64i1 = trunc <64 x i8> undef to <64 x i1>
  %v64i16_v64i1 = trunc <64 x i16> undef to <64 x i1>
  %v64i32_v64i1 = trunc <64 x i32> undef to <64 x i1>
  %v64i64_v64i1 = trunc <64 x i64> undef to <64 x i1>

  %vp_v64i16_v64i8 = call <64 x i8> @llvm.vp.trunc.v64i8.v64i16(<64 x i16> undef, <64 x i1> undef, i32 undef)
  %vp_v64i32_v64i8 = call <64 x i8> @llvm.vp.trunc.v64i8.v64i32(<64 x i32> undef, <64 x i1> undef, i32 undef)
  %vp_v64i64_v64i8 = call <64 x i8> @llvm.vp.trunc.v64i8.v64i64(<64 x i64> undef, <64 x i1> undef, i32 undef)
  %vp_v64i32_v64i16 = call <64 x i16> @llvm.vp.trunc.v64i16.v64i32(<64 x i32> undef, <64 x i1> undef, i32 undef)
  %vp_v64i64_v64i16 = call <64 x i16> @llvm.vp.trunc.v64i16.v64i64(<64 x i64> undef, <64 x i1> undef, i32 undef)
  %vp_v64i64_v64i32 = call <64 x i32> @llvm.vp.trunc.v64i32.v64i64(<64 x i64> undef, <64 x i1> undef, i32 undef)
  %vp_v64i8_v64i1 = call <64 x i1> @llvm.vp.trunc.v64i1.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
  %vp_v64i16_v64i1 = call <64 x i1> @llvm.vp.trunc.v64i1.v64i16(<64 x i16> undef, <64 x i1> undef, i32 undef)
  %vp_v64i32_v64i1 = call <64 x i1> @llvm.vp.trunc.v64i1.v64i32(<64 x i32> undef, <64 x i1> undef, i32 undef)
  %vp_v64i64_v64i1 = call <64 x i1> @llvm.vp.trunc.v64i1.v64i64(<64 x i64> undef, <64 x i1> undef, i32 undef)

  %v128i16_v128i8 = trunc <128 x i16> undef to <128 x i8>
  %v128i32_v128i8 = trunc <128 x i32> undef to <128 x i8>
  %v128i64_v128i8 = trunc <128 x i64> undef to <128 x i8>
  %v128i32_v128i16 = trunc <128 x i32> undef to <128 x i16>
  %v128i64_v128i16 = trunc <128 x i64> undef to <128 x i16>
  %v128i64_v128i32 = trunc <128 x i64> undef to <128 x i32>
  %v128i8_v128i1 = trunc <128 x i8> undef to <128 x i1>
  %v128i16_v128i1 = trunc <128 x i16> undef to <128 x i1>
  %v128i32_v128i1 = trunc <128 x i32> undef to <128 x i1>
  %v128i64_v128i1 = trunc <128 x i64> undef to <128 x i1>

  %vp_v128i16_v128i8 = call <128 x i8> @llvm.vp.trunc.v128i8.v128i16(<128 x i16> undef, <128 x i1> undef, i32 undef)
  %vp_v128i32_v128i8 = call <128 x i8> @llvm.vp.trunc.v128i8.v128i32(<128 x i32> undef, <128 x i1> undef, i32 undef)
  %vp_v128i64_v128i8 = call <128 x i8> @llvm.vp.trunc.v128i8.v128i64(<128 x i64> undef, <128 x i1> undef, i32 undef)
  %vp_v128i32_v128i16 = call <128 x i16> @llvm.vp.trunc.v128i16.v128i32(<128 x i32> undef, <128 x i1> undef, i32 undef)
  %vp_v128i64_v128i16 = call <128 x i16> @llvm.vp.trunc.v128i16.v128i64(<128 x i64> undef, <128 x i1> undef, i32 undef)
  %vp_v128i64_v128i32 = call <128 x i32> @llvm.vp.trunc.v128i32.v128i64(<128 x i64> undef, <128 x i1> undef, i32 undef)
  %vp_v128i8_v128i1 = call <128 x i1> @llvm.vp.trunc.v128i1.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
  %vp_v128i16_v128i1 = call <128 x i1> @llvm.vp.trunc.v128i1.v128i16(<128 x i16> undef, <128 x i1> undef, i32 undef)
  %vp_v128i32_v128i1 = call <128 x i1> @llvm.vp.trunc.v128i1.v128i32(<128 x i32> undef, <128 x i1> undef, i32 undef)
  %vp_v128i64_v128i1 = call <128 x i1> @llvm.vp.trunc.v128i1.v128i64(<128 x i64> undef, <128 x i1> undef, i32 undef)

  %v256i16_v256i8 = trunc <256 x i16> undef to <256 x i8>
  %v256i32_v256i8 = trunc <256 x i32> undef to <256 x i8>
  %v256i64_v256i8 = trunc <256 x i64> undef to <256 x i8>
  %v256i32_v256i16 = trunc <256 x i32> undef to <256 x i16>
  %v256i64_v256i16 = trunc <256 x i64> undef to <256 x i16>
  %v256i64_v256i32 = trunc <256 x i64> undef to <256 x i32>
  %v256i8_v256i1 = trunc <256 x i8> undef to <256 x i1>
  %v256i16_v256i1 = trunc <256 x i16> undef to <256 x i1>
  %v256i32_v256i1 = trunc <256 x i32> undef to <256 x i1>
  %v256i64_v256i1 = trunc <256 x i64> undef to <256 x i1>

  %vp_v256i16_v256i8 = call <256 x i8> @llvm.vp.trunc.v256i8.v256i16(<256 x i16> undef, <256 x i1> undef, i32 undef)
  %vp_v256i32_v256i8 = call <256 x i8> @llvm.vp.trunc.v256i8.v256i32(<256 x i32> undef, <256 x i1> undef, i32 undef)
  %vp_v256i64_v256i8 = call <256 x i8> @llvm.vp.trunc.v256i8.v256i64(<256 x i64> undef, <256 x i1> undef, i32 undef)
  %vp_v256i32_v256i16 = call <256 x i16> @llvm.vp.trunc.v256i16.v256i32(<256 x i32> undef, <256 x i1> undef, i32 undef)
  %vp_v256i64_v256i16 = call <256 x i16> @llvm.vp.trunc.v256i16.v256i64(<256 x i64> undef, <256 x i1> undef, i32 undef)
  %vp_v256i64_v256i32 = call <256 x i32> @llvm.vp.trunc.v256i32.v256i64(<256 x i64> undef, <256 x i1> undef, i32 undef)
  %vp_v256i8_v256i1 = call <256 x i1> @llvm.vp.trunc.v256i1.v256i8(<256 x i8> undef, <256 x i1> undef, i32 undef)
  %vp_v256i16_v256i1 = call <256 x i1> @llvm.vp.trunc.v256i1.v256i16(<256 x i16> undef, <256 x i1> undef, i32 undef)
  %vp_v256i32_v256i1 = call <256 x i1> @llvm.vp.trunc.v256i1.v256i32(<256 x i32> undef, <256 x i1> undef, i32 undef)
  %vp_v256i64_v256i1 = call <256 x i1> @llvm.vp.trunc.v256i1.v256i64(<256 x i64> undef, <256 x i1> undef, i32 undef)

  %nxv1i16_nxv1i8 = trunc <vscale x 1 x i16> undef to <vscale x 1 x i8>
  %nxv1i32_nxv1i8 = trunc <vscale x 1 x i32> undef to <vscale x 1 x i8>
  %nxv1i64_nxv1i8 = trunc <vscale x 1 x i64> undef to <vscale x 1 x i8>
  %nxv1i32_nxv1i16 = trunc <vscale x 1 x i32> undef to <vscale x 1 x i16>
  %nxv1i64_nxv1i16 = trunc <vscale x 1 x i64> undef to <vscale x 1 x i16>
  %nxv1i64_nxv1i32 = trunc <vscale x 1 x i64> undef to <vscale x 1 x i32>
  %nxv1i8_nxv1i1 = trunc <vscale x 1 x i8> undef to <vscale x 1 x i1>
  %nxv1i16_nxv1i1 = trunc <vscale x 1 x i16> undef to <vscale x 1 x i1>
  %nxv1i32_nxv1i1 = trunc <vscale x 1 x i32> undef to <vscale x 1 x i1>
  %nxv1i64_nxv1i1 = trunc <vscale x 1 x i64> undef to <vscale x 1 x i1>

  %vp_nxv1i16_nxv1i8 = call <vscale x 1 x i8> @llvm.vp.trunc.nxv1i8.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i32_nxv1i8 = call <vscale x 1 x i8> @llvm.vp.trunc.nxv1i8.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i64_nxv1i8 = call <vscale x 1 x i8> @llvm.vp.trunc.nxv1i8.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i32_nxv1i16 = call <vscale x 1 x i16> @llvm.vp.trunc.nxv1i16.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i64_nxv1i16 = call <vscale x 1 x i16> @llvm.vp.trunc.nxv1i16.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i64_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.trunc.nxv1i32.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i8_nxv1i1 = call <vscale x 1 x i1> @llvm.vp.trunc.nxv1i1.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i16_nxv1i1 = call <vscale x 1 x i1> @llvm.vp.trunc.nxv1i1.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i32_nxv1i1 = call <vscale x 1 x i1> @llvm.vp.trunc.nxv1i1.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1i64_nxv1i1 = call <vscale x 1 x i1> @llvm.vp.trunc.nxv1i1.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)

  %nxv2i16_nxv2i8 = trunc <vscale x 2 x i16> undef to <vscale x 2 x i8>
  %nxv2i32_nxv2i8 = trunc <vscale x 2 x i32> undef to <vscale x 2 x i8>
  %nxv2i64_nxv2i8 = trunc <vscale x 2 x i64> undef to <vscale x 2 x i8>
  %nxv2i32_nxv2i16 = trunc <vscale x 2 x i32> undef to <vscale x 2 x i16>
  %nxv2i64_nxv2i16 = trunc <vscale x 2 x i64> undef to <vscale x 2 x i16>
  %nxv2i64_nxv2i32 = trunc <vscale x 2 x i64> undef to <vscale x 2 x i32>
  %nxv2i8_nxv2i1 = trunc <vscale x 2 x i8> undef to <vscale x 2 x i1>
  %nxv2i16_nxv2i1 = trunc <vscale x 2 x i16> undef to <vscale x 2 x i1>
  %nxv2i32_nxv2i1 = trunc <vscale x 2 x i32> undef to <vscale x 2 x i1>
  %nxv2i64_nxv2i1 = trunc <vscale x 2 x i64> undef to <vscale x 2 x i1>

  %vp_nxv2i16_nxv2i8 = call <vscale x 2 x i8> @llvm.vp.trunc.nxv2i8.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i32_nxv2i8 = call <vscale x 2 x i8> @llvm.vp.trunc.nxv2i8.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i64_nxv2i8 = call <vscale x 2 x i8> @llvm.vp.trunc.nxv2i8.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i32_nxv2i16 = call <vscale x 2 x i16> @llvm.vp.trunc.nxv2i16.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i64_nxv2i16 = call <vscale x 2 x i16> @llvm.vp.trunc.nxv2i16.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i64_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.trunc.nxv2i32.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i8_nxv2i1 = call <vscale x 2 x i1> @llvm.vp.trunc.nxv2i1.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i16_nxv2i1 = call <vscale x 2 x i1> @llvm.vp.trunc.nxv2i1.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i32_nxv2i1 = call <vscale x 2 x i1> @llvm.vp.trunc.nxv2i1.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2i64_nxv2i1 = call <vscale x 2 x i1> @llvm.vp.trunc.nxv2i1.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i1> undef, i32 undef)

  %nxv4i16_nxv4i8 = trunc <vscale x 4 x i16> undef to <vscale x 4 x i8>
  %nxv4i32_nxv4i8 = trunc <vscale x 4 x i32> undef to <vscale x 4 x i8>
  %nxv4i64_nxv4i8 = trunc <vscale x 4 x i64> undef to <vscale x 4 x i8>
  %nxv4i32_nxv4i16 = trunc <vscale x 4 x i32> undef to <vscale x 4 x i16>
  %nxv4i64_nxv4i16 = trunc <vscale x 4 x i64> undef to <vscale x 4 x i16>
  %nxv4i64_nxv4i32 = trunc <vscale x 4 x i64> undef to <vscale x 4 x i32>
  %nxv4i8_nxv4i1 = trunc <vscale x 4 x i8> undef to <vscale x 4 x i1>
  %nxv4i16_nxv4i1 = trunc <vscale x 4 x i16> undef to <vscale x 4 x i1>
  %nxv4i32_nxv4i1 = trunc <vscale x 4 x i32> undef to <vscale x 4 x i1>
  %nxv4i64_nxv4i1 = trunc <vscale x 4 x i64> undef to <vscale x 4 x i1>

  %vp_nxv4i16_nxv4i8 = call <vscale x 4 x i8> @llvm.vp.trunc.nxv4i8.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i32_nxv4i8 = call <vscale x 4 x i8> @llvm.vp.trunc.nxv4i8.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i64_nxv4i8 = call <vscale x 4 x i8> @llvm.vp.trunc.nxv4i8.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i32_nxv4i16 = call <vscale x 4 x i16> @llvm.vp.trunc.nxv4i16.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i64_nxv4i16 = call <vscale x 4 x i16> @llvm.vp.trunc.nxv4i16.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i64_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.trunc.nxv4i32.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i8_nxv4i1 = call <vscale x 4 x i1> @llvm.vp.trunc.nxv4i1.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i16_nxv4i1 = call <vscale x 4 x i1> @llvm.vp.trunc.nxv4i1.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i32_nxv4i1 = call <vscale x 4 x i1> @llvm.vp.trunc.nxv4i1.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4i64_nxv4i1 = call <vscale x 4 x i1> @llvm.vp.trunc.nxv4i1.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i1> undef, i32 undef)

  %nxv8i16_nxv8i8 = trunc <vscale x 8 x i16> undef to <vscale x 8 x i8>
  %nxv8i32_nxv8i8 = trunc <vscale x 8 x i32> undef to <vscale x 8 x i8>
  %nxv8i64_nxv8i8 = trunc <vscale x 8 x i64> undef to <vscale x 8 x i8>
  %nxv8i32_nxv8i16 = trunc <vscale x 8 x i32> undef to <vscale x 8 x i16>
  %nxv8i64_nxv8i16 = trunc <vscale x 8 x i64> undef to <vscale x 8 x i16>
  %nxv8i64_nxv8i32 = trunc <vscale x 8 x i64> undef to <vscale x 8 x i32>
  %nxv8i8_nxv8i1 = trunc <vscale x 8 x i8> undef to <vscale x 8 x i1>
  %nxv8i16_nxv8i1 = trunc <vscale x 8 x i16> undef to <vscale x 8 x i1>
  %nxv8i32_nxv8i1 = trunc <vscale x 8 x i32> undef to <vscale x 8 x i1>
  %nxv8i64_nxv8i1 = trunc <vscale x 8 x i64> undef to <vscale x 8 x i1>

  %vp_nxv8i16_nxv8i8 = call <vscale x 8 x i8> @llvm.vp.trunc.nxv8i8.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i32_nxv8i8 = call <vscale x 8 x i8> @llvm.vp.trunc.nxv8i8.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i64_nxv8i8 = call <vscale x 8 x i8> @llvm.vp.trunc.nxv8i8.nxv8i64(<vscale x 8 x i64> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i32_nxv8i16 = call <vscale x 8 x i16> @llvm.vp.trunc.nxv8i16.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i64_nxv8i16 = call <vscale x 8 x i16> @llvm.vp.trunc.nxv8i16.nxv8i64(<vscale x 8 x i64> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i64_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.trunc.nxv8i32.nxv8i64(<vscale x 8 x i64> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i8_nxv8i1 = call <vscale x 8 x i1> @llvm.vp.trunc.nxv8i1.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i16_nxv8i1 = call <vscale x 8 x i1> @llvm.vp.trunc.nxv8i1.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i32_nxv8i1 = call <vscale x 8 x i1> @llvm.vp.trunc.nxv8i1.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8i64_nxv8i1 = call <vscale x 8 x i1> @llvm.vp.trunc.nxv8i1.nxv8i64(<vscale x 8 x i64> undef, <vscale x 8 x i1> undef, i32 undef)

  %nxv16i16_nxv16i8 = trunc <vscale x 16 x i16> undef to <vscale x 16 x i8>
  %nxv16i32_nxv16i8 = trunc <vscale x 16 x i32> undef to <vscale x 16 x i8>
  %nxv16i64_nxv16i8 = trunc <vscale x 16 x i64> undef to <vscale x 16 x i8>
  %nxv16i32_nxv16i16 = trunc <vscale x 16 x i32> undef to <vscale x 16 x i16>
  %nxv16i64_nxv16i16 = trunc <vscale x 16 x i64> undef to <vscale x 16 x i16>
  %nxv16i64_nxv16i32 = trunc <vscale x 16 x i64> undef to <vscale x 16 x i32>
  %nxv16i8_nxv16i1 = trunc <vscale x 16 x i8> undef to <vscale x 16 x i1>
  %nxv16i16_nxv16i1 = trunc <vscale x 16 x i16> undef to <vscale x 16 x i1>
  %nxv16i32_nxv16i1 = trunc <vscale x 16 x i32> undef to <vscale x 16 x i1>
  %nxv16i64_nxv16i1 = trunc <vscale x 16 x i64> undef to <vscale x 16 x i1>

  %vp_nxv16i16_nxv16i8 = call <vscale x 16 x i8> @llvm.vp.trunc.nxv16i8.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i32_nxv16i8 = call <vscale x 16 x i8> @llvm.vp.trunc.nxv16i8.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i64_nxv16i8 = call <vscale x 16 x i8> @llvm.vp.trunc.nxv16i8.nxv16i64(<vscale x 16 x i64> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i32_nxv16i16 = call <vscale x 16 x i16> @llvm.vp.trunc.nxv16i16.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i64_nxv16i16 = call <vscale x 16 x i16> @llvm.vp.trunc.nxv16i16.nxv16i64(<vscale x 16 x i64> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i64_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.trunc.nxv16i32.nxv16i64(<vscale x 16 x i64> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i8_nxv16i1 = call <vscale x 16 x i1> @llvm.vp.trunc.nxv16i1.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i16_nxv16i1 = call <vscale x 16 x i1> @llvm.vp.trunc.nxv16i1.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i32_nxv16i1 = call <vscale x 16 x i1> @llvm.vp.trunc.nxv16i1.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16i64_nxv16i1 = call <vscale x 16 x i1> @llvm.vp.trunc.nxv16i1.nxv16i64(<vscale x 16 x i64> undef, <vscale x 16 x i1> undef, i32 undef)

  %nxv32i16_nxv32i8 = trunc <vscale x 32 x i16> undef to <vscale x 32 x i8>
  %nxv32i32_nxv32i8 = trunc <vscale x 32 x i32> undef to <vscale x 32 x i8>
  %nxv32i64_nxv32i8 = trunc <vscale x 32 x i64> undef to <vscale x 32 x i8>
  %nxv32i32_nxv32i16 = trunc <vscale x 32 x i32> undef to <vscale x 32 x i16>
  %nxv32i64_nxv32i16 = trunc <vscale x 32 x i64> undef to <vscale x 32 x i16>
  %nxv32i64_nxv32i32 = trunc <vscale x 32 x i64> undef to <vscale x 32 x i32>
  %nxv32i8_nxv32i1 = trunc <vscale x 32 x i8> undef to <vscale x 32 x i1>
  %nxv32i16_nxv32i1 = trunc <vscale x 32 x i16> undef to <vscale x 32 x i1>
  %nxv32i32_nxv32i1 = trunc <vscale x 32 x i32> undef to <vscale x 32 x i1>
  %nxv32i64_nxv32i1 = trunc <vscale x 32 x i64> undef to <vscale x 32 x i1>

  %vp_nxv32i16_nxv32i8 = call <vscale x 32 x i8> @llvm.vp.trunc.nxv32i8.nxv32i16(<vscale x 32 x i16> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i32_nxv32i8 = call <vscale x 32 x i8> @llvm.vp.trunc.nxv32i8.nxv32i32(<vscale x 32 x i32> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i64_nxv32i8 = call <vscale x 32 x i8> @llvm.vp.trunc.nxv32i8.nxv32i64(<vscale x 32 x i64> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i32_nxv32i16 = call <vscale x 32 x i16> @llvm.vp.trunc.nxv32i16.nxv32i32(<vscale x 32 x i32> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i64_nxv32i16 = call <vscale x 32 x i16> @llvm.vp.trunc.nxv32i16.nxv32i64(<vscale x 32 x i64> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i64_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.trunc.nxv32i32.nxv32i64(<vscale x 32 x i64> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i8_nxv32i1 = call <vscale x 32 x i1> @llvm.vp.trunc.nxv32i1.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i16_nxv32i1 = call <vscale x 32 x i1> @llvm.vp.trunc.nxv32i1.nxv32i16(<vscale x 32 x i16> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i32_nxv32i1 = call <vscale x 32 x i1> @llvm.vp.trunc.nxv32i1.nxv32i32(<vscale x 32 x i32> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32i64_nxv32i1 = call <vscale x 32 x i1> @llvm.vp.trunc.nxv32i1.nxv32i64(<vscale x 32 x i64> undef, <vscale x 32 x i1> undef, i32 undef)

  %nxv64i16_nxv64i8 = trunc <vscale x 64 x i16> undef to <vscale x 64 x i8>
  %nxv64i32_nxv64i8 = trunc <vscale x 64 x i32> undef to <vscale x 64 x i8>
  %nxv64i64_nxv64i8 = trunc <vscale x 64 x i64> undef to <vscale x 64 x i8>
  %nxv64i32_nxv64i16 = trunc <vscale x 64 x i32> undef to <vscale x 64 x i16>
  %nxv64i64_nxv64i16 = trunc <vscale x 64 x i64> undef to <vscale x 64 x i16>
  %nxv64i64_nxv64i32 = trunc <vscale x 64 x i64> undef to <vscale x 64 x i32>
  %nxv64i8_nxv64i1 = trunc <vscale x 64 x i8> undef to <vscale x 64 x i1>
  %nxv64i16_nxv64i1 = trunc <vscale x 64 x i16> undef to <vscale x 64 x i1>
  %nxv64i32_nxv64i1 = trunc <vscale x 64 x i32> undef to <vscale x 64 x i1>
  %nxv64i64_nxv64i1 = trunc <vscale x 64 x i64> undef to <vscale x 64 x i1>

  %vp_nxv64i16_nxv64i8 = call <vscale x 64 x i8> @llvm.vp.trunc.nxv64i8.nxv64i16(<vscale x 64 x i16> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i32_nxv64i8 = call <vscale x 64 x i8> @llvm.vp.trunc.nxv64i8.nxv64i32(<vscale x 64 x i32> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i64_nxv64i8 = call <vscale x 64 x i8> @llvm.vp.trunc.nxv64i8.nxv64i64(<vscale x 64 x i64> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i32_nxv64i16 = call <vscale x 64 x i16> @llvm.vp.trunc.nxv64i16.nxv64i32(<vscale x 64 x i32> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i64_nxv64i16 = call <vscale x 64 x i16> @llvm.vp.trunc.nxv64i16.nxv64i64(<vscale x 64 x i64> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i64_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.trunc.nxv64i32.nxv64i64(<vscale x 64 x i64> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i8_nxv64i1 = call <vscale x 64 x i1> @llvm.vp.trunc.nxv64i1.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i16_nxv64i1 = call <vscale x 64 x i1> @llvm.vp.trunc.nxv64i1.nxv64i16(<vscale x 64 x i16> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i32_nxv64i1 = call <vscale x 64 x i1> @llvm.vp.trunc.nxv64i1.nxv64i32(<vscale x 64 x i32> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64i64_nxv64i1 = call <vscale x 64 x i1> @llvm.vp.trunc.nxv64i1.nxv64i64(<vscale x 64 x i64> undef, <vscale x 64 x i1> undef, i32 undef)

  ret void
}

define void @fpext() {
; ZVE64D-LABEL: 'fpext'
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2f16_v2f32 = fpext <2 x half> undef to <2 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2f16_v2f64 = fpext <2 x half> undef to <2 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2f32_v2f64 = fpext <2 x float> undef to <2 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2f16_v2f32 = call <2 x float> @llvm.vp.fpext.v2f32.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2f16_v2f64 = call <2 x double> @llvm.vp.fpext.v2f64.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2f32_v2f64 = call <2 x double> @llvm.vp.fpext.v2f64.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4f16_v4f32 = fpext <4 x half> undef to <4 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v4f16_v4f64 = fpext <4 x half> undef to <4 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4f32_v4f64 = fpext <4 x float> undef to <4 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v4f16_v4f32 = call <4 x float> @llvm.vp.fpext.v4f32.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v4f16_v4f64 = call <4 x double> @llvm.vp.fpext.v4f64.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4f32_v4f64 = call <4 x double> @llvm.vp.fpext.v4f64.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8f16_v8f32 = fpext <8 x half> undef to <8 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %v8f16_v8f64 = fpext <8 x half> undef to <8 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8f32_v8f64 = fpext <8 x float> undef to <8 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8f16_v8f32 = call <8 x float> @llvm.vp.fpext.v8f32.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_v8f16_v8f64 = call <8 x double> @llvm.vp.fpext.v8f64.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v8f32_v8f64 = call <8 x double> @llvm.vp.fpext.v8f64.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v16f16_v16f32 = fpext <16 x half> undef to <16 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %v16f16_v16f64 = fpext <16 x half> undef to <16 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v16f32_v16f64 = fpext <16 x float> undef to <16 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v16f16_v16f32 = call <16 x float> @llvm.vp.fpext.v16f32.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %vp_v16f16_v16f64 = call <16 x double> @llvm.vp.fpext.v16f64.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v16f32_v16f64 = call <16 x double> @llvm.vp.fpext.v16f64.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v32f16_v32f32 = fpext <32 x half> undef to <32 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %v32f16_v32f64 = fpext <32 x half> undef to <32 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %v32f32_v32f64 = fpext <32 x float> undef to <32 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v32f16_v32f32 = call <32 x float> @llvm.vp.fpext.v32f32.v32f16(<32 x half> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_v32f16_v32f64 = call <32 x double> @llvm.vp.fpext.v32f64.v32f16(<32 x half> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_v32f32_v32f64 = call <32 x double> @llvm.vp.fpext.v32f64.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %v64f16_v64f32 = fpext <64 x half> undef to <64 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %v64f16_v64f64 = fpext <64 x half> undef to <64 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %v64f32_v64f64 = fpext <64 x float> undef to <64 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_v64f16_v64f32 = call <64 x float> @llvm.vp.fpext.v64f32.v64f16(<64 x half> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %vp_v64f16_v64f64 = call <64 x double> @llvm.vp.fpext.v64f64.v64f16(<64 x half> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_v64f32_v64f64 = call <64 x double> @llvm.vp.fpext.v64f64.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %v128f16_v128f32 = fpext <128 x half> undef to <128 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 102 for instruction: %v128f16_v128f64 = fpext <128 x half> undef to <128 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %v128f32_v128f64 = fpext <128 x float> undef to <128 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_v128f16_v128f32 = call <128 x float> @llvm.vp.fpext.v128f32.v128f16(<128 x half> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 102 for instruction: %vp_v128f16_v128f64 = call <128 x double> @llvm.vp.fpext.v128f64.v128f16(<128 x half> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %vp_v128f32_v128f64 = call <128 x double> @llvm.vp.fpext.v128f64.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1f16_nxv1f32 = fpext <vscale x 1 x half> undef to <vscale x 1 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1f16_nxv1f64 = fpext <vscale x 1 x half> undef to <vscale x 1 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1f32_nxv1f64 = fpext <vscale x 1 x float> undef to <vscale x 1 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1f16_nxv1f32 = call <vscale x 1 x float> @llvm.vp.fpext.nxv1f32.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1f16_nxv1f64 = call <vscale x 1 x double> @llvm.vp.fpext.nxv1f64.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1f32_nxv1f64 = call <vscale x 1 x double> @llvm.vp.fpext.nxv1f64.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv2f16_nxv2f32 = fpext <vscale x 2 x half> undef to <vscale x 2 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv2f16_nxv2f64 = fpext <vscale x 2 x half> undef to <vscale x 2 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2f32_nxv2f64 = fpext <vscale x 2 x float> undef to <vscale x 2 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv2f16_nxv2f32 = call <vscale x 2 x float> @llvm.vp.fpext.nxv2f32.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv2f16_nxv2f64 = call <vscale x 2 x double> @llvm.vp.fpext.nxv2f64.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2f32_nxv2f64 = call <vscale x 2 x double> @llvm.vp.fpext.nxv2f64.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4f16_nxv4f32 = fpext <vscale x 4 x half> undef to <vscale x 4 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %nxv4f16_nxv4f64 = fpext <vscale x 4 x half> undef to <vscale x 4 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv4f32_nxv4f64 = fpext <vscale x 4 x float> undef to <vscale x 4 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4f16_nxv4f32 = call <vscale x 4 x float> @llvm.vp.fpext.nxv4f32.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_nxv4f16_nxv4f64 = call <vscale x 4 x double> @llvm.vp.fpext.nxv4f64.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv4f32_nxv4f64 = call <vscale x 4 x double> @llvm.vp.fpext.nxv4f64.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv8f16_nxv8f32 = fpext <vscale x 8 x half> undef to <vscale x 8 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %nxv8f16_nxv8f64 = fpext <vscale x 8 x half> undef to <vscale x 8 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv8f32_nxv8f64 = fpext <vscale x 8 x float> undef to <vscale x 8 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv8f16_nxv8f32 = call <vscale x 8 x float> @llvm.vp.fpext.nxv8f32.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %vp_nxv8f16_nxv8f64 = call <vscale x 8 x double> @llvm.vp.fpext.nxv8f64.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv8f32_nxv8f64 = call <vscale x 8 x double> @llvm.vp.fpext.nxv8f64.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv16f16_nxv16f32 = fpext <vscale x 16 x half> undef to <vscale x 16 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %nxv16f16_nxv16f64 = fpext <vscale x 16 x half> undef to <vscale x 16 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %nxv16f32_nxv16f64 = fpext <vscale x 16 x float> undef to <vscale x 16 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv16f16_nxv16f32 = call <vscale x 16 x float> @llvm.vp.fpext.nxv16f32.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_nxv16f16_nxv16f64 = call <vscale x 16 x double> @llvm.vp.fpext.nxv16f64.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_nxv16f32_nxv16f64 = call <vscale x 16 x double> @llvm.vp.fpext.nxv16f64.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %nxv32f16_nxv32f32 = fpext <vscale x 32 x half> undef to <vscale x 32 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %nxv32f16_nxv32f64 = fpext <vscale x 32 x half> undef to <vscale x 32 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %nxv32f32_nxv32f64 = fpext <vscale x 32 x float> undef to <vscale x 32 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_nxv32f16_nxv32f32 = call <vscale x 32 x float> @llvm.vp.fpext.nxv32f32.nxv32f16(<vscale x 32 x half> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %vp_nxv32f16_nxv32f64 = call <vscale x 32 x double> @llvm.vp.fpext.nxv32f64.nxv32f16(<vscale x 32 x half> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_nxv32f32_nxv32f64 = call <vscale x 32 x double> @llvm.vp.fpext.nxv32f64.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %nxv64f16_nxv64f32 = fpext <vscale x 64 x half> undef to <vscale x 64 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 102 for instruction: %nxv64f16_nxv64f64 = fpext <vscale x 64 x half> undef to <vscale x 64 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %nxv64f32_nxv64f64 = fpext <vscale x 64 x float> undef to <vscale x 64 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_nxv64f16_nxv64f32 = call <vscale x 64 x float> @llvm.vp.fpext.nxv64f32.nxv64f16(<vscale x 64 x half> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 102 for instruction: %vp_nxv64f16_nxv64f64 = call <vscale x 64 x double> @llvm.vp.fpext.nxv64f64.nxv64f16(<vscale x 64 x half> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %vp_nxv64f32_nxv64f64 = call <vscale x 64 x double> @llvm.vp.fpext.nxv64f64.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; ZVE64X-LABEL: 'fpext'
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2f16_v2f32 = fpext <2 x half> undef to <2 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2f16_v2f64 = fpext <2 x half> undef to <2 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2f32_v2f64 = fpext <2 x float> undef to <2 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2f16_v2f32 = call <2 x float> @llvm.vp.fpext.v2f32.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2f16_v2f64 = call <2 x double> @llvm.vp.fpext.v2f64.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2f32_v2f64 = call <2 x double> @llvm.vp.fpext.v2f64.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v4f16_v4f32 = fpext <4 x half> undef to <4 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v4f16_v4f64 = fpext <4 x half> undef to <4 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v4f32_v4f64 = fpext <4 x float> undef to <4 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v4f16_v4f32 = call <4 x float> @llvm.vp.fpext.v4f32.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v4f16_v4f64 = call <4 x double> @llvm.vp.fpext.v4f64.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v4f32_v4f64 = call <4 x double> @llvm.vp.fpext.v4f64.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v8f16_v8f32 = fpext <8 x half> undef to <8 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v8f16_v8f64 = fpext <8 x half> undef to <8 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v8f32_v8f64 = fpext <8 x float> undef to <8 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v8f16_v8f32 = call <8 x float> @llvm.vp.fpext.v8f32.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v8f16_v8f64 = call <8 x double> @llvm.vp.fpext.v8f64.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v8f32_v8f64 = call <8 x double> @llvm.vp.fpext.v8f64.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v16f16_v16f32 = fpext <16 x half> undef to <16 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v16f16_v16f64 = fpext <16 x half> undef to <16 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v16f32_v16f64 = fpext <16 x float> undef to <16 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v16f16_v16f32 = call <16 x float> @llvm.vp.fpext.v16f32.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v16f16_v16f64 = call <16 x double> @llvm.vp.fpext.v16f64.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v16f32_v16f64 = call <16 x double> @llvm.vp.fpext.v16f64.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %v32f16_v32f32 = fpext <32 x half> undef to <32 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %v32f16_v32f64 = fpext <32 x half> undef to <32 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %v32f32_v32f64 = fpext <32 x float> undef to <32 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %vp_v32f16_v32f32 = call <32 x float> @llvm.vp.fpext.v32f32.v32f16(<32 x half> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %vp_v32f16_v32f64 = call <32 x double> @llvm.vp.fpext.v32f64.v32f16(<32 x half> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %vp_v32f32_v32f64 = call <32 x double> @llvm.vp.fpext.v32f64.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %v64f16_v64f32 = fpext <64 x half> undef to <64 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %v64f16_v64f64 = fpext <64 x half> undef to <64 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %v64f32_v64f64 = fpext <64 x float> undef to <64 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %vp_v64f16_v64f32 = call <64 x float> @llvm.vp.fpext.v64f32.v64f16(<64 x half> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %vp_v64f16_v64f64 = call <64 x double> @llvm.vp.fpext.v64f64.v64f16(<64 x half> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %vp_v64f32_v64f64 = call <64 x double> @llvm.vp.fpext.v64f64.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %v128f16_v128f32 = fpext <128 x half> undef to <128 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %v128f16_v128f64 = fpext <128 x half> undef to <128 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %v128f32_v128f64 = fpext <128 x float> undef to <128 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %vp_v128f16_v128f32 = call <128 x float> @llvm.vp.fpext.v128f32.v128f16(<128 x half> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %vp_v128f16_v128f64 = call <128 x double> @llvm.vp.fpext.v128f64.v128f16(<128 x half> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %vp_v128f32_v128f64 = call <128 x double> @llvm.vp.fpext.v128f64.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f16_nxv1f32 = fpext <vscale x 1 x half> undef to <vscale x 1 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f16_nxv1f64 = fpext <vscale x 1 x half> undef to <vscale x 1 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f32_nxv1f64 = fpext <vscale x 1 x float> undef to <vscale x 1 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f16_nxv1f32 = call <vscale x 1 x float> @llvm.vp.fpext.nxv1f32.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f16_nxv1f64 = call <vscale x 1 x double> @llvm.vp.fpext.nxv1f64.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f32_nxv1f64 = call <vscale x 1 x double> @llvm.vp.fpext.nxv1f64.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f16_nxv2f32 = fpext <vscale x 2 x half> undef to <vscale x 2 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f16_nxv2f64 = fpext <vscale x 2 x half> undef to <vscale x 2 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f32_nxv2f64 = fpext <vscale x 2 x float> undef to <vscale x 2 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f16_nxv2f32 = call <vscale x 2 x float> @llvm.vp.fpext.nxv2f32.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f16_nxv2f64 = call <vscale x 2 x double> @llvm.vp.fpext.nxv2f64.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f32_nxv2f64 = call <vscale x 2 x double> @llvm.vp.fpext.nxv2f64.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f16_nxv4f32 = fpext <vscale x 4 x half> undef to <vscale x 4 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f16_nxv4f64 = fpext <vscale x 4 x half> undef to <vscale x 4 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f32_nxv4f64 = fpext <vscale x 4 x float> undef to <vscale x 4 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f16_nxv4f32 = call <vscale x 4 x float> @llvm.vp.fpext.nxv4f32.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f16_nxv4f64 = call <vscale x 4 x double> @llvm.vp.fpext.nxv4f64.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f32_nxv4f64 = call <vscale x 4 x double> @llvm.vp.fpext.nxv4f64.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f16_nxv8f32 = fpext <vscale x 8 x half> undef to <vscale x 8 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f16_nxv8f64 = fpext <vscale x 8 x half> undef to <vscale x 8 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f32_nxv8f64 = fpext <vscale x 8 x float> undef to <vscale x 8 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f16_nxv8f32 = call <vscale x 8 x float> @llvm.vp.fpext.nxv8f32.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f16_nxv8f64 = call <vscale x 8 x double> @llvm.vp.fpext.nxv8f64.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f32_nxv8f64 = call <vscale x 8 x double> @llvm.vp.fpext.nxv8f64.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f16_nxv16f32 = fpext <vscale x 16 x half> undef to <vscale x 16 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f16_nxv16f64 = fpext <vscale x 16 x half> undef to <vscale x 16 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f32_nxv16f64 = fpext <vscale x 16 x float> undef to <vscale x 16 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f16_nxv16f32 = call <vscale x 16 x float> @llvm.vp.fpext.nxv16f32.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f16_nxv16f64 = call <vscale x 16 x double> @llvm.vp.fpext.nxv16f64.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f32_nxv16f64 = call <vscale x 16 x double> @llvm.vp.fpext.nxv16f64.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f16_nxv32f32 = fpext <vscale x 32 x half> undef to <vscale x 32 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f16_nxv32f64 = fpext <vscale x 32 x half> undef to <vscale x 32 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f32_nxv32f64 = fpext <vscale x 32 x float> undef to <vscale x 32 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f16_nxv32f32 = call <vscale x 32 x float> @llvm.vp.fpext.nxv32f32.nxv32f16(<vscale x 32 x half> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f16_nxv32f64 = call <vscale x 32 x double> @llvm.vp.fpext.nxv32f64.nxv32f16(<vscale x 32 x half> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f32_nxv32f64 = call <vscale x 32 x double> @llvm.vp.fpext.nxv32f64.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f16_nxv64f32 = fpext <vscale x 64 x half> undef to <vscale x 64 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f16_nxv64f64 = fpext <vscale x 64 x half> undef to <vscale x 64 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f32_nxv64f64 = fpext <vscale x 64 x float> undef to <vscale x 64 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f16_nxv64f32 = call <vscale x 64 x float> @llvm.vp.fpext.nxv64f32.nxv64f16(<vscale x 64 x half> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f16_nxv64f64 = call <vscale x 64 x double> @llvm.vp.fpext.nxv64f64.nxv64f16(<vscale x 64 x half> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f32_nxv64f64 = call <vscale x 64 x double> @llvm.vp.fpext.nxv64f64.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  %v2f16_v2f32 = fpext <2 x half> undef to <2 x float>
  %v2f16_v2f64 = fpext <2 x half> undef to <2 x double>
  %v2f32_v2f64 = fpext <2 x float> undef to <2 x double>

  %vp_v2f16_v2f32 = call <2 x float> @llvm.vp.fpext.v2f32.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
  %vp_v2f16_v2f64 = call <2 x double> @llvm.vp.fpext.v2f64.v2f16(<2 x half> undef, <2 x i1> undef, i32 undef)
  %vp_v2f32_v2f64 = call <2 x double> @llvm.vp.fpext.v2f64.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)

  %v4f16_v4f32 = fpext <4 x half> undef to <4 x float>
  %v4f16_v4f64 = fpext <4 x half> undef to <4 x double>
  %v4f32_v4f64 = fpext <4 x float> undef to <4 x double>

  %vp_v4f16_v4f32 = call <4 x float> @llvm.vp.fpext.v4f32.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
  %vp_v4f16_v4f64 = call <4 x double> @llvm.vp.fpext.v4f64.v4f16(<4 x half> undef, <4 x i1> undef, i32 undef)
  %vp_v4f32_v4f64 = call <4 x double> @llvm.vp.fpext.v4f64.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)

  %v8f16_v8f32 = fpext <8 x half> undef to <8 x float>
  %v8f16_v8f64 = fpext <8 x half> undef to <8 x double>
  %v8f32_v8f64 = fpext <8 x float> undef to <8 x double>

  %vp_v8f16_v8f32 = call <8 x float> @llvm.vp.fpext.v8f32.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
  %vp_v8f16_v8f64 = call <8 x double> @llvm.vp.fpext.v8f64.v8f16(<8 x half> undef, <8 x i1> undef, i32 undef)
  %vp_v8f32_v8f64 = call <8 x double> @llvm.vp.fpext.v8f64.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)

  %v16f16_v16f32 = fpext <16 x half> undef to <16 x float>
  %v16f16_v16f64 = fpext <16 x half> undef to <16 x double>
  %v16f32_v16f64 = fpext <16 x float> undef to <16 x double>

  %vp_v16f16_v16f32 = call <16 x float> @llvm.vp.fpext.v16f32.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
  %vp_v16f16_v16f64 = call <16 x double> @llvm.vp.fpext.v16f64.v16f16(<16 x half> undef, <16 x i1> undef, i32 undef)
  %vp_v16f32_v16f64 = call <16 x double> @llvm.vp.fpext.v16f64.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)

  %v32f16_v32f32 = fpext <32 x half> undef to <32 x float>
  %v32f16_v32f64 = fpext <32 x half> undef to <32 x double>
  %v32f32_v32f64 = fpext <32 x float> undef to <32 x double>

  %vp_v32f16_v32f32 = call <32 x float> @llvm.vp.fpext.v32f32.v32f16(<32 x half> undef, <32 x i1> undef, i32 undef)
  %vp_v32f16_v32f64 = call <32 x double> @llvm.vp.fpext.v32f64.v32f16(<32 x half> undef, <32 x i1> undef, i32 undef)
  %vp_v32f32_v32f64 = call <32 x double> @llvm.vp.fpext.v32f64.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)

  %v64f16_v64f32 = fpext <64 x half> undef to <64 x float>
  %v64f16_v64f64 = fpext <64 x half> undef to <64 x double>
  %v64f32_v64f64 = fpext <64 x float> undef to <64 x double>

  %vp_v64f16_v64f32 = call <64 x float> @llvm.vp.fpext.v64f32.v64f16(<64 x half> undef, <64 x i1> undef, i32 undef)
  %vp_v64f16_v64f64 = call <64 x double> @llvm.vp.fpext.v64f64.v64f16(<64 x half> undef, <64 x i1> undef, i32 undef)
  %vp_v64f32_v64f64 = call <64 x double> @llvm.vp.fpext.v64f64.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)

  %v128f16_v128f32 = fpext <128 x half> undef to <128 x float>
  %v128f16_v128f64 = fpext <128 x half> undef to <128 x double>
  %v128f32_v128f64 = fpext <128 x float> undef to <128 x double>

  %vp_v128f16_v128f32 = call <128 x float> @llvm.vp.fpext.v128f32.v128f16(<128 x half> undef, <128 x i1> undef, i32 undef)
  %vp_v128f16_v128f64 = call <128 x double> @llvm.vp.fpext.v128f64.v128f16(<128 x half> undef, <128 x i1> undef, i32 undef)
  %vp_v128f32_v128f64 = call <128 x double> @llvm.vp.fpext.v128f64.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)

  %nxv1f16_nxv1f32 = fpext <vscale x 1 x half> undef to <vscale x 1 x float>
  %nxv1f16_nxv1f64 = fpext <vscale x 1 x half> undef to <vscale x 1 x double>
  %nxv1f32_nxv1f64 = fpext <vscale x 1 x float> undef to <vscale x 1 x double>

  %vp_nxv1f16_nxv1f32 = call <vscale x 1 x float> @llvm.vp.fpext.nxv1f32.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1f16_nxv1f64 = call <vscale x 1 x double> @llvm.vp.fpext.nxv1f64.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1f32_nxv1f64 = call <vscale x 1 x double> @llvm.vp.fpext.nxv1f64.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)

  %nxv2f16_nxv2f32 = fpext <vscale x 2 x half> undef to <vscale x 2 x float>
  %nxv2f16_nxv2f64 = fpext <vscale x 2 x half> undef to <vscale x 2 x double>
  %nxv2f32_nxv2f64 = fpext <vscale x 2 x float> undef to <vscale x 2 x double>

  %vp_nxv2f16_nxv2f32 = call <vscale x 2 x float> @llvm.vp.fpext.nxv2f32.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2f16_nxv2f64 = call <vscale x 2 x double> @llvm.vp.fpext.nxv2f64.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2f32_nxv2f64 = call <vscale x 2 x double> @llvm.vp.fpext.nxv2f64.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)

  %nxv4f16_nxv4f32 = fpext <vscale x 4 x half> undef to <vscale x 4 x float>
  %nxv4f16_nxv4f64 = fpext <vscale x 4 x half> undef to <vscale x 4 x double>
  %nxv4f32_nxv4f64 = fpext <vscale x 4 x float> undef to <vscale x 4 x double>

  %vp_nxv4f16_nxv4f32 = call <vscale x 4 x float> @llvm.vp.fpext.nxv4f32.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4f16_nxv4f64 = call <vscale x 4 x double> @llvm.vp.fpext.nxv4f64.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4f32_nxv4f64 = call <vscale x 4 x double> @llvm.vp.fpext.nxv4f64.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)

  %nxv8f16_nxv8f32 = fpext <vscale x 8 x half> undef to <vscale x 8 x float>
  %nxv8f16_nxv8f64 = fpext <vscale x 8 x half> undef to <vscale x 8 x double>
  %nxv8f32_nxv8f64 = fpext <vscale x 8 x float> undef to <vscale x 8 x double>

  %vp_nxv8f16_nxv8f32 = call <vscale x 8 x float> @llvm.vp.fpext.nxv8f32.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8f16_nxv8f64 = call <vscale x 8 x double> @llvm.vp.fpext.nxv8f64.nxv8f16(<vscale x 8 x half> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8f32_nxv8f64 = call <vscale x 8 x double> @llvm.vp.fpext.nxv8f64.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)

  %nxv16f16_nxv16f32 = fpext <vscale x 16 x half> undef to <vscale x 16 x float>
  %nxv16f16_nxv16f64 = fpext <vscale x 16 x half> undef to <vscale x 16 x double>
  %nxv16f32_nxv16f64 = fpext <vscale x 16 x float> undef to <vscale x 16 x double>

  %vp_nxv16f16_nxv16f32 = call <vscale x 16 x float> @llvm.vp.fpext.nxv16f32.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16f16_nxv16f64 = call <vscale x 16 x double> @llvm.vp.fpext.nxv16f64.nxv16f16(<vscale x 16 x half> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16f32_nxv16f64 = call <vscale x 16 x double> @llvm.vp.fpext.nxv16f64.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)

  %nxv32f16_nxv32f32 = fpext <vscale x 32 x half> undef to <vscale x 32 x float>
  %nxv32f16_nxv32f64 = fpext <vscale x 32 x half> undef to <vscale x 32 x double>
  %nxv32f32_nxv32f64 = fpext <vscale x 32 x float> undef to <vscale x 32 x double>

  %vp_nxv32f16_nxv32f32 = call <vscale x 32 x float> @llvm.vp.fpext.nxv32f32.nxv32f16(<vscale x 32 x half> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32f16_nxv32f64 = call <vscale x 32 x double> @llvm.vp.fpext.nxv32f64.nxv32f16(<vscale x 32 x half> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32f32_nxv32f64 = call <vscale x 32 x double> @llvm.vp.fpext.nxv32f64.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)

  %nxv64f16_nxv64f32 = fpext <vscale x 64 x half> undef to <vscale x 64 x float>
  %nxv64f16_nxv64f64 = fpext <vscale x 64 x half> undef to <vscale x 64 x double>
  %nxv64f32_nxv64f64 = fpext <vscale x 64 x float> undef to <vscale x 64 x double>

  %vp_nxv64f16_nxv64f32 = call <vscale x 64 x float> @llvm.vp.fpext.nxv64f32.nxv64f16(<vscale x 64 x half> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64f16_nxv64f64 = call <vscale x 64 x double> @llvm.vp.fpext.nxv64f64.nxv64f16(<vscale x 64 x half> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64f32_nxv64f64 = call <vscale x 64 x double> @llvm.vp.fpext.nxv64f64.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)

  ret void
}

define void @fptrunc() {
; ZVE64D-LABEL: 'fptrunc'
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2f32_v2f16 = fptrunc <2 x float> undef to <2 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2f64_v2f16 = fptrunc <2 x double> undef to <2 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2f64_v2f32 = fptrunc <2 x double> undef to <2 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2f32_v2f16 = call <2 x half> @llvm.vp.fptrunc.v2f16.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2f64_v2f16 = call <2 x half> @llvm.vp.fptrunc.v2f16.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2f64_v2f32 = call <2 x float> @llvm.vp.fptrunc.v2f32.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4f32_v4f16 = fptrunc <4 x float> undef to <4 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4f64_v4f16 = fptrunc <4 x double> undef to <4 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4f64_v4f32 = fptrunc <4 x double> undef to <4 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v4f32_v4f16 = call <4 x half> @llvm.vp.fptrunc.v4f16.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4f64_v4f16 = call <4 x half> @llvm.vp.fptrunc.v4f16.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v4f64_v4f32 = call <4 x float> @llvm.vp.fptrunc.v4f32.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8f32_v8f16 = fptrunc <8 x float> undef to <8 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v8f64_v8f16 = fptrunc <8 x double> undef to <8 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8f64_v8f32 = fptrunc <8 x double> undef to <8 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v8f32_v8f16 = call <8 x half> @llvm.vp.fptrunc.v8f16.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v8f64_v8f16 = call <8 x half> @llvm.vp.fptrunc.v8f16.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8f64_v8f32 = call <8 x float> @llvm.vp.fptrunc.v8f32.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v16f32_v16f16 = fptrunc <16 x float> undef to <16 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %v16f64_v16f16 = fptrunc <16 x double> undef to <16 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v16f64_v16f32 = fptrunc <16 x double> undef to <16 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v16f32_v16f16 = call <16 x half> @llvm.vp.fptrunc.v16f16.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_v16f64_v16f16 = call <16 x half> @llvm.vp.fptrunc.v16f16.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v16f64_v16f32 = call <16 x float> @llvm.vp.fptrunc.v16f32.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v32f32_v32f16 = fptrunc <32 x float> undef to <32 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %v32f64_v32f16 = fptrunc <32 x double> undef to <32 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %v32f64_v32f32 = fptrunc <32 x double> undef to <32 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v32f32_v32f16 = call <32 x half> @llvm.vp.fptrunc.v32f16.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %vp_v32f64_v32f16 = call <32 x half> @llvm.vp.fptrunc.v32f16.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %vp_v32f64_v32f32 = call <32 x float> @llvm.vp.fptrunc.v32f32.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %v64f32_v64f16 = fptrunc <64 x float> undef to <64 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %v64f64_v64f16 = fptrunc <64 x double> undef to <64 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %v64f64_v64f32 = fptrunc <64 x double> undef to <64 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %vp_v64f32_v64f16 = call <64 x half> @llvm.vp.fptrunc.v64f16.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %vp_v64f64_v64f16 = call <64 x half> @llvm.vp.fptrunc.v64f16.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %vp_v64f64_v64f32 = call <64 x float> @llvm.vp.fptrunc.v64f32.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %v128f32_v128f16 = fptrunc <128 x float> undef to <128 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %v128f64_v128f16 = fptrunc <128 x double> undef to <128 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %v128f64_v128f32 = fptrunc <128 x double> undef to <128 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %vp_v128f32_v128f16 = call <128 x half> @llvm.vp.fptrunc.v128f16.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %vp_v128f64_v128f16 = call <128 x half> @llvm.vp.fptrunc.v128f16.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %vp_v128f64_v128f32 = call <128 x float> @llvm.vp.fptrunc.v128f32.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1f32_nxv1f16 = fptrunc <vscale x 1 x float> undef to <vscale x 1 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1f64_nxv1f16 = fptrunc <vscale x 1 x double> undef to <vscale x 1 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1f64_nxv1f32 = fptrunc <vscale x 1 x double> undef to <vscale x 1 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1f32_nxv1f16 = call <vscale x 1 x half> @llvm.vp.fptrunc.nxv1f16.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1f64_nxv1f16 = call <vscale x 1 x half> @llvm.vp.fptrunc.nxv1f16.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1f64_nxv1f32 = call <vscale x 1 x float> @llvm.vp.fptrunc.nxv1f32.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv2f32_nxv1f16 = fptrunc <vscale x 2 x float> undef to <vscale x 2 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2f64_nxv1f16 = fptrunc <vscale x 2 x double> undef to <vscale x 2 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv2f64_nxv1f32 = fptrunc <vscale x 2 x double> undef to <vscale x 2 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv2f32_nxv2f16 = call <vscale x 2 x half> @llvm.vp.fptrunc.nxv2f16.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2f64_nxv2f16 = call <vscale x 2 x half> @llvm.vp.fptrunc.nxv2f16.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv2f64_nxv2f32 = call <vscale x 2 x float> @llvm.vp.fptrunc.nxv2f32.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv4f32_nxv4f16 = fptrunc <vscale x 4 x float> undef to <vscale x 4 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv4f64_nxv4f16 = fptrunc <vscale x 4 x double> undef to <vscale x 4 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4f64_nxv4f32 = fptrunc <vscale x 4 x double> undef to <vscale x 4 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv4f32_nxv4f16 = call <vscale x 4 x half> @llvm.vp.fptrunc.nxv4f16.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv4f64_nxv4f16 = call <vscale x 4 x half> @llvm.vp.fptrunc.nxv4f16.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4f64_nxv4f32 = call <vscale x 4 x float> @llvm.vp.fptrunc.nxv4f32.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv8f32_nxv8f16 = fptrunc <vscale x 8 x float> undef to <vscale x 8 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %nxv8f64_nxv8f16 = fptrunc <vscale x 8 x double> undef to <vscale x 8 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv8f64_nxv8f32 = fptrunc <vscale x 8 x double> undef to <vscale x 8 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv8f32_nxv8f16 = call <vscale x 8 x half> @llvm.vp.fptrunc.nxv8f16.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_nxv8f64_nxv8f16 = call <vscale x 8 x half> @llvm.vp.fptrunc.nxv8f16.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv8f64_nxv8f32 = call <vscale x 8 x float> @llvm.vp.fptrunc.nxv8f32.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv16f32_nxv16f16 = fptrunc <vscale x 16 x float> undef to <vscale x 16 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %nxv16f64_nxv16f16 = fptrunc <vscale x 16 x double> undef to <vscale x 16 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %nxv16f64_nxv16f32 = fptrunc <vscale x 16 x double> undef to <vscale x 16 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv16f32_nxv16f16 = call <vscale x 16 x half> @llvm.vp.fptrunc.nxv16f16.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %vp_nxv16f64_nxv16f16 = call <vscale x 16 x half> @llvm.vp.fptrunc.nxv16f16.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %vp_nxv16f64_nxv16f32 = call <vscale x 16 x float> @llvm.vp.fptrunc.nxv16f32.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %nxv32f32_nxv32f16 = fptrunc <vscale x 32 x float> undef to <vscale x 32 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %nxv32f64_nxv32f16 = fptrunc <vscale x 32 x double> undef to <vscale x 32 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %nxv32f64_nxv32f32 = fptrunc <vscale x 32 x double> undef to <vscale x 32 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %vp_nxv32f32_nxv32f16 = call <vscale x 32 x half> @llvm.vp.fptrunc.nxv32f16.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %vp_nxv32f64_nxv32f16 = call <vscale x 32 x half> @llvm.vp.fptrunc.nxv32f16.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %vp_nxv32f64_nxv32f32 = call <vscale x 32 x float> @llvm.vp.fptrunc.nxv32f32.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %nxv64f32_nxv64f16 = fptrunc <vscale x 64 x float> undef to <vscale x 64 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %nxv64f64_nxv64f16 = fptrunc <vscale x 64 x double> undef to <vscale x 64 x half>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %nxv64f64_nxv64f32 = fptrunc <vscale x 64 x double> undef to <vscale x 64 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %vp_nxv64f32_nxv64f16 = call <vscale x 64 x half> @llvm.vp.fptrunc.nxv64f16.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %vp_nxv64f64_nxv64f16 = call <vscale x 64 x half> @llvm.vp.fptrunc.nxv64f16.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %vp_nxv64f64_nxv64f32 = call <vscale x 64 x float> @llvm.vp.fptrunc.nxv64f32.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; ZVE64X-LABEL: 'fptrunc'
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2f32_v2f16 = fptrunc <2 x float> undef to <2 x half>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2f64_v2f16 = fptrunc <2 x double> undef to <2 x half>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2f64_v2f32 = fptrunc <2 x double> undef to <2 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2f32_v2f16 = call <2 x half> @llvm.vp.fptrunc.v2f16.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2f64_v2f16 = call <2 x half> @llvm.vp.fptrunc.v2f16.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2f64_v2f32 = call <2 x float> @llvm.vp.fptrunc.v2f32.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v4f32_v4f16 = fptrunc <4 x float> undef to <4 x half>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v4f64_v4f16 = fptrunc <4 x double> undef to <4 x half>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v4f64_v4f32 = fptrunc <4 x double> undef to <4 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v4f32_v4f16 = call <4 x half> @llvm.vp.fptrunc.v4f16.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v4f64_v4f16 = call <4 x half> @llvm.vp.fptrunc.v4f16.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v4f64_v4f32 = call <4 x float> @llvm.vp.fptrunc.v4f32.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v8f32_v8f16 = fptrunc <8 x float> undef to <8 x half>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v8f64_v8f16 = fptrunc <8 x double> undef to <8 x half>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v8f64_v8f32 = fptrunc <8 x double> undef to <8 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v8f32_v8f16 = call <8 x half> @llvm.vp.fptrunc.v8f16.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v8f64_v8f16 = call <8 x half> @llvm.vp.fptrunc.v8f16.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v8f64_v8f32 = call <8 x float> @llvm.vp.fptrunc.v8f32.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v16f32_v16f16 = fptrunc <16 x float> undef to <16 x half>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v16f64_v16f16 = fptrunc <16 x double> undef to <16 x half>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v16f64_v16f32 = fptrunc <16 x double> undef to <16 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v16f32_v16f16 = call <16 x half> @llvm.vp.fptrunc.v16f16.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v16f64_v16f16 = call <16 x half> @llvm.vp.fptrunc.v16f16.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v16f64_v16f32 = call <16 x float> @llvm.vp.fptrunc.v16f32.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %v32f32_v32f16 = fptrunc <32 x float> undef to <32 x half>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %v32f64_v32f16 = fptrunc <32 x double> undef to <32 x half>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %v32f64_v32f32 = fptrunc <32 x double> undef to <32 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %vp_v32f32_v32f16 = call <32 x half> @llvm.vp.fptrunc.v32f16.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %vp_v32f64_v32f16 = call <32 x half> @llvm.vp.fptrunc.v32f16.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %vp_v32f64_v32f32 = call <32 x float> @llvm.vp.fptrunc.v32f32.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %v64f32_v64f16 = fptrunc <64 x float> undef to <64 x half>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %v64f64_v64f16 = fptrunc <64 x double> undef to <64 x half>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %v64f64_v64f32 = fptrunc <64 x double> undef to <64 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %vp_v64f32_v64f16 = call <64 x half> @llvm.vp.fptrunc.v64f16.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %vp_v64f64_v64f16 = call <64 x half> @llvm.vp.fptrunc.v64f16.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %vp_v64f64_v64f32 = call <64 x float> @llvm.vp.fptrunc.v64f32.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %v128f32_v128f16 = fptrunc <128 x float> undef to <128 x half>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %v128f64_v128f16 = fptrunc <128 x double> undef to <128 x half>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %v128f64_v128f32 = fptrunc <128 x double> undef to <128 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %vp_v128f32_v128f16 = call <128 x half> @llvm.vp.fptrunc.v128f16.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %vp_v128f64_v128f16 = call <128 x half> @llvm.vp.fptrunc.v128f16.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %vp_v128f64_v128f32 = call <128 x float> @llvm.vp.fptrunc.v128f32.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f32_nxv1f16 = fptrunc <vscale x 1 x float> undef to <vscale x 1 x half>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f64_nxv1f16 = fptrunc <vscale x 1 x double> undef to <vscale x 1 x half>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f64_nxv1f32 = fptrunc <vscale x 1 x double> undef to <vscale x 1 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f32_nxv1f16 = call <vscale x 1 x half> @llvm.vp.fptrunc.nxv1f16.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f64_nxv1f16 = call <vscale x 1 x half> @llvm.vp.fptrunc.nxv1f16.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f64_nxv1f32 = call <vscale x 1 x float> @llvm.vp.fptrunc.nxv1f32.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f32_nxv1f16 = fptrunc <vscale x 2 x float> undef to <vscale x 2 x half>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f64_nxv1f16 = fptrunc <vscale x 2 x double> undef to <vscale x 2 x half>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f64_nxv1f32 = fptrunc <vscale x 2 x double> undef to <vscale x 2 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f32_nxv2f16 = call <vscale x 2 x half> @llvm.vp.fptrunc.nxv2f16.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f64_nxv2f16 = call <vscale x 2 x half> @llvm.vp.fptrunc.nxv2f16.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f64_nxv2f32 = call <vscale x 2 x float> @llvm.vp.fptrunc.nxv2f32.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f32_nxv4f16 = fptrunc <vscale x 4 x float> undef to <vscale x 4 x half>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f64_nxv4f16 = fptrunc <vscale x 4 x double> undef to <vscale x 4 x half>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f64_nxv4f32 = fptrunc <vscale x 4 x double> undef to <vscale x 4 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f32_nxv4f16 = call <vscale x 4 x half> @llvm.vp.fptrunc.nxv4f16.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f64_nxv4f16 = call <vscale x 4 x half> @llvm.vp.fptrunc.nxv4f16.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f64_nxv4f32 = call <vscale x 4 x float> @llvm.vp.fptrunc.nxv4f32.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f32_nxv8f16 = fptrunc <vscale x 8 x float> undef to <vscale x 8 x half>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f64_nxv8f16 = fptrunc <vscale x 8 x double> undef to <vscale x 8 x half>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f64_nxv8f32 = fptrunc <vscale x 8 x double> undef to <vscale x 8 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f32_nxv8f16 = call <vscale x 8 x half> @llvm.vp.fptrunc.nxv8f16.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f64_nxv8f16 = call <vscale x 8 x half> @llvm.vp.fptrunc.nxv8f16.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f64_nxv8f32 = call <vscale x 8 x float> @llvm.vp.fptrunc.nxv8f32.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f32_nxv16f16 = fptrunc <vscale x 16 x float> undef to <vscale x 16 x half>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f64_nxv16f16 = fptrunc <vscale x 16 x double> undef to <vscale x 16 x half>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f64_nxv16f32 = fptrunc <vscale x 16 x double> undef to <vscale x 16 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f32_nxv16f16 = call <vscale x 16 x half> @llvm.vp.fptrunc.nxv16f16.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f64_nxv16f16 = call <vscale x 16 x half> @llvm.vp.fptrunc.nxv16f16.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f64_nxv16f32 = call <vscale x 16 x float> @llvm.vp.fptrunc.nxv16f32.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f32_nxv32f16 = fptrunc <vscale x 32 x float> undef to <vscale x 32 x half>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f64_nxv32f16 = fptrunc <vscale x 32 x double> undef to <vscale x 32 x half>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f64_nxv32f32 = fptrunc <vscale x 32 x double> undef to <vscale x 32 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f32_nxv32f16 = call <vscale x 32 x half> @llvm.vp.fptrunc.nxv32f16.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f64_nxv32f16 = call <vscale x 32 x half> @llvm.vp.fptrunc.nxv32f16.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f64_nxv32f32 = call <vscale x 32 x float> @llvm.vp.fptrunc.nxv32f32.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f32_nxv64f16 = fptrunc <vscale x 64 x float> undef to <vscale x 64 x half>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f64_nxv64f16 = fptrunc <vscale x 64 x double> undef to <vscale x 64 x half>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f64_nxv64f32 = fptrunc <vscale x 64 x double> undef to <vscale x 64 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f32_nxv64f16 = call <vscale x 64 x half> @llvm.vp.fptrunc.nxv64f16.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f64_nxv64f16 = call <vscale x 64 x half> @llvm.vp.fptrunc.nxv64f16.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f64_nxv64f32 = call <vscale x 64 x float> @llvm.vp.fptrunc.nxv64f32.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  %v2f32_v2f16 = fptrunc <2 x float> undef to <2 x half>
  %v2f64_v2f16 = fptrunc <2 x double> undef to <2 x half>
  %v2f64_v2f32 = fptrunc <2 x double> undef to <2 x float>

  %vp_v2f32_v2f16 = call <2 x half> @llvm.vp.fptrunc.v2f16.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
  %vp_v2f64_v2f16 = call <2 x half> @llvm.vp.fptrunc.v2f16.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
  %vp_v2f64_v2f32 = call <2 x float> @llvm.vp.fptrunc.v2f32.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)

  %v4f32_v4f16 = fptrunc <4 x float> undef to <4 x half>
  %v4f64_v4f16 = fptrunc <4 x double> undef to <4 x half>
  %v4f64_v4f32 = fptrunc <4 x double> undef to <4 x float>

  %vp_v4f32_v4f16 = call <4 x half> @llvm.vp.fptrunc.v4f16.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
  %vp_v4f64_v4f16 = call <4 x half> @llvm.vp.fptrunc.v4f16.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
  %vp_v4f64_v4f32 = call <4 x float> @llvm.vp.fptrunc.v4f32.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)

  %v8f32_v8f16 = fptrunc <8 x float> undef to <8 x half>
  %v8f64_v8f16 = fptrunc <8 x double> undef to <8 x half>
  %v8f64_v8f32 = fptrunc <8 x double> undef to <8 x float>

  %vp_v8f32_v8f16 = call <8 x half> @llvm.vp.fptrunc.v8f16.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
  %vp_v8f64_v8f16 = call <8 x half> @llvm.vp.fptrunc.v8f16.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
  %vp_v8f64_v8f32 = call <8 x float> @llvm.vp.fptrunc.v8f32.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)

  %v16f32_v16f16 = fptrunc <16 x float> undef to <16 x half>
  %v16f64_v16f16 = fptrunc <16 x double> undef to <16 x half>
  %v16f64_v16f32 = fptrunc <16 x double> undef to <16 x float>

  %vp_v16f32_v16f16 = call <16 x half> @llvm.vp.fptrunc.v16f16.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
  %vp_v16f64_v16f16 = call <16 x half> @llvm.vp.fptrunc.v16f16.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
  %vp_v16f64_v16f32 = call <16 x float> @llvm.vp.fptrunc.v16f32.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)

  %v32f32_v32f16 = fptrunc <32 x float> undef to <32 x half>
  %v32f64_v32f16 = fptrunc <32 x double> undef to <32 x half>
  %v32f64_v32f32 = fptrunc <32 x double> undef to <32 x float>

  %vp_v32f32_v32f16 = call <32 x half> @llvm.vp.fptrunc.v32f16.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
  %vp_v32f64_v32f16 = call <32 x half> @llvm.vp.fptrunc.v32f16.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
  %vp_v32f64_v32f32 = call <32 x float> @llvm.vp.fptrunc.v32f32.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)

  %v64f32_v64f16 = fptrunc <64 x float> undef to <64 x half>
  %v64f64_v64f16 = fptrunc <64 x double> undef to <64 x half>
  %v64f64_v64f32 = fptrunc <64 x double> undef to <64 x float>

  %vp_v64f32_v64f16 = call <64 x half> @llvm.vp.fptrunc.v64f16.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
  %vp_v64f64_v64f16 = call <64 x half> @llvm.vp.fptrunc.v64f16.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
  %vp_v64f64_v64f32 = call <64 x float> @llvm.vp.fptrunc.v64f32.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)

  %v128f32_v128f16 = fptrunc <128 x float> undef to <128 x half>
  %v128f64_v128f16 = fptrunc <128 x double> undef to <128 x half>
  %v128f64_v128f32 = fptrunc <128 x double> undef to <128 x float>

  %vp_v128f32_v128f16 = call <128 x half> @llvm.vp.fptrunc.v128f16.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
  %vp_v128f64_v128f16 = call <128 x half> @llvm.vp.fptrunc.v128f16.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
  %vp_v128f64_v128f32 = call <128 x float> @llvm.vp.fptrunc.v128f32.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)

  %nxv1f32_nxv1f16 = fptrunc <vscale x 1 x float> undef to <vscale x 1 x half>
  %nxv1f64_nxv1f16 = fptrunc <vscale x 1 x double> undef to <vscale x 1 x half>
  %nxv1f64_nxv1f32 = fptrunc <vscale x 1 x double> undef to <vscale x 1 x float>

  %vp_nxv1f32_nxv1f16 = call <vscale x 1 x half> @llvm.vp.fptrunc.nxv1f16.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1f64_nxv1f16 = call <vscale x 1 x half> @llvm.vp.fptrunc.nxv1f16.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1f64_nxv1f32 = call <vscale x 1 x float> @llvm.vp.fptrunc.nxv1f32.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)

  %nxv2f32_nxv1f16 = fptrunc <vscale x 2 x float> undef to <vscale x 2 x half>
  %nxv2f64_nxv1f16 = fptrunc <vscale x 2 x double> undef to <vscale x 2 x half>
  %nxv2f64_nxv1f32 = fptrunc <vscale x 2 x double> undef to <vscale x 2 x float>

  %vp_nxv2f32_nxv2f16 = call <vscale x 2 x half> @llvm.vp.fptrunc.nxv2f16.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2f64_nxv2f16 = call <vscale x 2 x half> @llvm.vp.fptrunc.nxv2f16.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2f64_nxv2f32 = call <vscale x 2 x float> @llvm.vp.fptrunc.nxv2f32.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)

  %nxv4f32_nxv4f16 = fptrunc <vscale x 4 x float> undef to <vscale x 4 x half>
  %nxv4f64_nxv4f16 = fptrunc <vscale x 4 x double> undef to <vscale x 4 x half>
  %nxv4f64_nxv4f32 = fptrunc <vscale x 4 x double> undef to <vscale x 4 x float>

  %vp_nxv4f32_nxv4f16 = call <vscale x 4 x half> @llvm.vp.fptrunc.nxv4f16.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4f64_nxv4f16 = call <vscale x 4 x half> @llvm.vp.fptrunc.nxv4f16.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4f64_nxv4f32 = call <vscale x 4 x float> @llvm.vp.fptrunc.nxv4f32.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)

  %nxv8f32_nxv8f16 = fptrunc <vscale x 8 x float> undef to <vscale x 8 x half>
  %nxv8f64_nxv8f16 = fptrunc <vscale x 8 x double> undef to <vscale x 8 x half>
  %nxv8f64_nxv8f32 = fptrunc <vscale x 8 x double> undef to <vscale x 8 x float>

  %vp_nxv8f32_nxv8f16 = call <vscale x 8 x half> @llvm.vp.fptrunc.nxv8f16.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8f64_nxv8f16 = call <vscale x 8 x half> @llvm.vp.fptrunc.nxv8f16.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8f64_nxv8f32 = call <vscale x 8 x float> @llvm.vp.fptrunc.nxv8f32.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)

  %nxv16f32_nxv16f16 = fptrunc <vscale x 16 x float> undef to <vscale x 16 x half>
  %nxv16f64_nxv16f16 = fptrunc <vscale x 16 x double> undef to <vscale x 16 x half>
  %nxv16f64_nxv16f32 = fptrunc <vscale x 16 x double> undef to <vscale x 16 x float>

  %vp_nxv16f32_nxv16f16 = call <vscale x 16 x half> @llvm.vp.fptrunc.nxv16f16.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16f64_nxv16f16 = call <vscale x 16 x half> @llvm.vp.fptrunc.nxv16f16.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16f64_nxv16f32 = call <vscale x 16 x float> @llvm.vp.fptrunc.nxv16f32.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)

  %nxv32f32_nxv32f16 = fptrunc <vscale x 32 x float> undef to <vscale x 32 x half>
  %nxv32f64_nxv32f16 = fptrunc <vscale x 32 x double> undef to <vscale x 32 x half>
  %nxv32f64_nxv32f32 = fptrunc <vscale x 32 x double> undef to <vscale x 32 x float>

  %vp_nxv32f32_nxv32f16 = call <vscale x 32 x half> @llvm.vp.fptrunc.nxv32f16.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32f64_nxv32f16 = call <vscale x 32 x half> @llvm.vp.fptrunc.nxv32f16.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32f64_nxv32f32 = call <vscale x 32 x float> @llvm.vp.fptrunc.nxv32f32.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)

  %nxv64f32_nxv64f16 = fptrunc <vscale x 64 x float> undef to <vscale x 64 x half>
  %nxv64f64_nxv64f16 = fptrunc <vscale x 64 x double> undef to <vscale x 64 x half>
  %nxv64f64_nxv64f32 = fptrunc <vscale x 64 x double> undef to <vscale x 64 x float>

  %vp_nxv64f32_nxv64f16 = call <vscale x 64 x half> @llvm.vp.fptrunc.nxv64f16.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64f64_nxv64f16 = call <vscale x 64 x half> @llvm.vp.fptrunc.nxv64f16.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64f64_nxv64f32 = call <vscale x 64 x float> @llvm.vp.fptrunc.nxv64f32.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)

  ret void
}

define void @fptosi() {
; ZVE64D-LABEL: 'fptosi'
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2f32_v2i8 = fptosi <2 x float> undef to <2 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2f64_v2i8 = fptosi <2 x double> undef to <2 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2f32_v2i16 = fptosi <2 x float> undef to <2 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2f64_v2i16 = fptosi <2 x double> undef to <2 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2f32_v2i32 = fptosi <2 x float> undef to <2 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2f64_v2i32 = fptosi <2 x double> undef to <2 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2f32_v2i64 = fptosi <2 x float> undef to <2 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2f64_v2i64 = fptosi <2 x double> undef to <2 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2f32_v2i1 = fptosi <2 x float> undef to <2 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2f64_v2i1 = fptosi <2 x double> undef to <2 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2f32_v2i8 = call <2 x i8> @llvm.vp.fptosi.v2i8.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2f64_v2i8 = call <2 x i8> @llvm.vp.fptosi.v2i8.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2f32_v2i16 = call <2 x i16> @llvm.vp.fptosi.v2i16.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2f64_v2i16 = call <2 x i16> @llvm.vp.fptosi.v2i16.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2f32_v2i32 = call <2 x i32> @llvm.vp.fptosi.v2i32.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2f64_v2i32 = call <2 x i32> @llvm.vp.fptosi.v2i32.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2f32_v2i64 = call <2 x i64> @llvm.vp.fptosi.v2i64.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2f64_v2i64 = call <2 x i64> @llvm.vp.fptosi.v2i64.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2f32_v2i1 = call <2 x i1> @llvm.vp.fptosi.v2i1.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2f64_v2i1 = call <2 x i1> @llvm.vp.fptosi.v2i1.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4f32_v4i8 = fptosi <4 x float> undef to <4 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v4f64_v4i8 = fptosi <4 x double> undef to <4 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4f32_v4i16 = fptosi <4 x float> undef to <4 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4f64_v4i16 = fptosi <4 x double> undef to <4 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4f32_v4i32 = fptosi <4 x float> undef to <4 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4f64_v4i32 = fptosi <4 x double> undef to <4 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4f32_v4i64 = fptosi <4 x float> undef to <4 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4f64_v4i64 = fptosi <4 x double> undef to <4 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v4f32_v4i1 = fptosi <4 x float> undef to <4 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v4f64_v4i1 = fptosi <4 x double> undef to <4 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4f32_v4i8 = call <4 x i8> @llvm.vp.fptosi.v4i8.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v4f64_v4i8 = call <4 x i8> @llvm.vp.fptosi.v4i8.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v4f32_v4i16 = call <4 x i16> @llvm.vp.fptosi.v4i16.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4f64_v4i16 = call <4 x i16> @llvm.vp.fptosi.v4i16.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v4f32_v4i32 = call <4 x i32> @llvm.vp.fptosi.v4i32.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v4f64_v4i32 = call <4 x i32> @llvm.vp.fptosi.v4i32.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4f32_v4i64 = call <4 x i64> @llvm.vp.fptosi.v4i64.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4f64_v4i64 = call <4 x i64> @llvm.vp.fptosi.v4i64.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v4f32_v4i1 = call <4 x i1> @llvm.vp.fptosi.v4i1.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v4f64_v4i1 = call <4 x i1> @llvm.vp.fptosi.v4i1.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8f32_v8i8 = fptosi <8 x float> undef to <8 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8f64_v8i8 = fptosi <8 x double> undef to <8 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8f32_v8i16 = fptosi <8 x float> undef to <8 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v8f64_v8i16 = fptosi <8 x double> undef to <8 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8f32_v8i32 = fptosi <8 x float> undef to <8 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8f64_v8i32 = fptosi <8 x double> undef to <8 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8f32_v8i64 = fptosi <8 x float> undef to <8 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8f64_v8i64 = fptosi <8 x double> undef to <8 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v8f32_v8i1 = fptosi <8 x float> undef to <8 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %v8f64_v8i1 = fptosi <8 x double> undef to <8 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8f32_v8i8 = call <8 x i8> @llvm.vp.fptosi.v8i8.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v8f64_v8i8 = call <8 x i8> @llvm.vp.fptosi.v8i8.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v8f32_v8i16 = call <8 x i16> @llvm.vp.fptosi.v8i16.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v8f64_v8i16 = call <8 x i16> @llvm.vp.fptosi.v8i16.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8f32_v8i32 = call <8 x i32> @llvm.vp.fptosi.v8i32.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8f64_v8i32 = call <8 x i32> @llvm.vp.fptosi.v8i32.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v8f32_v8i64 = call <8 x i64> @llvm.vp.fptosi.v8i64.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v8f64_v8i64 = call <8 x i64> @llvm.vp.fptosi.v8i64.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v8f32_v8i1 = call <8 x i1> @llvm.vp.fptosi.v8i1.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_v8f64_v8i1 = call <8 x i1> @llvm.vp.fptosi.v8i1.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v16f32_v16i8 = fptosi <16 x float> undef to <16 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v16f64_v16i8 = fptosi <16 x double> undef to <16 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v16f32_v16i16 = fptosi <16 x float> undef to <16 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %v16f64_v16i16 = fptosi <16 x double> undef to <16 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v16f32_v16i32 = fptosi <16 x float> undef to <16 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v16f64_v16i32 = fptosi <16 x double> undef to <16 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v16f32_v16i64 = fptosi <16 x float> undef to <16 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v16f64_v16i64 = fptosi <16 x double> undef to <16 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %v16f32_v16i1 = fptosi <16 x float> undef to <16 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %v16f64_v16i1 = fptosi <16 x double> undef to <16 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v16f32_v16i8 = call <16 x i8> @llvm.vp.fptosi.v16i8.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v16f64_v16i8 = call <16 x i8> @llvm.vp.fptosi.v16i8.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v16f32_v16i16 = call <16 x i16> @llvm.vp.fptosi.v16i16.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_v16f64_v16i16 = call <16 x i16> @llvm.vp.fptosi.v16i16.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v16f32_v16i32 = call <16 x i32> @llvm.vp.fptosi.v16i32.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v16f64_v16i32 = call <16 x i32> @llvm.vp.fptosi.v16i32.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v16f32_v16i64 = call <16 x i64> @llvm.vp.fptosi.v16i64.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v16f64_v16i64 = call <16 x i64> @llvm.vp.fptosi.v16i64.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_v16f32_v16i1 = call <16 x i1> @llvm.vp.fptosi.v16i1.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %vp_v16f64_v16i1 = call <16 x i1> @llvm.vp.fptosi.v16i1.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %v32f32_v32i8 = fptosi <32 x float> undef to <32 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v32f64_v32i8 = fptosi <32 x double> undef to <32 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v32f32_v32i16 = fptosi <32 x float> undef to <32 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %v32f64_v32i16 = fptosi <32 x double> undef to <32 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v32f32_v32i32 = fptosi <32 x float> undef to <32 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %v32f64_v32i32 = fptosi <32 x double> undef to <32 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %v32f32_v32i64 = fptosi <32 x float> undef to <32 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v32f64_v32i64 = fptosi <32 x double> undef to <32 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %v32f32_v32i1 = fptosi <32 x float> undef to <32 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %v32f64_v32i1 = fptosi <32 x double> undef to <32 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_v32f32_v32i8 = call <32 x i8> @llvm.vp.fptosi.v32i8.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v32f64_v32i8 = call <32 x i8> @llvm.vp.fptosi.v32i8.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v32f32_v32i16 = call <32 x i16> @llvm.vp.fptosi.v32i16.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %vp_v32f64_v32i16 = call <32 x i16> @llvm.vp.fptosi.v32i16.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v32f32_v32i32 = call <32 x i32> @llvm.vp.fptosi.v32i32.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %vp_v32f64_v32i32 = call <32 x i32> @llvm.vp.fptosi.v32i32.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_v32f32_v32i64 = call <32 x i64> @llvm.vp.fptosi.v32i64.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v32f64_v32i64 = call <32 x i64> @llvm.vp.fptosi.v32i64.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %vp_v32f32_v32i1 = call <32 x i1> @llvm.vp.fptosi.v32i1.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_v32f64_v32i1 = call <32 x i1> @llvm.vp.fptosi.v32i1.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %v64f32_v64i8 = fptosi <64 x float> undef to <64 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v64f64_v64i8 = fptosi <64 x double> undef to <64 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %v64f32_v64i16 = fptosi <64 x float> undef to <64 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %v64f64_v64i16 = fptosi <64 x double> undef to <64 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v64f32_v64i32 = fptosi <64 x float> undef to <64 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %v64f64_v64i32 = fptosi <64 x double> undef to <64 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %v64f32_v64i64 = fptosi <64 x float> undef to <64 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v64f64_v64i64 = fptosi <64 x double> undef to <64 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %v64f32_v64i1 = fptosi <64 x float> undef to <64 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %v64f64_v64i1 = fptosi <64 x double> undef to <64 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %vp_v64f32_v64i8 = call <64 x i8> @llvm.vp.fptosi.v64i8.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v64f64_v64i8 = call <64 x i8> @llvm.vp.fptosi.v64i8.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %vp_v64f32_v64i16 = call <64 x i16> @llvm.vp.fptosi.v64i16.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %vp_v64f64_v64i16 = call <64 x i16> @llvm.vp.fptosi.v64i16.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v64f32_v64i32 = call <64 x i32> @llvm.vp.fptosi.v64i32.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %vp_v64f64_v64i32 = call <64 x i32> @llvm.vp.fptosi.v64i32.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_v64f32_v64i64 = call <64 x i64> @llvm.vp.fptosi.v64i64.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v64f64_v64i64 = call <64 x i64> @llvm.vp.fptosi.v64i64.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_v64f32_v64i1 = call <64 x i1> @llvm.vp.fptosi.v64i1.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %vp_v64f64_v64i1 = call <64 x i1> @llvm.vp.fptosi.v64i1.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %v128f32_v128i8 = fptosi <128 x float> undef to <128 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v128f64_v128i8 = fptosi <128 x double> undef to <128 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %v128f32_v128i16 = fptosi <128 x float> undef to <128 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %v128f64_v128i16 = fptosi <128 x double> undef to <128 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v128f32_v128i32 = fptosi <128 x float> undef to <128 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %v128f64_v128i32 = fptosi <128 x double> undef to <128 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %v128f32_v128i64 = fptosi <128 x float> undef to <128 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v128f64_v128i64 = fptosi <128 x double> undef to <128 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %v128f32_v128i1 = fptosi <128 x float> undef to <128 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 103 for instruction: %v128f64_v128i1 = fptosi <128 x double> undef to <128 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %vp_v128f32_v128i8 = call <128 x i8> @llvm.vp.fptosi.v128i8.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v128f64_v128i8 = call <128 x i8> @llvm.vp.fptosi.v128i8.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %vp_v128f32_v128i16 = call <128 x i16> @llvm.vp.fptosi.v128i16.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %vp_v128f64_v128i16 = call <128 x i16> @llvm.vp.fptosi.v128i16.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v128f32_v128i32 = call <128 x i32> @llvm.vp.fptosi.v128i32.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %vp_v128f64_v128i32 = call <128 x i32> @llvm.vp.fptosi.v128i32.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %vp_v128f32_v128i64 = call <128 x i64> @llvm.vp.fptosi.v128i64.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v128f64_v128i64 = call <128 x i64> @llvm.vp.fptosi.v128i64.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %vp_v128f32_v128i1 = call <128 x i1> @llvm.vp.fptosi.v128i1.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 103 for instruction: %vp_v128f64_v128i1 = call <128 x i1> @llvm.vp.fptosi.v128i1.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1f32_nxv1i8 = fptosi <vscale x 1 x float> undef to <vscale x 1 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv1f64_nxv1i8 = fptosi <vscale x 1 x double> undef to <vscale x 1 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1f32_nxv1i16 = fptosi <vscale x 1 x float> undef to <vscale x 1 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1f64_nxv1i16 = fptosi <vscale x 1 x double> undef to <vscale x 1 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1f32_nxv1i32 = fptosi <vscale x 1 x float> undef to <vscale x 1 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1f64_nxv1i32 = fptosi <vscale x 1 x double> undef to <vscale x 1 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1f32_nxv1i64 = fptosi <vscale x 1 x float> undef to <vscale x 1 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1f64_nxv1i64 = fptosi <vscale x 1 x double> undef to <vscale x 1 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv1f32_nxv1i1 = fptosi <vscale x 1 x float> undef to <vscale x 1 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv1f64_nxv1i1 = fptosi <vscale x 1 x double> undef to <vscale x 1 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1f32_nxv1i8 = call <vscale x 1 x i8> @llvm.vp.fptosi.nxv1i8.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv1f64_nxv1i8 = call <vscale x 1 x i8> @llvm.vp.fptosi.nxv1i8.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1f32_nxv1i16 = call <vscale x 1 x i16> @llvm.vp.fptosi.nxv1i16.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1f64_nxv1i16 = call <vscale x 1 x i16> @llvm.vp.fptosi.nxv1i16.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1f32_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.fptosi.nxv1i32.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1f64_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.fptosi.nxv1i32.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1f32_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.fptosi.nxv1i64.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1f64_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.fptosi.nxv1i64.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv1f32_nxv1i1 = call <vscale x 1 x i1> @llvm.vp.fptosi.nxv1i1.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv1f64_nxv1i1 = call <vscale x 1 x i1> @llvm.vp.fptosi.nxv1i1.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2f32_nxv2i8 = fptosi <vscale x 2 x float> undef to <vscale x 2 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv2f64_nxv2i8 = fptosi <vscale x 2 x double> undef to <vscale x 2 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv2f32_nxv2i16 = fptosi <vscale x 2 x float> undef to <vscale x 2 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2f64_nxv2i16 = fptosi <vscale x 2 x double> undef to <vscale x 2 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv2f32_nxv2i32 = fptosi <vscale x 2 x float> undef to <vscale x 2 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv2f64_nxv2i32 = fptosi <vscale x 2 x double> undef to <vscale x 2 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2f32_nxv2i64 = fptosi <vscale x 2 x float> undef to <vscale x 2 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2f64_nxv2i64 = fptosi <vscale x 2 x double> undef to <vscale x 2 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv2f32_nxv2i1 = fptosi <vscale x 2 x float> undef to <vscale x 2 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv2f64_nxv2i1 = fptosi <vscale x 2 x double> undef to <vscale x 2 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2f32_nxv2i8 = call <vscale x 2 x i8> @llvm.vp.fptosi.nxv2i8.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv2f64_nxv2i8 = call <vscale x 2 x i8> @llvm.vp.fptosi.nxv2i8.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv2f32_nxv2i16 = call <vscale x 2 x i16> @llvm.vp.fptosi.nxv2i16.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2f64_nxv2i16 = call <vscale x 2 x i16> @llvm.vp.fptosi.nxv2i16.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv2f32_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.fptosi.nxv2i32.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv2f64_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.fptosi.nxv2i32.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2f32_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.fptosi.nxv2i64.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2f64_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.fptosi.nxv2i64.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv2f32_nxv2i1 = call <vscale x 2 x i1> @llvm.vp.fptosi.nxv2i1.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv2f64_nxv2i1 = call <vscale x 2 x i1> @llvm.vp.fptosi.nxv2i1.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4f32_nxv4i8 = fptosi <vscale x 4 x float> undef to <vscale x 4 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv4f64_nxv4i8 = fptosi <vscale x 4 x double> undef to <vscale x 4 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv4f32_nxv4i16 = fptosi <vscale x 4 x float> undef to <vscale x 4 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv4f64_nxv4i16 = fptosi <vscale x 4 x double> undef to <vscale x 4 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4f32_nxv4i32 = fptosi <vscale x 4 x float> undef to <vscale x 4 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4f64_nxv4i32 = fptosi <vscale x 4 x double> undef to <vscale x 4 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv4f32_nxv4i64 = fptosi <vscale x 4 x float> undef to <vscale x 4 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv4f64_nxv4i64 = fptosi <vscale x 4 x double> undef to <vscale x 4 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv4f32_nxv4i1 = fptosi <vscale x 4 x float> undef to <vscale x 4 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %nxv4f64_nxv4i1 = fptosi <vscale x 4 x double> undef to <vscale x 4 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4f32_nxv4i8 = call <vscale x 4 x i8> @llvm.vp.fptosi.nxv4i8.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv4f64_nxv4i8 = call <vscale x 4 x i8> @llvm.vp.fptosi.nxv4i8.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv4f32_nxv4i16 = call <vscale x 4 x i16> @llvm.vp.fptosi.nxv4i16.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv4f64_nxv4i16 = call <vscale x 4 x i16> @llvm.vp.fptosi.nxv4i16.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4f32_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.fptosi.nxv4i32.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4f64_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.fptosi.nxv4i32.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv4f32_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.fptosi.nxv4i64.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv4f64_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.fptosi.nxv4i64.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv4f32_nxv4i1 = call <vscale x 4 x i1> @llvm.vp.fptosi.nxv4i1.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_nxv4f64_nxv4i1 = call <vscale x 4 x i1> @llvm.vp.fptosi.nxv4i1.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv8f32_nxv8i8 = fptosi <vscale x 8 x float> undef to <vscale x 8 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %nxv8f64_nxv8i8 = fptosi <vscale x 8 x double> undef to <vscale x 8 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv8f32_nxv8i16 = fptosi <vscale x 8 x float> undef to <vscale x 8 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %nxv8f64_nxv8i16 = fptosi <vscale x 8 x double> undef to <vscale x 8 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv8f32_nxv8i32 = fptosi <vscale x 8 x float> undef to <vscale x 8 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv8f64_nxv8i32 = fptosi <vscale x 8 x double> undef to <vscale x 8 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv8f32_nxv8i64 = fptosi <vscale x 8 x float> undef to <vscale x 8 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv8f64_nxv8i64 = fptosi <vscale x 8 x double> undef to <vscale x 8 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %nxv8f32_nxv8i1 = fptosi <vscale x 8 x float> undef to <vscale x 8 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %nxv8f64_nxv8i1 = fptosi <vscale x 8 x double> undef to <vscale x 8 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv8f32_nxv8i8 = call <vscale x 8 x i8> @llvm.vp.fptosi.nxv8i8.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_nxv8f64_nxv8i8 = call <vscale x 8 x i8> @llvm.vp.fptosi.nxv8i8.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv8f32_nxv8i16 = call <vscale x 8 x i16> @llvm.vp.fptosi.nxv8i16.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_nxv8f64_nxv8i16 = call <vscale x 8 x i16> @llvm.vp.fptosi.nxv8i16.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv8f32_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.fptosi.nxv8i32.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv8f64_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.fptosi.nxv8i32.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv8f32_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.fptosi.nxv8i64.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv8f64_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.fptosi.nxv8i64.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_nxv8f32_nxv8i1 = call <vscale x 8 x i1> @llvm.vp.fptosi.nxv8i1.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %vp_nxv8f64_nxv8i1 = call <vscale x 8 x i1> @llvm.vp.fptosi.nxv8i1.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %nxv16f32_nxv16i8 = fptosi <vscale x 16 x float> undef to <vscale x 16 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %nxv16f64_nxv16i8 = fptosi <vscale x 16 x double> undef to <vscale x 16 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv16f32_nxv16i16 = fptosi <vscale x 16 x float> undef to <vscale x 16 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %nxv16f64_nxv16i16 = fptosi <vscale x 16 x double> undef to <vscale x 16 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv16f32_nxv16i32 = fptosi <vscale x 16 x float> undef to <vscale x 16 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %nxv16f64_nxv16i32 = fptosi <vscale x 16 x double> undef to <vscale x 16 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %nxv16f32_nxv16i64 = fptosi <vscale x 16 x float> undef to <vscale x 16 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv16f64_nxv16i64 = fptosi <vscale x 16 x double> undef to <vscale x 16 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %nxv16f32_nxv16i1 = fptosi <vscale x 16 x float> undef to <vscale x 16 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %nxv16f64_nxv16i1 = fptosi <vscale x 16 x double> undef to <vscale x 16 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_nxv16f32_nxv16i8 = call <vscale x 16 x i8> @llvm.vp.fptosi.nxv16i8.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_nxv16f64_nxv16i8 = call <vscale x 16 x i8> @llvm.vp.fptosi.nxv16i8.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv16f32_nxv16i16 = call <vscale x 16 x i16> @llvm.vp.fptosi.nxv16i16.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %vp_nxv16f64_nxv16i16 = call <vscale x 16 x i16> @llvm.vp.fptosi.nxv16i16.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv16f32_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.fptosi.nxv16i32.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %vp_nxv16f64_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.fptosi.nxv16i32.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_nxv16f32_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.fptosi.nxv16i64.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv16f64_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.fptosi.nxv16i64.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %vp_nxv16f32_nxv16i1 = call <vscale x 16 x i1> @llvm.vp.fptosi.nxv16i1.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_nxv16f64_nxv16i1 = call <vscale x 16 x i1> @llvm.vp.fptosi.nxv16i1.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %nxv32f32_nxv32i8 = fptosi <vscale x 32 x float> undef to <vscale x 32 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %nxv32f64_nxv32i8 = fptosi <vscale x 32 x double> undef to <vscale x 32 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %nxv32f32_nxv32i16 = fptosi <vscale x 32 x float> undef to <vscale x 32 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %nxv32f64_nxv32i16 = fptosi <vscale x 32 x double> undef to <vscale x 32 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv32f32_nxv32i32 = fptosi <vscale x 32 x float> undef to <vscale x 32 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %nxv32f64_nxv32i32 = fptosi <vscale x 32 x double> undef to <vscale x 32 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %nxv32f32_nxv32i64 = fptosi <vscale x 32 x float> undef to <vscale x 32 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv32f64_nxv32i64 = fptosi <vscale x 32 x double> undef to <vscale x 32 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %nxv32f32_nxv32i1 = fptosi <vscale x 32 x float> undef to <vscale x 32 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %nxv32f64_nxv32i1 = fptosi <vscale x 32 x double> undef to <vscale x 32 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %vp_nxv32f32_nxv32i8 = call <vscale x 32 x i8> @llvm.vp.fptosi.nxv32i8.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_nxv32f64_nxv32i8 = call <vscale x 32 x i8> @llvm.vp.fptosi.nxv32i8.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %vp_nxv32f32_nxv32i16 = call <vscale x 32 x i16> @llvm.vp.fptosi.nxv32i16.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %vp_nxv32f64_nxv32i16 = call <vscale x 32 x i16> @llvm.vp.fptosi.nxv32i16.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv32f32_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.fptosi.nxv32i32.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %vp_nxv32f64_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.fptosi.nxv32i32.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_nxv32f32_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.fptosi.nxv32i64.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv32f64_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.fptosi.nxv32i64.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_nxv32f32_nxv32i1 = call <vscale x 32 x i1> @llvm.vp.fptosi.nxv32i1.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %vp_nxv32f64_nxv32i1 = call <vscale x 32 x i1> @llvm.vp.fptosi.nxv32i1.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %nxv64f32_nxv64i8 = fptosi <vscale x 64 x float> undef to <vscale x 64 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %nxv64f64_nxv64i8 = fptosi <vscale x 64 x double> undef to <vscale x 64 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %nxv64f32_nxv64i16 = fptosi <vscale x 64 x float> undef to <vscale x 64 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %nxv64f64_nxv64i16 = fptosi <vscale x 64 x double> undef to <vscale x 64 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv64f32_nxv64i32 = fptosi <vscale x 64 x float> undef to <vscale x 64 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %nxv64f64_nxv64i32 = fptosi <vscale x 64 x double> undef to <vscale x 64 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %nxv64f32_nxv64i64 = fptosi <vscale x 64 x float> undef to <vscale x 64 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv64f64_nxv64i64 = fptosi <vscale x 64 x double> undef to <vscale x 64 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %nxv64f32_nxv64i1 = fptosi <vscale x 64 x float> undef to <vscale x 64 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 103 for instruction: %nxv64f64_nxv64i1 = fptosi <vscale x 64 x double> undef to <vscale x 64 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %vp_nxv64f32_nxv64i8 = call <vscale x 64 x i8> @llvm.vp.fptosi.nxv64i8.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_nxv64f64_nxv64i8 = call <vscale x 64 x i8> @llvm.vp.fptosi.nxv64i8.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %vp_nxv64f32_nxv64i16 = call <vscale x 64 x i16> @llvm.vp.fptosi.nxv64i16.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %vp_nxv64f64_nxv64i16 = call <vscale x 64 x i16> @llvm.vp.fptosi.nxv64i16.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv64f32_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.fptosi.nxv64i32.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %vp_nxv64f64_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.fptosi.nxv64i32.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %vp_nxv64f32_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.fptosi.nxv64i64.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv64f64_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.fptosi.nxv64i64.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %vp_nxv64f32_nxv64i1 = call <vscale x 64 x i1> @llvm.vp.fptosi.nxv64i1.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 103 for instruction: %vp_nxv64f64_nxv64i1 = call <vscale x 64 x i1> @llvm.vp.fptosi.nxv64i1.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; ZVE64X-LABEL: 'fptosi'
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v2f32_v2i8 = fptosi <2 x float> undef to <2 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v2f64_v2i8 = fptosi <2 x double> undef to <2 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v2f32_v2i16 = fptosi <2 x float> undef to <2 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v2f64_v2i16 = fptosi <2 x double> undef to <2 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v2f32_v2i32 = fptosi <2 x float> undef to <2 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v2f64_v2i32 = fptosi <2 x double> undef to <2 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2f32_v2i64 = fptosi <2 x float> undef to <2 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2f64_v2i64 = fptosi <2 x double> undef to <2 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %v2f32_v2i1 = fptosi <2 x float> undef to <2 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %v2f64_v2i1 = fptosi <2 x double> undef to <2 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v2f32_v2i8 = call <2 x i8> @llvm.vp.fptosi.v2i8.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v2f64_v2i8 = call <2 x i8> @llvm.vp.fptosi.v2i8.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v2f32_v2i16 = call <2 x i16> @llvm.vp.fptosi.v2i16.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v2f64_v2i16 = call <2 x i16> @llvm.vp.fptosi.v2i16.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v2f32_v2i32 = call <2 x i32> @llvm.vp.fptosi.v2i32.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v2f64_v2i32 = call <2 x i32> @llvm.vp.fptosi.v2i32.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2f32_v2i64 = call <2 x i64> @llvm.vp.fptosi.v2i64.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2f64_v2i64 = call <2 x i64> @llvm.vp.fptosi.v2i64.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %vp_v2f32_v2i1 = call <2 x i1> @llvm.vp.fptosi.v2i1.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %vp_v2f64_v2i1 = call <2 x i1> @llvm.vp.fptosi.v2i1.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v4f32_v4i8 = fptosi <4 x float> undef to <4 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v4f64_v4i8 = fptosi <4 x double> undef to <4 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v4f32_v4i16 = fptosi <4 x float> undef to <4 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v4f64_v4i16 = fptosi <4 x double> undef to <4 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v4f32_v4i32 = fptosi <4 x float> undef to <4 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v4f64_v4i32 = fptosi <4 x double> undef to <4 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4f32_v4i64 = fptosi <4 x float> undef to <4 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4f64_v4i64 = fptosi <4 x double> undef to <4 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 39 for instruction: %v4f32_v4i1 = fptosi <4 x float> undef to <4 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 39 for instruction: %v4f64_v4i1 = fptosi <4 x double> undef to <4 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v4f32_v4i8 = call <4 x i8> @llvm.vp.fptosi.v4i8.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v4f64_v4i8 = call <4 x i8> @llvm.vp.fptosi.v4i8.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v4f32_v4i16 = call <4 x i16> @llvm.vp.fptosi.v4i16.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v4f64_v4i16 = call <4 x i16> @llvm.vp.fptosi.v4i16.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v4f32_v4i32 = call <4 x i32> @llvm.vp.fptosi.v4i32.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v4f64_v4i32 = call <4 x i32> @llvm.vp.fptosi.v4i32.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v4f32_v4i64 = call <4 x i64> @llvm.vp.fptosi.v4i64.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v4f64_v4i64 = call <4 x i64> @llvm.vp.fptosi.v4i64.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 39 for instruction: %vp_v4f32_v4i1 = call <4 x i1> @llvm.vp.fptosi.v4i1.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 39 for instruction: %vp_v4f64_v4i1 = call <4 x i1> @llvm.vp.fptosi.v4i1.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v8f32_v8i8 = fptosi <8 x float> undef to <8 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v8f64_v8i8 = fptosi <8 x double> undef to <8 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v8f32_v8i16 = fptosi <8 x float> undef to <8 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v8f64_v8i16 = fptosi <8 x double> undef to <8 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v8f32_v8i32 = fptosi <8 x float> undef to <8 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v8f64_v8i32 = fptosi <8 x double> undef to <8 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v8f32_v8i64 = fptosi <8 x float> undef to <8 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v8f64_v8i64 = fptosi <8 x double> undef to <8 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 79 for instruction: %v8f32_v8i1 = fptosi <8 x float> undef to <8 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 79 for instruction: %v8f64_v8i1 = fptosi <8 x double> undef to <8 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v8f32_v8i8 = call <8 x i8> @llvm.vp.fptosi.v8i8.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v8f64_v8i8 = call <8 x i8> @llvm.vp.fptosi.v8i8.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v8f32_v8i16 = call <8 x i16> @llvm.vp.fptosi.v8i16.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v8f64_v8i16 = call <8 x i16> @llvm.vp.fptosi.v8i16.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v8f32_v8i32 = call <8 x i32> @llvm.vp.fptosi.v8i32.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v8f64_v8i32 = call <8 x i32> @llvm.vp.fptosi.v8i32.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v8f32_v8i64 = call <8 x i64> @llvm.vp.fptosi.v8i64.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v8f64_v8i64 = call <8 x i64> @llvm.vp.fptosi.v8i64.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 79 for instruction: %vp_v8f32_v8i1 = call <8 x i1> @llvm.vp.fptosi.v8i1.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 79 for instruction: %vp_v8f64_v8i1 = call <8 x i1> @llvm.vp.fptosi.v8i1.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v16f32_v16i8 = fptosi <16 x float> undef to <16 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v16f64_v16i8 = fptosi <16 x double> undef to <16 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v16f32_v16i16 = fptosi <16 x float> undef to <16 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v16f64_v16i16 = fptosi <16 x double> undef to <16 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v16f32_v16i32 = fptosi <16 x float> undef to <16 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v16f64_v16i32 = fptosi <16 x double> undef to <16 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v16f32_v16i64 = fptosi <16 x float> undef to <16 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v16f64_v16i64 = fptosi <16 x double> undef to <16 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 159 for instruction: %v16f32_v16i1 = fptosi <16 x float> undef to <16 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 159 for instruction: %v16f64_v16i1 = fptosi <16 x double> undef to <16 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v16f32_v16i8 = call <16 x i8> @llvm.vp.fptosi.v16i8.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v16f64_v16i8 = call <16 x i8> @llvm.vp.fptosi.v16i8.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v16f32_v16i16 = call <16 x i16> @llvm.vp.fptosi.v16i16.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v16f64_v16i16 = call <16 x i16> @llvm.vp.fptosi.v16i16.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v16f32_v16i32 = call <16 x i32> @llvm.vp.fptosi.v16i32.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v16f64_v16i32 = call <16 x i32> @llvm.vp.fptosi.v16i32.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v16f32_v16i64 = call <16 x i64> @llvm.vp.fptosi.v16i64.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v16f64_v16i64 = call <16 x i64> @llvm.vp.fptosi.v16i64.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 159 for instruction: %vp_v16f32_v16i1 = call <16 x i1> @llvm.vp.fptosi.v16i1.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 159 for instruction: %vp_v16f64_v16i1 = call <16 x i1> @llvm.vp.fptosi.v16i1.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %v32f32_v32i8 = fptosi <32 x float> undef to <32 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %v32f64_v32i8 = fptosi <32 x double> undef to <32 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %v32f32_v32i16 = fptosi <32 x float> undef to <32 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %v32f64_v32i16 = fptosi <32 x double> undef to <32 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %v32f32_v32i32 = fptosi <32 x float> undef to <32 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %v32f64_v32i32 = fptosi <32 x double> undef to <32 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 62 for instruction: %v32f32_v32i64 = fptosi <32 x float> undef to <32 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 62 for instruction: %v32f64_v32i64 = fptosi <32 x double> undef to <32 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 319 for instruction: %v32f32_v32i1 = fptosi <32 x float> undef to <32 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 319 for instruction: %v32f64_v32i1 = fptosi <32 x double> undef to <32 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %vp_v32f32_v32i8 = call <32 x i8> @llvm.vp.fptosi.v32i8.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %vp_v32f64_v32i8 = call <32 x i8> @llvm.vp.fptosi.v32i8.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %vp_v32f32_v32i16 = call <32 x i16> @llvm.vp.fptosi.v32i16.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %vp_v32f64_v32i16 = call <32 x i16> @llvm.vp.fptosi.v32i16.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %vp_v32f32_v32i32 = call <32 x i32> @llvm.vp.fptosi.v32i32.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %vp_v32f64_v32i32 = call <32 x i32> @llvm.vp.fptosi.v32i32.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 62 for instruction: %vp_v32f32_v32i64 = call <32 x i64> @llvm.vp.fptosi.v32i64.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 62 for instruction: %vp_v32f64_v32i64 = call <32 x i64> @llvm.vp.fptosi.v32i64.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 319 for instruction: %vp_v32f32_v32i1 = call <32 x i1> @llvm.vp.fptosi.v32i1.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 319 for instruction: %vp_v32f64_v32i1 = call <32 x i1> @llvm.vp.fptosi.v32i1.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %v64f32_v64i8 = fptosi <64 x float> undef to <64 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %v64f64_v64i8 = fptosi <64 x double> undef to <64 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %v64f32_v64i16 = fptosi <64 x float> undef to <64 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %v64f64_v64i16 = fptosi <64 x double> undef to <64 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 254 for instruction: %v64f32_v64i32 = fptosi <64 x float> undef to <64 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 254 for instruction: %v64f64_v64i32 = fptosi <64 x double> undef to <64 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 124 for instruction: %v64f32_v64i64 = fptosi <64 x float> undef to <64 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 124 for instruction: %v64f64_v64i64 = fptosi <64 x double> undef to <64 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 639 for instruction: %v64f32_v64i1 = fptosi <64 x float> undef to <64 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 639 for instruction: %v64f64_v64i1 = fptosi <64 x double> undef to <64 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %vp_v64f32_v64i8 = call <64 x i8> @llvm.vp.fptosi.v64i8.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %vp_v64f64_v64i8 = call <64 x i8> @llvm.vp.fptosi.v64i8.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %vp_v64f32_v64i16 = call <64 x i16> @llvm.vp.fptosi.v64i16.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %vp_v64f64_v64i16 = call <64 x i16> @llvm.vp.fptosi.v64i16.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 254 for instruction: %vp_v64f32_v64i32 = call <64 x i32> @llvm.vp.fptosi.v64i32.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 254 for instruction: %vp_v64f64_v64i32 = call <64 x i32> @llvm.vp.fptosi.v64i32.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 124 for instruction: %vp_v64f32_v64i64 = call <64 x i64> @llvm.vp.fptosi.v64i64.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 124 for instruction: %vp_v64f64_v64i64 = call <64 x i64> @llvm.vp.fptosi.v64i64.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 639 for instruction: %vp_v64f32_v64i1 = call <64 x i1> @llvm.vp.fptosi.v64i1.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 639 for instruction: %vp_v64f64_v64i1 = call <64 x i1> @llvm.vp.fptosi.v64i1.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 511 for instruction: %v128f32_v128i8 = fptosi <128 x float> undef to <128 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 511 for instruction: %v128f64_v128i8 = fptosi <128 x double> undef to <128 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 510 for instruction: %v128f32_v128i16 = fptosi <128 x float> undef to <128 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 510 for instruction: %v128f64_v128i16 = fptosi <128 x double> undef to <128 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 508 for instruction: %v128f32_v128i32 = fptosi <128 x float> undef to <128 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 508 for instruction: %v128f64_v128i32 = fptosi <128 x double> undef to <128 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 248 for instruction: %v128f32_v128i64 = fptosi <128 x float> undef to <128 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 248 for instruction: %v128f64_v128i64 = fptosi <128 x double> undef to <128 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 1279 for instruction: %v128f32_v128i1 = fptosi <128 x float> undef to <128 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 1279 for instruction: %v128f64_v128i1 = fptosi <128 x double> undef to <128 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 511 for instruction: %vp_v128f32_v128i8 = call <128 x i8> @llvm.vp.fptosi.v128i8.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 511 for instruction: %vp_v128f64_v128i8 = call <128 x i8> @llvm.vp.fptosi.v128i8.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 510 for instruction: %vp_v128f32_v128i16 = call <128 x i16> @llvm.vp.fptosi.v128i16.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 510 for instruction: %vp_v128f64_v128i16 = call <128 x i16> @llvm.vp.fptosi.v128i16.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 508 for instruction: %vp_v128f32_v128i32 = call <128 x i32> @llvm.vp.fptosi.v128i32.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 508 for instruction: %vp_v128f64_v128i32 = call <128 x i32> @llvm.vp.fptosi.v128i32.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 248 for instruction: %vp_v128f32_v128i64 = call <128 x i64> @llvm.vp.fptosi.v128i64.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 248 for instruction: %vp_v128f64_v128i64 = call <128 x i64> @llvm.vp.fptosi.v128i64.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 1279 for instruction: %vp_v128f32_v128i1 = call <128 x i1> @llvm.vp.fptosi.v128i1.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 1279 for instruction: %vp_v128f64_v128i1 = call <128 x i1> @llvm.vp.fptosi.v128i1.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f32_nxv1i8 = fptosi <vscale x 1 x float> undef to <vscale x 1 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f64_nxv1i8 = fptosi <vscale x 1 x double> undef to <vscale x 1 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f32_nxv1i16 = fptosi <vscale x 1 x float> undef to <vscale x 1 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f64_nxv1i16 = fptosi <vscale x 1 x double> undef to <vscale x 1 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f32_nxv1i32 = fptosi <vscale x 1 x float> undef to <vscale x 1 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f64_nxv1i32 = fptosi <vscale x 1 x double> undef to <vscale x 1 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f32_nxv1i64 = fptosi <vscale x 1 x float> undef to <vscale x 1 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f64_nxv1i64 = fptosi <vscale x 1 x double> undef to <vscale x 1 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f32_nxv1i1 = fptosi <vscale x 1 x float> undef to <vscale x 1 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f64_nxv1i1 = fptosi <vscale x 1 x double> undef to <vscale x 1 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f32_nxv1i8 = call <vscale x 1 x i8> @llvm.vp.fptosi.nxv1i8.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f64_nxv1i8 = call <vscale x 1 x i8> @llvm.vp.fptosi.nxv1i8.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f32_nxv1i16 = call <vscale x 1 x i16> @llvm.vp.fptosi.nxv1i16.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f64_nxv1i16 = call <vscale x 1 x i16> @llvm.vp.fptosi.nxv1i16.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f32_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.fptosi.nxv1i32.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f64_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.fptosi.nxv1i32.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f32_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.fptosi.nxv1i64.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f64_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.fptosi.nxv1i64.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f32_nxv1i1 = call <vscale x 1 x i1> @llvm.vp.fptosi.nxv1i1.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f64_nxv1i1 = call <vscale x 1 x i1> @llvm.vp.fptosi.nxv1i1.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f32_nxv2i8 = fptosi <vscale x 2 x float> undef to <vscale x 2 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f64_nxv2i8 = fptosi <vscale x 2 x double> undef to <vscale x 2 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f32_nxv2i16 = fptosi <vscale x 2 x float> undef to <vscale x 2 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f64_nxv2i16 = fptosi <vscale x 2 x double> undef to <vscale x 2 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f32_nxv2i32 = fptosi <vscale x 2 x float> undef to <vscale x 2 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f64_nxv2i32 = fptosi <vscale x 2 x double> undef to <vscale x 2 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f32_nxv2i64 = fptosi <vscale x 2 x float> undef to <vscale x 2 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f64_nxv2i64 = fptosi <vscale x 2 x double> undef to <vscale x 2 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f32_nxv2i1 = fptosi <vscale x 2 x float> undef to <vscale x 2 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f64_nxv2i1 = fptosi <vscale x 2 x double> undef to <vscale x 2 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f32_nxv2i8 = call <vscale x 2 x i8> @llvm.vp.fptosi.nxv2i8.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f64_nxv2i8 = call <vscale x 2 x i8> @llvm.vp.fptosi.nxv2i8.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f32_nxv2i16 = call <vscale x 2 x i16> @llvm.vp.fptosi.nxv2i16.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f64_nxv2i16 = call <vscale x 2 x i16> @llvm.vp.fptosi.nxv2i16.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f32_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.fptosi.nxv2i32.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f64_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.fptosi.nxv2i32.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f32_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.fptosi.nxv2i64.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f64_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.fptosi.nxv2i64.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f32_nxv2i1 = call <vscale x 2 x i1> @llvm.vp.fptosi.nxv2i1.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f64_nxv2i1 = call <vscale x 2 x i1> @llvm.vp.fptosi.nxv2i1.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f32_nxv4i8 = fptosi <vscale x 4 x float> undef to <vscale x 4 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f64_nxv4i8 = fptosi <vscale x 4 x double> undef to <vscale x 4 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f32_nxv4i16 = fptosi <vscale x 4 x float> undef to <vscale x 4 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f64_nxv4i16 = fptosi <vscale x 4 x double> undef to <vscale x 4 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f32_nxv4i32 = fptosi <vscale x 4 x float> undef to <vscale x 4 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f64_nxv4i32 = fptosi <vscale x 4 x double> undef to <vscale x 4 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f32_nxv4i64 = fptosi <vscale x 4 x float> undef to <vscale x 4 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f64_nxv4i64 = fptosi <vscale x 4 x double> undef to <vscale x 4 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f32_nxv4i1 = fptosi <vscale x 4 x float> undef to <vscale x 4 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f64_nxv4i1 = fptosi <vscale x 4 x double> undef to <vscale x 4 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f32_nxv4i8 = call <vscale x 4 x i8> @llvm.vp.fptosi.nxv4i8.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f64_nxv4i8 = call <vscale x 4 x i8> @llvm.vp.fptosi.nxv4i8.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f32_nxv4i16 = call <vscale x 4 x i16> @llvm.vp.fptosi.nxv4i16.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f64_nxv4i16 = call <vscale x 4 x i16> @llvm.vp.fptosi.nxv4i16.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f32_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.fptosi.nxv4i32.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f64_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.fptosi.nxv4i32.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f32_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.fptosi.nxv4i64.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f64_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.fptosi.nxv4i64.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f32_nxv4i1 = call <vscale x 4 x i1> @llvm.vp.fptosi.nxv4i1.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f64_nxv4i1 = call <vscale x 4 x i1> @llvm.vp.fptosi.nxv4i1.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f32_nxv8i8 = fptosi <vscale x 8 x float> undef to <vscale x 8 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f64_nxv8i8 = fptosi <vscale x 8 x double> undef to <vscale x 8 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f32_nxv8i16 = fptosi <vscale x 8 x float> undef to <vscale x 8 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f64_nxv8i16 = fptosi <vscale x 8 x double> undef to <vscale x 8 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f32_nxv8i32 = fptosi <vscale x 8 x float> undef to <vscale x 8 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f64_nxv8i32 = fptosi <vscale x 8 x double> undef to <vscale x 8 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f32_nxv8i64 = fptosi <vscale x 8 x float> undef to <vscale x 8 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f64_nxv8i64 = fptosi <vscale x 8 x double> undef to <vscale x 8 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f32_nxv8i1 = fptosi <vscale x 8 x float> undef to <vscale x 8 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f64_nxv8i1 = fptosi <vscale x 8 x double> undef to <vscale x 8 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f32_nxv8i8 = call <vscale x 8 x i8> @llvm.vp.fptosi.nxv8i8.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f64_nxv8i8 = call <vscale x 8 x i8> @llvm.vp.fptosi.nxv8i8.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f32_nxv8i16 = call <vscale x 8 x i16> @llvm.vp.fptosi.nxv8i16.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f64_nxv8i16 = call <vscale x 8 x i16> @llvm.vp.fptosi.nxv8i16.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f32_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.fptosi.nxv8i32.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f64_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.fptosi.nxv8i32.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f32_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.fptosi.nxv8i64.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f64_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.fptosi.nxv8i64.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f32_nxv8i1 = call <vscale x 8 x i1> @llvm.vp.fptosi.nxv8i1.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f64_nxv8i1 = call <vscale x 8 x i1> @llvm.vp.fptosi.nxv8i1.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f32_nxv16i8 = fptosi <vscale x 16 x float> undef to <vscale x 16 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f64_nxv16i8 = fptosi <vscale x 16 x double> undef to <vscale x 16 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f32_nxv16i16 = fptosi <vscale x 16 x float> undef to <vscale x 16 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f64_nxv16i16 = fptosi <vscale x 16 x double> undef to <vscale x 16 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f32_nxv16i32 = fptosi <vscale x 16 x float> undef to <vscale x 16 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f64_nxv16i32 = fptosi <vscale x 16 x double> undef to <vscale x 16 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f32_nxv16i64 = fptosi <vscale x 16 x float> undef to <vscale x 16 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f64_nxv16i64 = fptosi <vscale x 16 x double> undef to <vscale x 16 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f32_nxv16i1 = fptosi <vscale x 16 x float> undef to <vscale x 16 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f64_nxv16i1 = fptosi <vscale x 16 x double> undef to <vscale x 16 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f32_nxv16i8 = call <vscale x 16 x i8> @llvm.vp.fptosi.nxv16i8.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f64_nxv16i8 = call <vscale x 16 x i8> @llvm.vp.fptosi.nxv16i8.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f32_nxv16i16 = call <vscale x 16 x i16> @llvm.vp.fptosi.nxv16i16.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f64_nxv16i16 = call <vscale x 16 x i16> @llvm.vp.fptosi.nxv16i16.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f32_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.fptosi.nxv16i32.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f64_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.fptosi.nxv16i32.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f32_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.fptosi.nxv16i64.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f64_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.fptosi.nxv16i64.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f32_nxv16i1 = call <vscale x 16 x i1> @llvm.vp.fptosi.nxv16i1.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f64_nxv16i1 = call <vscale x 16 x i1> @llvm.vp.fptosi.nxv16i1.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f32_nxv32i8 = fptosi <vscale x 32 x float> undef to <vscale x 32 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f64_nxv32i8 = fptosi <vscale x 32 x double> undef to <vscale x 32 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f32_nxv32i16 = fptosi <vscale x 32 x float> undef to <vscale x 32 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f64_nxv32i16 = fptosi <vscale x 32 x double> undef to <vscale x 32 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f32_nxv32i32 = fptosi <vscale x 32 x float> undef to <vscale x 32 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f64_nxv32i32 = fptosi <vscale x 32 x double> undef to <vscale x 32 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f32_nxv32i64 = fptosi <vscale x 32 x float> undef to <vscale x 32 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f64_nxv32i64 = fptosi <vscale x 32 x double> undef to <vscale x 32 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f32_nxv32i1 = fptosi <vscale x 32 x float> undef to <vscale x 32 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f64_nxv32i1 = fptosi <vscale x 32 x double> undef to <vscale x 32 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f32_nxv32i8 = call <vscale x 32 x i8> @llvm.vp.fptosi.nxv32i8.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f64_nxv32i8 = call <vscale x 32 x i8> @llvm.vp.fptosi.nxv32i8.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f32_nxv32i16 = call <vscale x 32 x i16> @llvm.vp.fptosi.nxv32i16.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f64_nxv32i16 = call <vscale x 32 x i16> @llvm.vp.fptosi.nxv32i16.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f32_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.fptosi.nxv32i32.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f64_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.fptosi.nxv32i32.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f32_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.fptosi.nxv32i64.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f64_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.fptosi.nxv32i64.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f32_nxv32i1 = call <vscale x 32 x i1> @llvm.vp.fptosi.nxv32i1.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f64_nxv32i1 = call <vscale x 32 x i1> @llvm.vp.fptosi.nxv32i1.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f32_nxv64i8 = fptosi <vscale x 64 x float> undef to <vscale x 64 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f64_nxv64i8 = fptosi <vscale x 64 x double> undef to <vscale x 64 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f32_nxv64i16 = fptosi <vscale x 64 x float> undef to <vscale x 64 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f64_nxv64i16 = fptosi <vscale x 64 x double> undef to <vscale x 64 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f32_nxv64i32 = fptosi <vscale x 64 x float> undef to <vscale x 64 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f64_nxv64i32 = fptosi <vscale x 64 x double> undef to <vscale x 64 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f32_nxv64i64 = fptosi <vscale x 64 x float> undef to <vscale x 64 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f64_nxv64i64 = fptosi <vscale x 64 x double> undef to <vscale x 64 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f32_nxv64i1 = fptosi <vscale x 64 x float> undef to <vscale x 64 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f64_nxv64i1 = fptosi <vscale x 64 x double> undef to <vscale x 64 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f32_nxv64i8 = call <vscale x 64 x i8> @llvm.vp.fptosi.nxv64i8.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f64_nxv64i8 = call <vscale x 64 x i8> @llvm.vp.fptosi.nxv64i8.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f32_nxv64i16 = call <vscale x 64 x i16> @llvm.vp.fptosi.nxv64i16.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f64_nxv64i16 = call <vscale x 64 x i16> @llvm.vp.fptosi.nxv64i16.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f32_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.fptosi.nxv64i32.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f64_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.fptosi.nxv64i32.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f32_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.fptosi.nxv64i64.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f64_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.fptosi.nxv64i64.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f32_nxv64i1 = call <vscale x 64 x i1> @llvm.vp.fptosi.nxv64i1.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f64_nxv64i1 = call <vscale x 64 x i1> @llvm.vp.fptosi.nxv64i1.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  %v2f32_v2i8 = fptosi <2 x float> undef to <2 x i8>
  %v2f64_v2i8 = fptosi <2 x double> undef to <2 x i8>
  %v2f32_v2i16 = fptosi <2 x float> undef to <2 x i16>
  %v2f64_v2i16 = fptosi <2 x double> undef to <2 x i16>
  %v2f32_v2i32 = fptosi <2 x float> undef to <2 x i32>
  %v2f64_v2i32 = fptosi <2 x double> undef to <2 x i32>
  %v2f32_v2i64 = fptosi <2 x float> undef to <2 x i64>
  %v2f64_v2i64 = fptosi <2 x double> undef to <2 x i64>
  %v2f32_v2i1 = fptosi <2 x float> undef to <2 x i1>
  %v2f64_v2i1 = fptosi <2 x double> undef to <2 x i1>

  %vp_v2f32_v2i8 = call <2 x i8> @llvm.vp.fptosi.v2i8.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
  %vp_v2f64_v2i8 = call <2 x i8> @llvm.vp.fptosi.v2i8.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
  %vp_v2f32_v2i16 = call <2 x i16> @llvm.vp.fptosi.v2i16.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
  %vp_v2f64_v2i16 = call <2 x i16> @llvm.vp.fptosi.v2i16.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
  %vp_v2f32_v2i32 = call <2 x i32> @llvm.vp.fptosi.v2i32.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
  %vp_v2f64_v2i32 = call <2 x i32> @llvm.vp.fptosi.v2i32.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
  %vp_v2f32_v2i64 = call <2 x i64> @llvm.vp.fptosi.v2i64.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
  %vp_v2f64_v2i64 = call <2 x i64> @llvm.vp.fptosi.v2i64.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
  %vp_v2f32_v2i1 = call <2 x i1> @llvm.vp.fptosi.v2i1.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
  %vp_v2f64_v2i1 = call <2 x i1> @llvm.vp.fptosi.v2i1.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)

  %v4f32_v4i8 = fptosi <4 x float> undef to <4 x i8>
  %v4f64_v4i8 = fptosi <4 x double> undef to <4 x i8>
  %v4f32_v4i16 = fptosi <4 x float> undef to <4 x i16>
  %v4f64_v4i16 = fptosi <4 x double> undef to <4 x i16>
  %v4f32_v4i32 = fptosi <4 x float> undef to <4 x i32>
  %v4f64_v4i32 = fptosi <4 x double> undef to <4 x i32>
  %v4f32_v4i64 = fptosi <4 x float> undef to <4 x i64>
  %v4f64_v4i64 = fptosi <4 x double> undef to <4 x i64>
  %v4f32_v4i1 = fptosi <4 x float> undef to <4 x i1>
  %v4f64_v4i1 = fptosi <4 x double> undef to <4 x i1>

  %vp_v4f32_v4i8 = call <4 x i8> @llvm.vp.fptosi.v4i8.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
  %vp_v4f64_v4i8 = call <4 x i8> @llvm.vp.fptosi.v4i8.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
  %vp_v4f32_v4i16 = call <4 x i16> @llvm.vp.fptosi.v4i16.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
  %vp_v4f64_v4i16 = call <4 x i16> @llvm.vp.fptosi.v4i16.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
  %vp_v4f32_v4i32 = call <4 x i32> @llvm.vp.fptosi.v4i32.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
  %vp_v4f64_v4i32 = call <4 x i32> @llvm.vp.fptosi.v4i32.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
  %vp_v4f32_v4i64 = call <4 x i64> @llvm.vp.fptosi.v4i64.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
  %vp_v4f64_v4i64 = call <4 x i64> @llvm.vp.fptosi.v4i64.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
  %vp_v4f32_v4i1 = call <4 x i1> @llvm.vp.fptosi.v4i1.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
  %vp_v4f64_v4i1 = call <4 x i1> @llvm.vp.fptosi.v4i1.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)

  %v8f32_v8i8 = fptosi <8 x float> undef to <8 x i8>
  %v8f64_v8i8 = fptosi <8 x double> undef to <8 x i8>
  %v8f32_v8i16 = fptosi <8 x float> undef to <8 x i16>
  %v8f64_v8i16 = fptosi <8 x double> undef to <8 x i16>
  %v8f32_v8i32 = fptosi <8 x float> undef to <8 x i32>
  %v8f64_v8i32 = fptosi <8 x double> undef to <8 x i32>
  %v8f32_v8i64 = fptosi <8 x float> undef to <8 x i64>
  %v8f64_v8i64 = fptosi <8 x double> undef to <8 x i64>
  %v8f32_v8i1 = fptosi <8 x float> undef to <8 x i1>
  %v8f64_v8i1 = fptosi <8 x double> undef to <8 x i1>

  %vp_v8f32_v8i8 = call <8 x i8> @llvm.vp.fptosi.v8i8.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
  %vp_v8f64_v8i8 = call <8 x i8> @llvm.vp.fptosi.v8i8.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
  %vp_v8f32_v8i16 = call <8 x i16> @llvm.vp.fptosi.v8i16.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
  %vp_v8f64_v8i16 = call <8 x i16> @llvm.vp.fptosi.v8i16.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
  %vp_v8f32_v8i32 = call <8 x i32> @llvm.vp.fptosi.v8i32.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
  %vp_v8f64_v8i32 = call <8 x i32> @llvm.vp.fptosi.v8i32.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
  %vp_v8f32_v8i64 = call <8 x i64> @llvm.vp.fptosi.v8i64.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
  %vp_v8f64_v8i64 = call <8 x i64> @llvm.vp.fptosi.v8i64.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
  %vp_v8f32_v8i1 = call <8 x i1> @llvm.vp.fptosi.v8i1.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
  %vp_v8f64_v8i1 = call <8 x i1> @llvm.vp.fptosi.v8i1.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)

  %v16f32_v16i8 = fptosi <16 x float> undef to <16 x i8>
  %v16f64_v16i8 = fptosi <16 x double> undef to <16 x i8>
  %v16f32_v16i16 = fptosi <16 x float> undef to <16 x i16>
  %v16f64_v16i16 = fptosi <16 x double> undef to <16 x i16>
  %v16f32_v16i32 = fptosi <16 x float> undef to <16 x i32>
  %v16f64_v16i32 = fptosi <16 x double> undef to <16 x i32>
  %v16f32_v16i64 = fptosi <16 x float> undef to <16 x i64>
  %v16f64_v16i64 = fptosi <16 x double> undef to <16 x i64>
  %v16f32_v16i1 = fptosi <16 x float> undef to <16 x i1>
  %v16f64_v16i1 = fptosi <16 x double> undef to <16 x i1>

  %vp_v16f32_v16i8 = call <16 x i8> @llvm.vp.fptosi.v16i8.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
  %vp_v16f64_v16i8 = call <16 x i8> @llvm.vp.fptosi.v16i8.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
  %vp_v16f32_v16i16 = call <16 x i16> @llvm.vp.fptosi.v16i16.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
  %vp_v16f64_v16i16 = call <16 x i16> @llvm.vp.fptosi.v16i16.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
  %vp_v16f32_v16i32 = call <16 x i32> @llvm.vp.fptosi.v16i32.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
  %vp_v16f64_v16i32 = call <16 x i32> @llvm.vp.fptosi.v16i32.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
  %vp_v16f32_v16i64 = call <16 x i64> @llvm.vp.fptosi.v16i64.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
  %vp_v16f64_v16i64 = call <16 x i64> @llvm.vp.fptosi.v16i64.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
  %vp_v16f32_v16i1 = call <16 x i1> @llvm.vp.fptosi.v16i1.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
  %vp_v16f64_v16i1 = call <16 x i1> @llvm.vp.fptosi.v16i1.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)

  %v32f32_v32i8 = fptosi <32 x float> undef to <32 x i8>
  %v32f64_v32i8 = fptosi <32 x double> undef to <32 x i8>
  %v32f32_v32i16 = fptosi <32 x float> undef to <32 x i16>
  %v32f64_v32i16 = fptosi <32 x double> undef to <32 x i16>
  %v32f32_v32i32 = fptosi <32 x float> undef to <32 x i32>
  %v32f64_v32i32 = fptosi <32 x double> undef to <32 x i32>
  %v32f32_v32i64 = fptosi <32 x float> undef to <32 x i64>
  %v32f64_v32i64 = fptosi <32 x double> undef to <32 x i64>
  %v32f32_v32i1 = fptosi <32 x float> undef to <32 x i1>
  %v32f64_v32i1 = fptosi <32 x double> undef to <32 x i1>

  %vp_v32f32_v32i8 = call <32 x i8> @llvm.vp.fptosi.v32i8.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
  %vp_v32f64_v32i8 = call <32 x i8> @llvm.vp.fptosi.v32i8.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
  %vp_v32f32_v32i16 = call <32 x i16> @llvm.vp.fptosi.v32i16.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
  %vp_v32f64_v32i16 = call <32 x i16> @llvm.vp.fptosi.v32i16.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
  %vp_v32f32_v32i32 = call <32 x i32> @llvm.vp.fptosi.v32i32.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
  %vp_v32f64_v32i32 = call <32 x i32> @llvm.vp.fptosi.v32i32.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
  %vp_v32f32_v32i64 = call <32 x i64> @llvm.vp.fptosi.v32i64.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
  %vp_v32f64_v32i64 = call <32 x i64> @llvm.vp.fptosi.v32i64.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
  %vp_v32f32_v32i1 = call <32 x i1> @llvm.vp.fptosi.v32i1.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
  %vp_v32f64_v32i1 = call <32 x i1> @llvm.vp.fptosi.v32i1.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)

  %v64f32_v64i8 = fptosi <64 x float> undef to <64 x i8>
  %v64f64_v64i8 = fptosi <64 x double> undef to <64 x i8>
  %v64f32_v64i16 = fptosi <64 x float> undef to <64 x i16>
  %v64f64_v64i16 = fptosi <64 x double> undef to <64 x i16>
  %v64f32_v64i32 = fptosi <64 x float> undef to <64 x i32>
  %v64f64_v64i32 = fptosi <64 x double> undef to <64 x i32>
  %v64f32_v64i64 = fptosi <64 x float> undef to <64 x i64>
  %v64f64_v64i64 = fptosi <64 x double> undef to <64 x i64>
  %v64f32_v64i1 = fptosi <64 x float> undef to <64 x i1>
  %v64f64_v64i1 = fptosi <64 x double> undef to <64 x i1>

  %vp_v64f32_v64i8 = call <64 x i8> @llvm.vp.fptosi.v64i8.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
  %vp_v64f64_v64i8 = call <64 x i8> @llvm.vp.fptosi.v64i8.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
  %vp_v64f32_v64i16 = call <64 x i16> @llvm.vp.fptosi.v64i16.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
  %vp_v64f64_v64i16 = call <64 x i16> @llvm.vp.fptosi.v64i16.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
  %vp_v64f32_v64i32 = call <64 x i32> @llvm.vp.fptosi.v64i32.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
  %vp_v64f64_v64i32 = call <64 x i32> @llvm.vp.fptosi.v64i32.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
  %vp_v64f32_v64i64 = call <64 x i64> @llvm.vp.fptosi.v64i64.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
  %vp_v64f64_v64i64 = call <64 x i64> @llvm.vp.fptosi.v64i64.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
  %vp_v64f32_v64i1 = call <64 x i1> @llvm.vp.fptosi.v64i1.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
  %vp_v64f64_v64i1 = call <64 x i1> @llvm.vp.fptosi.v64i1.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)

  %v128f32_v128i8 = fptosi <128 x float> undef to <128 x i8>
  %v128f64_v128i8 = fptosi <128 x double> undef to <128 x i8>
  %v128f32_v128i16 = fptosi <128 x float> undef to <128 x i16>
  %v128f64_v128i16 = fptosi <128 x double> undef to <128 x i16>
  %v128f32_v128i32 = fptosi <128 x float> undef to <128 x i32>
  %v128f64_v128i32 = fptosi <128 x double> undef to <128 x i32>
  %v128f32_v128i64 = fptosi <128 x float> undef to <128 x i64>
  %v128f64_v128i64 = fptosi <128 x double> undef to <128 x i64>
  %v128f32_v128i1 = fptosi <128 x float> undef to <128 x i1>
  %v128f64_v128i1 = fptosi <128 x double> undef to <128 x i1>

  %vp_v128f32_v128i8 = call <128 x i8> @llvm.vp.fptosi.v128i8.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
  %vp_v128f64_v128i8 = call <128 x i8> @llvm.vp.fptosi.v128i8.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
  %vp_v128f32_v128i16 = call <128 x i16> @llvm.vp.fptosi.v128i16.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
  %vp_v128f64_v128i16 = call <128 x i16> @llvm.vp.fptosi.v128i16.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
  %vp_v128f32_v128i32 = call <128 x i32> @llvm.vp.fptosi.v128i32.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
  %vp_v128f64_v128i32 = call <128 x i32> @llvm.vp.fptosi.v128i32.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
  %vp_v128f32_v128i64 = call <128 x i64> @llvm.vp.fptosi.v128i64.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
  %vp_v128f64_v128i64 = call <128 x i64> @llvm.vp.fptosi.v128i64.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
  %vp_v128f32_v128i1 = call <128 x i1> @llvm.vp.fptosi.v128i1.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
  %vp_v128f64_v128i1 = call <128 x i1> @llvm.vp.fptosi.v128i1.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)

  %nxv1f32_nxv1i8 = fptosi <vscale x 1 x float> undef to <vscale x 1 x i8>
  %nxv1f64_nxv1i8 = fptosi <vscale x 1 x double> undef to <vscale x 1 x i8>
  %nxv1f32_nxv1i16 = fptosi <vscale x 1 x float> undef to <vscale x 1 x i16>
  %nxv1f64_nxv1i16 = fptosi <vscale x 1 x double> undef to <vscale x 1 x i16>
  %nxv1f32_nxv1i32 = fptosi <vscale x 1 x float> undef to <vscale x 1 x i32>
  %nxv1f64_nxv1i32 = fptosi <vscale x 1 x double> undef to <vscale x 1 x i32>
  %nxv1f32_nxv1i64 = fptosi <vscale x 1 x float> undef to <vscale x 1 x i64>
  %nxv1f64_nxv1i64 = fptosi <vscale x 1 x double> undef to <vscale x 1 x i64>
  %nxv1f32_nxv1i1 = fptosi <vscale x 1 x float> undef to <vscale x 1 x i1>
  %nxv1f64_nxv1i1 = fptosi <vscale x 1 x double> undef to <vscale x 1 x i1>

  %vp_nxv1f32_nxv1i8 = call <vscale x 1 x i8> @llvm.vp.fptosi.nxv1i8.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1f64_nxv1i8 = call <vscale x 1 x i8> @llvm.vp.fptosi.nxv1i8.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1f32_nxv1i16 = call <vscale x 1 x i16> @llvm.vp.fptosi.nxv1i16.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1f64_nxv1i16 = call <vscale x 1 x i16> @llvm.vp.fptosi.nxv1i16.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1f32_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.fptosi.nxv1i32.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1f64_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.fptosi.nxv1i32.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1f32_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.fptosi.nxv1i64.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1f64_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.fptosi.nxv1i64.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1f32_nxv1i1 = call <vscale x 1 x i1> @llvm.vp.fptosi.nxv1i1.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1f64_nxv1i1 = call <vscale x 1 x i1> @llvm.vp.fptosi.nxv1i1.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)

  %nxv2f32_nxv2i8 = fptosi <vscale x 2 x float> undef to <vscale x 2 x i8>
  %nxv2f64_nxv2i8 = fptosi <vscale x 2 x double> undef to <vscale x 2 x i8>
  %nxv2f32_nxv2i16 = fptosi <vscale x 2 x float> undef to <vscale x 2 x i16>
  %nxv2f64_nxv2i16 = fptosi <vscale x 2 x double> undef to <vscale x 2 x i16>
  %nxv2f32_nxv2i32 = fptosi <vscale x 2 x float> undef to <vscale x 2 x i32>
  %nxv2f64_nxv2i32 = fptosi <vscale x 2 x double> undef to <vscale x 2 x i32>
  %nxv2f32_nxv2i64 = fptosi <vscale x 2 x float> undef to <vscale x 2 x i64>
  %nxv2f64_nxv2i64 = fptosi <vscale x 2 x double> undef to <vscale x 2 x i64>
  %nxv2f32_nxv2i1 = fptosi <vscale x 2 x float> undef to <vscale x 2 x i1>
  %nxv2f64_nxv2i1 = fptosi <vscale x 2 x double> undef to <vscale x 2 x i1>

  %vp_nxv2f32_nxv2i8 = call <vscale x 2 x i8> @llvm.vp.fptosi.nxv2i8.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2f64_nxv2i8 = call <vscale x 2 x i8> @llvm.vp.fptosi.nxv2i8.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2f32_nxv2i16 = call <vscale x 2 x i16> @llvm.vp.fptosi.nxv2i16.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2f64_nxv2i16 = call <vscale x 2 x i16> @llvm.vp.fptosi.nxv2i16.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2f32_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.fptosi.nxv2i32.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2f64_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.fptosi.nxv2i32.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2f32_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.fptosi.nxv2i64.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2f64_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.fptosi.nxv2i64.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2f32_nxv2i1 = call <vscale x 2 x i1> @llvm.vp.fptosi.nxv2i1.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2f64_nxv2i1 = call <vscale x 2 x i1> @llvm.vp.fptosi.nxv2i1.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)

  %nxv4f32_nxv4i8 = fptosi <vscale x 4 x float> undef to <vscale x 4 x i8>
  %nxv4f64_nxv4i8 = fptosi <vscale x 4 x double> undef to <vscale x 4 x i8>
  %nxv4f32_nxv4i16 = fptosi <vscale x 4 x float> undef to <vscale x 4 x i16>
  %nxv4f64_nxv4i16 = fptosi <vscale x 4 x double> undef to <vscale x 4 x i16>
  %nxv4f32_nxv4i32 = fptosi <vscale x 4 x float> undef to <vscale x 4 x i32>
  %nxv4f64_nxv4i32 = fptosi <vscale x 4 x double> undef to <vscale x 4 x i32>
  %nxv4f32_nxv4i64 = fptosi <vscale x 4 x float> undef to <vscale x 4 x i64>
  %nxv4f64_nxv4i64 = fptosi <vscale x 4 x double> undef to <vscale x 4 x i64>
  %nxv4f32_nxv4i1 = fptosi <vscale x 4 x float> undef to <vscale x 4 x i1>
  %nxv4f64_nxv4i1 = fptosi <vscale x 4 x double> undef to <vscale x 4 x i1>

  %vp_nxv4f32_nxv4i8 = call <vscale x 4 x i8> @llvm.vp.fptosi.nxv4i8.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4f64_nxv4i8 = call <vscale x 4 x i8> @llvm.vp.fptosi.nxv4i8.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4f32_nxv4i16 = call <vscale x 4 x i16> @llvm.vp.fptosi.nxv4i16.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4f64_nxv4i16 = call <vscale x 4 x i16> @llvm.vp.fptosi.nxv4i16.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4f32_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.fptosi.nxv4i32.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4f64_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.fptosi.nxv4i32.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4f32_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.fptosi.nxv4i64.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4f64_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.fptosi.nxv4i64.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4f32_nxv4i1 = call <vscale x 4 x i1> @llvm.vp.fptosi.nxv4i1.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4f64_nxv4i1 = call <vscale x 4 x i1> @llvm.vp.fptosi.nxv4i1.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)

  %nxv8f32_nxv8i8 = fptosi <vscale x 8 x float> undef to <vscale x 8 x i8>
  %nxv8f64_nxv8i8 = fptosi <vscale x 8 x double> undef to <vscale x 8 x i8>
  %nxv8f32_nxv8i16 = fptosi <vscale x 8 x float> undef to <vscale x 8 x i16>
  %nxv8f64_nxv8i16 = fptosi <vscale x 8 x double> undef to <vscale x 8 x i16>
  %nxv8f32_nxv8i32 = fptosi <vscale x 8 x float> undef to <vscale x 8 x i32>
  %nxv8f64_nxv8i32 = fptosi <vscale x 8 x double> undef to <vscale x 8 x i32>
  %nxv8f32_nxv8i64 = fptosi <vscale x 8 x float> undef to <vscale x 8 x i64>
  %nxv8f64_nxv8i64 = fptosi <vscale x 8 x double> undef to <vscale x 8 x i64>
  %nxv8f32_nxv8i1 = fptosi <vscale x 8 x float> undef to <vscale x 8 x i1>
  %nxv8f64_nxv8i1 = fptosi <vscale x 8 x double> undef to <vscale x 8 x i1>

  %vp_nxv8f32_nxv8i8 = call <vscale x 8 x i8> @llvm.vp.fptosi.nxv8i8.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8f64_nxv8i8 = call <vscale x 8 x i8> @llvm.vp.fptosi.nxv8i8.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8f32_nxv8i16 = call <vscale x 8 x i16> @llvm.vp.fptosi.nxv8i16.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8f64_nxv8i16 = call <vscale x 8 x i16> @llvm.vp.fptosi.nxv8i16.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8f32_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.fptosi.nxv8i32.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8f64_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.fptosi.nxv8i32.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8f32_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.fptosi.nxv8i64.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8f64_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.fptosi.nxv8i64.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8f32_nxv8i1 = call <vscale x 8 x i1> @llvm.vp.fptosi.nxv8i1.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8f64_nxv8i1 = call <vscale x 8 x i1> @llvm.vp.fptosi.nxv8i1.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)

  %nxv16f32_nxv16i8 = fptosi <vscale x 16 x float> undef to <vscale x 16 x i8>
  %nxv16f64_nxv16i8 = fptosi <vscale x 16 x double> undef to <vscale x 16 x i8>
  %nxv16f32_nxv16i16 = fptosi <vscale x 16 x float> undef to <vscale x 16 x i16>
  %nxv16f64_nxv16i16 = fptosi <vscale x 16 x double> undef to <vscale x 16 x i16>
  %nxv16f32_nxv16i32 = fptosi <vscale x 16 x float> undef to <vscale x 16 x i32>
  %nxv16f64_nxv16i32 = fptosi <vscale x 16 x double> undef to <vscale x 16 x i32>
  %nxv16f32_nxv16i64 = fptosi <vscale x 16 x float> undef to <vscale x 16 x i64>
  %nxv16f64_nxv16i64 = fptosi <vscale x 16 x double> undef to <vscale x 16 x i64>
  %nxv16f32_nxv16i1 = fptosi <vscale x 16 x float> undef to <vscale x 16 x i1>
  %nxv16f64_nxv16i1 = fptosi <vscale x 16 x double> undef to <vscale x 16 x i1>

  %vp_nxv16f32_nxv16i8 = call <vscale x 16 x i8> @llvm.vp.fptosi.nxv16i8.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16f64_nxv16i8 = call <vscale x 16 x i8> @llvm.vp.fptosi.nxv16i8.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16f32_nxv16i16 = call <vscale x 16 x i16> @llvm.vp.fptosi.nxv16i16.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16f64_nxv16i16 = call <vscale x 16 x i16> @llvm.vp.fptosi.nxv16i16.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16f32_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.fptosi.nxv16i32.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16f64_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.fptosi.nxv16i32.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16f32_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.fptosi.nxv16i64.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16f64_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.fptosi.nxv16i64.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16f32_nxv16i1 = call <vscale x 16 x i1> @llvm.vp.fptosi.nxv16i1.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16f64_nxv16i1 = call <vscale x 16 x i1> @llvm.vp.fptosi.nxv16i1.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)

  %nxv32f32_nxv32i8 = fptosi <vscale x 32 x float> undef to <vscale x 32 x i8>
  %nxv32f64_nxv32i8 = fptosi <vscale x 32 x double> undef to <vscale x 32 x i8>
  %nxv32f32_nxv32i16 = fptosi <vscale x 32 x float> undef to <vscale x 32 x i16>
  %nxv32f64_nxv32i16 = fptosi <vscale x 32 x double> undef to <vscale x 32 x i16>
  %nxv32f32_nxv32i32 = fptosi <vscale x 32 x float> undef to <vscale x 32 x i32>
  %nxv32f64_nxv32i32 = fptosi <vscale x 32 x double> undef to <vscale x 32 x i32>
  %nxv32f32_nxv32i64 = fptosi <vscale x 32 x float> undef to <vscale x 32 x i64>
  %nxv32f64_nxv32i64 = fptosi <vscale x 32 x double> undef to <vscale x 32 x i64>
  %nxv32f32_nxv32i1 = fptosi <vscale x 32 x float> undef to <vscale x 32 x i1>
  %nxv32f64_nxv32i1 = fptosi <vscale x 32 x double> undef to <vscale x 32 x i1>

  %vp_nxv32f32_nxv32i8 = call <vscale x 32 x i8> @llvm.vp.fptosi.nxv32i8.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32f64_nxv32i8 = call <vscale x 32 x i8> @llvm.vp.fptosi.nxv32i8.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32f32_nxv32i16 = call <vscale x 32 x i16> @llvm.vp.fptosi.nxv32i16.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32f64_nxv32i16 = call <vscale x 32 x i16> @llvm.vp.fptosi.nxv32i16.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32f32_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.fptosi.nxv32i32.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32f64_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.fptosi.nxv32i32.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32f32_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.fptosi.nxv32i64.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32f64_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.fptosi.nxv32i64.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32f32_nxv32i1 = call <vscale x 32 x i1> @llvm.vp.fptosi.nxv32i1.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32f64_nxv32i1 = call <vscale x 32 x i1> @llvm.vp.fptosi.nxv32i1.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)

  %nxv64f32_nxv64i8 = fptosi <vscale x 64 x float> undef to <vscale x 64 x i8>
  %nxv64f64_nxv64i8 = fptosi <vscale x 64 x double> undef to <vscale x 64 x i8>
  %nxv64f32_nxv64i16 = fptosi <vscale x 64 x float> undef to <vscale x 64 x i16>
  %nxv64f64_nxv64i16 = fptosi <vscale x 64 x double> undef to <vscale x 64 x i16>
  %nxv64f32_nxv64i32 = fptosi <vscale x 64 x float> undef to <vscale x 64 x i32>
  %nxv64f64_nxv64i32 = fptosi <vscale x 64 x double> undef to <vscale x 64 x i32>
  %nxv64f32_nxv64i64 = fptosi <vscale x 64 x float> undef to <vscale x 64 x i64>
  %nxv64f64_nxv64i64 = fptosi <vscale x 64 x double> undef to <vscale x 64 x i64>
  %nxv64f32_nxv64i1 = fptosi <vscale x 64 x float> undef to <vscale x 64 x i1>
  %nxv64f64_nxv64i1 = fptosi <vscale x 64 x double> undef to <vscale x 64 x i1>

  %vp_nxv64f32_nxv64i8 = call <vscale x 64 x i8> @llvm.vp.fptosi.nxv64i8.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64f64_nxv64i8 = call <vscale x 64 x i8> @llvm.vp.fptosi.nxv64i8.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64f32_nxv64i16 = call <vscale x 64 x i16> @llvm.vp.fptosi.nxv64i16.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64f64_nxv64i16 = call <vscale x 64 x i16> @llvm.vp.fptosi.nxv64i16.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64f32_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.fptosi.nxv64i32.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64f64_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.fptosi.nxv64i32.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64f32_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.fptosi.nxv64i64.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64f64_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.fptosi.nxv64i64.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64f32_nxv64i1 = call <vscale x 64 x i1> @llvm.vp.fptosi.nxv64i1.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64f64_nxv64i1 = call <vscale x 64 x i1> @llvm.vp.fptosi.nxv64i1.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)

  ret void
}

define void @fptoui() {
; ZVE64D-LABEL: 'fptoui'
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2f32_v2i8 = fptoui <2 x float> undef to <2 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2f64_v2i8 = fptoui <2 x double> undef to <2 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2f32_v2i16 = fptoui <2 x float> undef to <2 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2f64_v2i16 = fptoui <2 x double> undef to <2 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2f32_v2i32 = fptoui <2 x float> undef to <2 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2f64_v2i32 = fptoui <2 x double> undef to <2 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2f32_v2i64 = fptoui <2 x float> undef to <2 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2f64_v2i64 = fptoui <2 x double> undef to <2 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2f32_v2i1 = fptoui <2 x float> undef to <2 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2f64_v2i1 = fptoui <2 x double> undef to <2 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2f32_v2i8 = call <2 x i8> @llvm.vp.fptoui.v2i8.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2f64_v2i8 = call <2 x i8> @llvm.vp.fptoui.v2i8.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2f32_v2i16 = call <2 x i16> @llvm.vp.fptoui.v2i16.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2f64_v2i16 = call <2 x i16> @llvm.vp.fptoui.v2i16.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2f32_v2i32 = call <2 x i32> @llvm.vp.fptoui.v2i32.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2f64_v2i32 = call <2 x i32> @llvm.vp.fptoui.v2i32.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2f32_v2i64 = call <2 x i64> @llvm.vp.fptoui.v2i64.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2f64_v2i64 = call <2 x i64> @llvm.vp.fptoui.v2i64.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2f32_v2i1 = call <2 x i1> @llvm.vp.fptoui.v2i1.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2f64_v2i1 = call <2 x i1> @llvm.vp.fptoui.v2i1.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4f32_v4i8 = fptoui <4 x float> undef to <4 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v4f64_v4i8 = fptoui <4 x double> undef to <4 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4f32_v4i16 = fptoui <4 x float> undef to <4 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4f64_v4i16 = fptoui <4 x double> undef to <4 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4f32_v4i32 = fptoui <4 x float> undef to <4 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4f64_v4i32 = fptoui <4 x double> undef to <4 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4f32_v4i64 = fptoui <4 x float> undef to <4 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4f64_v4i64 = fptoui <4 x double> undef to <4 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v4f32_v4i1 = fptoui <4 x float> undef to <4 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v4f64_v4i1 = fptoui <4 x double> undef to <4 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4f32_v4i8 = call <4 x i8> @llvm.vp.fptoui.v4i8.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v4f64_v4i8 = call <4 x i8> @llvm.vp.fptoui.v4i8.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v4f32_v4i16 = call <4 x i16> @llvm.vp.fptoui.v4i16.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4f64_v4i16 = call <4 x i16> @llvm.vp.fptoui.v4i16.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v4f32_v4i32 = call <4 x i32> @llvm.vp.fptoui.v4i32.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v4f64_v4i32 = call <4 x i32> @llvm.vp.fptoui.v4i32.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4f32_v4i64 = call <4 x i64> @llvm.vp.fptoui.v4i64.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4f64_v4i64 = call <4 x i64> @llvm.vp.fptoui.v4i64.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v4f32_v4i1 = call <4 x i1> @llvm.vp.fptoui.v4i1.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v4f64_v4i1 = call <4 x i1> @llvm.vp.fptoui.v4i1.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8f32_v8i8 = fptoui <8 x float> undef to <8 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8f64_v8i8 = fptoui <8 x double> undef to <8 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8f32_v8i16 = fptoui <8 x float> undef to <8 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v8f64_v8i16 = fptoui <8 x double> undef to <8 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8f32_v8i32 = fptoui <8 x float> undef to <8 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8f64_v8i32 = fptoui <8 x double> undef to <8 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8f32_v8i64 = fptoui <8 x float> undef to <8 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8f64_v8i64 = fptoui <8 x double> undef to <8 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v8f32_v8i1 = fptoui <8 x float> undef to <8 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %v8f64_v8i1 = fptoui <8 x double> undef to <8 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8f32_v8i8 = call <8 x i8> @llvm.vp.fptoui.v8i8.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v8f64_v8i8 = call <8 x i8> @llvm.vp.fptoui.v8i8.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v8f32_v8i16 = call <8 x i16> @llvm.vp.fptoui.v8i16.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v8f64_v8i16 = call <8 x i16> @llvm.vp.fptoui.v8i16.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8f32_v8i32 = call <8 x i32> @llvm.vp.fptoui.v8i32.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8f64_v8i32 = call <8 x i32> @llvm.vp.fptoui.v8i32.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v8f32_v8i64 = call <8 x i64> @llvm.vp.fptoui.v8i64.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v8f64_v8i64 = call <8 x i64> @llvm.vp.fptoui.v8i64.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v8f32_v8i1 = call <8 x i1> @llvm.vp.fptoui.v8i1.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_v8f64_v8i1 = call <8 x i1> @llvm.vp.fptoui.v8i1.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v16f32_v16i8 = fptoui <16 x float> undef to <16 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v16f64_v16i8 = fptoui <16 x double> undef to <16 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v16f32_v16i16 = fptoui <16 x float> undef to <16 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %v16f64_v16i16 = fptoui <16 x double> undef to <16 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v16f32_v16i32 = fptoui <16 x float> undef to <16 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v16f64_v16i32 = fptoui <16 x double> undef to <16 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v16f32_v16i64 = fptoui <16 x float> undef to <16 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v16f64_v16i64 = fptoui <16 x double> undef to <16 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %v16f32_v16i1 = fptoui <16 x float> undef to <16 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %v16f64_v16i1 = fptoui <16 x double> undef to <16 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v16f32_v16i8 = call <16 x i8> @llvm.vp.fptoui.v16i8.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v16f64_v16i8 = call <16 x i8> @llvm.vp.fptoui.v16i8.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v16f32_v16i16 = call <16 x i16> @llvm.vp.fptoui.v16i16.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_v16f64_v16i16 = call <16 x i16> @llvm.vp.fptoui.v16i16.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v16f32_v16i32 = call <16 x i32> @llvm.vp.fptoui.v16i32.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v16f64_v16i32 = call <16 x i32> @llvm.vp.fptoui.v16i32.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v16f32_v16i64 = call <16 x i64> @llvm.vp.fptoui.v16i64.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v16f64_v16i64 = call <16 x i64> @llvm.vp.fptoui.v16i64.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_v16f32_v16i1 = call <16 x i1> @llvm.vp.fptoui.v16i1.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %vp_v16f64_v16i1 = call <16 x i1> @llvm.vp.fptoui.v16i1.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %v32f32_v32i8 = fptoui <32 x float> undef to <32 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v32f64_v32i8 = fptoui <32 x double> undef to <32 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v32f32_v32i16 = fptoui <32 x float> undef to <32 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %v32f64_v32i16 = fptoui <32 x double> undef to <32 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v32f32_v32i32 = fptoui <32 x float> undef to <32 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %v32f64_v32i32 = fptoui <32 x double> undef to <32 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %v32f32_v32i64 = fptoui <32 x float> undef to <32 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v32f64_v32i64 = fptoui <32 x double> undef to <32 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %v32f32_v32i1 = fptoui <32 x float> undef to <32 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %v32f64_v32i1 = fptoui <32 x double> undef to <32 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_v32f32_v32i8 = call <32 x i8> @llvm.vp.fptoui.v32i8.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v32f64_v32i8 = call <32 x i8> @llvm.vp.fptoui.v32i8.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v32f32_v32i16 = call <32 x i16> @llvm.vp.fptoui.v32i16.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %vp_v32f64_v32i16 = call <32 x i16> @llvm.vp.fptoui.v32i16.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v32f32_v32i32 = call <32 x i32> @llvm.vp.fptoui.v32i32.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %vp_v32f64_v32i32 = call <32 x i32> @llvm.vp.fptoui.v32i32.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_v32f32_v32i64 = call <32 x i64> @llvm.vp.fptoui.v32i64.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v32f64_v32i64 = call <32 x i64> @llvm.vp.fptoui.v32i64.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %vp_v32f32_v32i1 = call <32 x i1> @llvm.vp.fptoui.v32i1.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_v32f64_v32i1 = call <32 x i1> @llvm.vp.fptoui.v32i1.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %v64f32_v64i8 = fptoui <64 x float> undef to <64 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v64f64_v64i8 = fptoui <64 x double> undef to <64 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %v64f32_v64i16 = fptoui <64 x float> undef to <64 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %v64f64_v64i16 = fptoui <64 x double> undef to <64 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v64f32_v64i32 = fptoui <64 x float> undef to <64 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %v64f64_v64i32 = fptoui <64 x double> undef to <64 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %v64f32_v64i64 = fptoui <64 x float> undef to <64 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v64f64_v64i64 = fptoui <64 x double> undef to <64 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %v64f32_v64i1 = fptoui <64 x float> undef to <64 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %v64f64_v64i1 = fptoui <64 x double> undef to <64 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %vp_v64f32_v64i8 = call <64 x i8> @llvm.vp.fptoui.v64i8.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v64f64_v64i8 = call <64 x i8> @llvm.vp.fptoui.v64i8.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %vp_v64f32_v64i16 = call <64 x i16> @llvm.vp.fptoui.v64i16.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %vp_v64f64_v64i16 = call <64 x i16> @llvm.vp.fptoui.v64i16.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v64f32_v64i32 = call <64 x i32> @llvm.vp.fptoui.v64i32.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %vp_v64f64_v64i32 = call <64 x i32> @llvm.vp.fptoui.v64i32.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_v64f32_v64i64 = call <64 x i64> @llvm.vp.fptoui.v64i64.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v64f64_v64i64 = call <64 x i64> @llvm.vp.fptoui.v64i64.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_v64f32_v64i1 = call <64 x i1> @llvm.vp.fptoui.v64i1.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %vp_v64f64_v64i1 = call <64 x i1> @llvm.vp.fptoui.v64i1.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %v128f32_v128i8 = fptoui <128 x float> undef to <128 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v128f64_v128i8 = fptoui <128 x double> undef to <128 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %v128f32_v128i16 = fptoui <128 x float> undef to <128 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %v128f64_v128i16 = fptoui <128 x double> undef to <128 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v128f32_v128i32 = fptoui <128 x float> undef to <128 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %v128f64_v128i32 = fptoui <128 x double> undef to <128 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %v128f32_v128i64 = fptoui <128 x float> undef to <128 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v128f64_v128i64 = fptoui <128 x double> undef to <128 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %v128f32_v128i1 = fptoui <128 x float> undef to <128 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 103 for instruction: %v128f64_v128i1 = fptoui <128 x double> undef to <128 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %vp_v128f32_v128i8 = call <128 x i8> @llvm.vp.fptoui.v128i8.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v128f64_v128i8 = call <128 x i8> @llvm.vp.fptoui.v128i8.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %vp_v128f32_v128i16 = call <128 x i16> @llvm.vp.fptoui.v128i16.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %vp_v128f64_v128i16 = call <128 x i16> @llvm.vp.fptoui.v128i16.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v128f32_v128i32 = call <128 x i32> @llvm.vp.fptoui.v128i32.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %vp_v128f64_v128i32 = call <128 x i32> @llvm.vp.fptoui.v128i32.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %vp_v128f32_v128i64 = call <128 x i64> @llvm.vp.fptoui.v128i64.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v128f64_v128i64 = call <128 x i64> @llvm.vp.fptoui.v128i64.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %vp_v128f32_v128i1 = call <128 x i1> @llvm.vp.fptoui.v128i1.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 103 for instruction: %vp_v128f64_v128i1 = call <128 x i1> @llvm.vp.fptoui.v128i1.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1f32_nxv1i8 = fptoui <vscale x 1 x float> undef to <vscale x 1 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv1f64_nxv1i8 = fptoui <vscale x 1 x double> undef to <vscale x 1 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1f32_nxv1i16 = fptoui <vscale x 1 x float> undef to <vscale x 1 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1f64_nxv1i16 = fptoui <vscale x 1 x double> undef to <vscale x 1 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1f32_nxv1i32 = fptoui <vscale x 1 x float> undef to <vscale x 1 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1f64_nxv1i32 = fptoui <vscale x 1 x double> undef to <vscale x 1 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1f32_nxv1i64 = fptoui <vscale x 1 x float> undef to <vscale x 1 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1f64_nxv1i64 = fptoui <vscale x 1 x double> undef to <vscale x 1 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv1f32_nxv1i1 = fptoui <vscale x 1 x float> undef to <vscale x 1 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv1f64_nxv1i1 = fptoui <vscale x 1 x double> undef to <vscale x 1 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1f32_nxv1i8 = call <vscale x 1 x i8> @llvm.vp.fptoui.nxv1i8.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv1f64_nxv1i8 = call <vscale x 1 x i8> @llvm.vp.fptoui.nxv1i8.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1f32_nxv1i16 = call <vscale x 1 x i16> @llvm.vp.fptoui.nxv1i16.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1f64_nxv1i16 = call <vscale x 1 x i16> @llvm.vp.fptoui.nxv1i16.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1f32_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.fptoui.nxv1i32.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1f64_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.fptoui.nxv1i32.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1f32_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.fptoui.nxv1i64.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1f64_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.fptoui.nxv1i64.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv1f32_nxv1i1 = call <vscale x 1 x i1> @llvm.vp.fptoui.nxv1i1.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv1f64_nxv1i1 = call <vscale x 1 x i1> @llvm.vp.fptoui.nxv1i1.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2f32_nxv2i8 = fptoui <vscale x 2 x float> undef to <vscale x 2 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv2f64_nxv2i8 = fptoui <vscale x 2 x double> undef to <vscale x 2 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv2f32_nxv2i16 = fptoui <vscale x 2 x float> undef to <vscale x 2 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2f64_nxv2i16 = fptoui <vscale x 2 x double> undef to <vscale x 2 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv2f32_nxv2i32 = fptoui <vscale x 2 x float> undef to <vscale x 2 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv2f64_nxv2i32 = fptoui <vscale x 2 x double> undef to <vscale x 2 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2f32_nxv2i64 = fptoui <vscale x 2 x float> undef to <vscale x 2 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2f64_nxv2i64 = fptoui <vscale x 2 x double> undef to <vscale x 2 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv2f32_nxv2i1 = fptoui <vscale x 2 x float> undef to <vscale x 2 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv2f64_nxv2i1 = fptoui <vscale x 2 x double> undef to <vscale x 2 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2f32_nxv2i8 = call <vscale x 2 x i8> @llvm.vp.fptoui.nxv2i8.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv2f64_nxv2i8 = call <vscale x 2 x i8> @llvm.vp.fptoui.nxv2i8.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv2f32_nxv2i16 = call <vscale x 2 x i16> @llvm.vp.fptoui.nxv2i16.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2f64_nxv2i16 = call <vscale x 2 x i16> @llvm.vp.fptoui.nxv2i16.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv2f32_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.fptoui.nxv2i32.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv2f64_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.fptoui.nxv2i32.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2f32_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.fptoui.nxv2i64.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2f64_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.fptoui.nxv2i64.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv2f32_nxv2i1 = call <vscale x 2 x i1> @llvm.vp.fptoui.nxv2i1.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv2f64_nxv2i1 = call <vscale x 2 x i1> @llvm.vp.fptoui.nxv2i1.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4f32_nxv4i8 = fptoui <vscale x 4 x float> undef to <vscale x 4 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv4f64_nxv4i8 = fptoui <vscale x 4 x double> undef to <vscale x 4 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv4f32_nxv4i16 = fptoui <vscale x 4 x float> undef to <vscale x 4 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv4f64_nxv4i16 = fptoui <vscale x 4 x double> undef to <vscale x 4 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4f32_nxv4i32 = fptoui <vscale x 4 x float> undef to <vscale x 4 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4f64_nxv4i32 = fptoui <vscale x 4 x double> undef to <vscale x 4 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv4f32_nxv4i64 = fptoui <vscale x 4 x float> undef to <vscale x 4 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv4f64_nxv4i64 = fptoui <vscale x 4 x double> undef to <vscale x 4 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv4f32_nxv4i1 = fptoui <vscale x 4 x float> undef to <vscale x 4 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %nxv4f64_nxv4i1 = fptoui <vscale x 4 x double> undef to <vscale x 4 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4f32_nxv4i8 = call <vscale x 4 x i8> @llvm.vp.fptoui.nxv4i8.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv4f64_nxv4i8 = call <vscale x 4 x i8> @llvm.vp.fptoui.nxv4i8.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv4f32_nxv4i16 = call <vscale x 4 x i16> @llvm.vp.fptoui.nxv4i16.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv4f64_nxv4i16 = call <vscale x 4 x i16> @llvm.vp.fptoui.nxv4i16.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4f32_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.fptoui.nxv4i32.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4f64_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.fptoui.nxv4i32.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv4f32_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.fptoui.nxv4i64.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv4f64_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.fptoui.nxv4i64.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv4f32_nxv4i1 = call <vscale x 4 x i1> @llvm.vp.fptoui.nxv4i1.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_nxv4f64_nxv4i1 = call <vscale x 4 x i1> @llvm.vp.fptoui.nxv4i1.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv8f32_nxv8i8 = fptoui <vscale x 8 x float> undef to <vscale x 8 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %nxv8f64_nxv8i8 = fptoui <vscale x 8 x double> undef to <vscale x 8 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv8f32_nxv8i16 = fptoui <vscale x 8 x float> undef to <vscale x 8 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %nxv8f64_nxv8i16 = fptoui <vscale x 8 x double> undef to <vscale x 8 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv8f32_nxv8i32 = fptoui <vscale x 8 x float> undef to <vscale x 8 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv8f64_nxv8i32 = fptoui <vscale x 8 x double> undef to <vscale x 8 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv8f32_nxv8i64 = fptoui <vscale x 8 x float> undef to <vscale x 8 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv8f64_nxv8i64 = fptoui <vscale x 8 x double> undef to <vscale x 8 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %nxv8f32_nxv8i1 = fptoui <vscale x 8 x float> undef to <vscale x 8 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %nxv8f64_nxv8i1 = fptoui <vscale x 8 x double> undef to <vscale x 8 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv8f32_nxv8i8 = call <vscale x 8 x i8> @llvm.vp.fptoui.nxv8i8.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_nxv8f64_nxv8i8 = call <vscale x 8 x i8> @llvm.vp.fptoui.nxv8i8.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv8f32_nxv8i16 = call <vscale x 8 x i16> @llvm.vp.fptoui.nxv8i16.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_nxv8f64_nxv8i16 = call <vscale x 8 x i16> @llvm.vp.fptoui.nxv8i16.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv8f32_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.fptoui.nxv8i32.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv8f64_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.fptoui.nxv8i32.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv8f32_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.fptoui.nxv8i64.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv8f64_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.fptoui.nxv8i64.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_nxv8f32_nxv8i1 = call <vscale x 8 x i1> @llvm.vp.fptoui.nxv8i1.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %vp_nxv8f64_nxv8i1 = call <vscale x 8 x i1> @llvm.vp.fptoui.nxv8i1.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %nxv16f32_nxv16i8 = fptoui <vscale x 16 x float> undef to <vscale x 16 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %nxv16f64_nxv16i8 = fptoui <vscale x 16 x double> undef to <vscale x 16 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv16f32_nxv16i16 = fptoui <vscale x 16 x float> undef to <vscale x 16 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %nxv16f64_nxv16i16 = fptoui <vscale x 16 x double> undef to <vscale x 16 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv16f32_nxv16i32 = fptoui <vscale x 16 x float> undef to <vscale x 16 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %nxv16f64_nxv16i32 = fptoui <vscale x 16 x double> undef to <vscale x 16 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %nxv16f32_nxv16i64 = fptoui <vscale x 16 x float> undef to <vscale x 16 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv16f64_nxv16i64 = fptoui <vscale x 16 x double> undef to <vscale x 16 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %nxv16f32_nxv16i1 = fptoui <vscale x 16 x float> undef to <vscale x 16 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %nxv16f64_nxv16i1 = fptoui <vscale x 16 x double> undef to <vscale x 16 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_nxv16f32_nxv16i8 = call <vscale x 16 x i8> @llvm.vp.fptoui.nxv16i8.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_nxv16f64_nxv16i8 = call <vscale x 16 x i8> @llvm.vp.fptoui.nxv16i8.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv16f32_nxv16i16 = call <vscale x 16 x i16> @llvm.vp.fptoui.nxv16i16.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %vp_nxv16f64_nxv16i16 = call <vscale x 16 x i16> @llvm.vp.fptoui.nxv16i16.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv16f32_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.fptoui.nxv16i32.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %vp_nxv16f64_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.fptoui.nxv16i32.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_nxv16f32_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.fptoui.nxv16i64.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv16f64_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.fptoui.nxv16i64.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %vp_nxv16f32_nxv16i1 = call <vscale x 16 x i1> @llvm.vp.fptoui.nxv16i1.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_nxv16f64_nxv16i1 = call <vscale x 16 x i1> @llvm.vp.fptoui.nxv16i1.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %nxv32f32_nxv32i8 = fptoui <vscale x 32 x float> undef to <vscale x 32 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %nxv32f64_nxv32i8 = fptoui <vscale x 32 x double> undef to <vscale x 32 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %nxv32f32_nxv32i16 = fptoui <vscale x 32 x float> undef to <vscale x 32 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %nxv32f64_nxv32i16 = fptoui <vscale x 32 x double> undef to <vscale x 32 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv32f32_nxv32i32 = fptoui <vscale x 32 x float> undef to <vscale x 32 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %nxv32f64_nxv32i32 = fptoui <vscale x 32 x double> undef to <vscale x 32 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %nxv32f32_nxv32i64 = fptoui <vscale x 32 x float> undef to <vscale x 32 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv32f64_nxv32i64 = fptoui <vscale x 32 x double> undef to <vscale x 32 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %nxv32f32_nxv32i1 = fptoui <vscale x 32 x float> undef to <vscale x 32 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %nxv32f64_nxv32i1 = fptoui <vscale x 32 x double> undef to <vscale x 32 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %vp_nxv32f32_nxv32i8 = call <vscale x 32 x i8> @llvm.vp.fptoui.nxv32i8.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_nxv32f64_nxv32i8 = call <vscale x 32 x i8> @llvm.vp.fptoui.nxv32i8.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %vp_nxv32f32_nxv32i16 = call <vscale x 32 x i16> @llvm.vp.fptoui.nxv32i16.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %vp_nxv32f64_nxv32i16 = call <vscale x 32 x i16> @llvm.vp.fptoui.nxv32i16.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv32f32_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.fptoui.nxv32i32.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %vp_nxv32f64_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.fptoui.nxv32i32.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_nxv32f32_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.fptoui.nxv32i64.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv32f64_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.fptoui.nxv32i64.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_nxv32f32_nxv32i1 = call <vscale x 32 x i1> @llvm.vp.fptoui.nxv32i1.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %vp_nxv32f64_nxv32i1 = call <vscale x 32 x i1> @llvm.vp.fptoui.nxv32i1.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %nxv64f32_nxv64i8 = fptoui <vscale x 64 x float> undef to <vscale x 64 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %nxv64f64_nxv64i8 = fptoui <vscale x 64 x double> undef to <vscale x 64 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %nxv64f32_nxv64i16 = fptoui <vscale x 64 x float> undef to <vscale x 64 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %nxv64f64_nxv64i16 = fptoui <vscale x 64 x double> undef to <vscale x 64 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv64f32_nxv64i32 = fptoui <vscale x 64 x float> undef to <vscale x 64 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %nxv64f64_nxv64i32 = fptoui <vscale x 64 x double> undef to <vscale x 64 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %nxv64f32_nxv64i64 = fptoui <vscale x 64 x float> undef to <vscale x 64 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv64f64_nxv64i64 = fptoui <vscale x 64 x double> undef to <vscale x 64 x i64>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %nxv64f32_nxv64i1 = fptoui <vscale x 64 x float> undef to <vscale x 64 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 103 for instruction: %nxv64f64_nxv64i1 = fptoui <vscale x 64 x double> undef to <vscale x 64 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %vp_nxv64f32_nxv64i8 = call <vscale x 64 x i8> @llvm.vp.fptoui.nxv64i8.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_nxv64f64_nxv64i8 = call <vscale x 64 x i8> @llvm.vp.fptoui.nxv64i8.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %vp_nxv64f32_nxv64i16 = call <vscale x 64 x i16> @llvm.vp.fptoui.nxv64i16.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %vp_nxv64f64_nxv64i16 = call <vscale x 64 x i16> @llvm.vp.fptoui.nxv64i16.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv64f32_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.fptoui.nxv64i32.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %vp_nxv64f64_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.fptoui.nxv64i32.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %vp_nxv64f32_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.fptoui.nxv64i64.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv64f64_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.fptoui.nxv64i64.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %vp_nxv64f32_nxv64i1 = call <vscale x 64 x i1> @llvm.vp.fptoui.nxv64i1.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 103 for instruction: %vp_nxv64f64_nxv64i1 = call <vscale x 64 x i1> @llvm.vp.fptoui.nxv64i1.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; ZVE64X-LABEL: 'fptoui'
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v2f32_v2i8 = fptoui <2 x float> undef to <2 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v2f64_v2i8 = fptoui <2 x double> undef to <2 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v2f32_v2i16 = fptoui <2 x float> undef to <2 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v2f64_v2i16 = fptoui <2 x double> undef to <2 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v2f32_v2i32 = fptoui <2 x float> undef to <2 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v2f64_v2i32 = fptoui <2 x double> undef to <2 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2f32_v2i64 = fptoui <2 x float> undef to <2 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2f64_v2i64 = fptoui <2 x double> undef to <2 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %v2f32_v2i1 = fptoui <2 x float> undef to <2 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %v2f64_v2i1 = fptoui <2 x double> undef to <2 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v2f32_v2i8 = call <2 x i8> @llvm.vp.fptoui.v2i8.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v2f64_v2i8 = call <2 x i8> @llvm.vp.fptoui.v2i8.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v2f32_v2i16 = call <2 x i16> @llvm.vp.fptoui.v2i16.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v2f64_v2i16 = call <2 x i16> @llvm.vp.fptoui.v2i16.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v2f32_v2i32 = call <2 x i32> @llvm.vp.fptoui.v2i32.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v2f64_v2i32 = call <2 x i32> @llvm.vp.fptoui.v2i32.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2f32_v2i64 = call <2 x i64> @llvm.vp.fptoui.v2i64.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2f64_v2i64 = call <2 x i64> @llvm.vp.fptoui.v2i64.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %vp_v2f32_v2i1 = call <2 x i1> @llvm.vp.fptoui.v2i1.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %vp_v2f64_v2i1 = call <2 x i1> @llvm.vp.fptoui.v2i1.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v4f32_v4i8 = fptoui <4 x float> undef to <4 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v4f64_v4i8 = fptoui <4 x double> undef to <4 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v4f32_v4i16 = fptoui <4 x float> undef to <4 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v4f64_v4i16 = fptoui <4 x double> undef to <4 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v4f32_v4i32 = fptoui <4 x float> undef to <4 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v4f64_v4i32 = fptoui <4 x double> undef to <4 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4f32_v4i64 = fptoui <4 x float> undef to <4 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4f64_v4i64 = fptoui <4 x double> undef to <4 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 39 for instruction: %v4f32_v4i1 = fptoui <4 x float> undef to <4 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 39 for instruction: %v4f64_v4i1 = fptoui <4 x double> undef to <4 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v4f32_v4i8 = call <4 x i8> @llvm.vp.fptoui.v4i8.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v4f64_v4i8 = call <4 x i8> @llvm.vp.fptoui.v4i8.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v4f32_v4i16 = call <4 x i16> @llvm.vp.fptoui.v4i16.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v4f64_v4i16 = call <4 x i16> @llvm.vp.fptoui.v4i16.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v4f32_v4i32 = call <4 x i32> @llvm.vp.fptoui.v4i32.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v4f64_v4i32 = call <4 x i32> @llvm.vp.fptoui.v4i32.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v4f32_v4i64 = call <4 x i64> @llvm.vp.fptoui.v4i64.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v4f64_v4i64 = call <4 x i64> @llvm.vp.fptoui.v4i64.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 39 for instruction: %vp_v4f32_v4i1 = call <4 x i1> @llvm.vp.fptoui.v4i1.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 39 for instruction: %vp_v4f64_v4i1 = call <4 x i1> @llvm.vp.fptoui.v4i1.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v8f32_v8i8 = fptoui <8 x float> undef to <8 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v8f64_v8i8 = fptoui <8 x double> undef to <8 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v8f32_v8i16 = fptoui <8 x float> undef to <8 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v8f64_v8i16 = fptoui <8 x double> undef to <8 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v8f32_v8i32 = fptoui <8 x float> undef to <8 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v8f64_v8i32 = fptoui <8 x double> undef to <8 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v8f32_v8i64 = fptoui <8 x float> undef to <8 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v8f64_v8i64 = fptoui <8 x double> undef to <8 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 79 for instruction: %v8f32_v8i1 = fptoui <8 x float> undef to <8 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 79 for instruction: %v8f64_v8i1 = fptoui <8 x double> undef to <8 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v8f32_v8i8 = call <8 x i8> @llvm.vp.fptoui.v8i8.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v8f64_v8i8 = call <8 x i8> @llvm.vp.fptoui.v8i8.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v8f32_v8i16 = call <8 x i16> @llvm.vp.fptoui.v8i16.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v8f64_v8i16 = call <8 x i16> @llvm.vp.fptoui.v8i16.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v8f32_v8i32 = call <8 x i32> @llvm.vp.fptoui.v8i32.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v8f64_v8i32 = call <8 x i32> @llvm.vp.fptoui.v8i32.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v8f32_v8i64 = call <8 x i64> @llvm.vp.fptoui.v8i64.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v8f64_v8i64 = call <8 x i64> @llvm.vp.fptoui.v8i64.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 79 for instruction: %vp_v8f32_v8i1 = call <8 x i1> @llvm.vp.fptoui.v8i1.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 79 for instruction: %vp_v8f64_v8i1 = call <8 x i1> @llvm.vp.fptoui.v8i1.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v16f32_v16i8 = fptoui <16 x float> undef to <16 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v16f64_v16i8 = fptoui <16 x double> undef to <16 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v16f32_v16i16 = fptoui <16 x float> undef to <16 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v16f64_v16i16 = fptoui <16 x double> undef to <16 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v16f32_v16i32 = fptoui <16 x float> undef to <16 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v16f64_v16i32 = fptoui <16 x double> undef to <16 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v16f32_v16i64 = fptoui <16 x float> undef to <16 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v16f64_v16i64 = fptoui <16 x double> undef to <16 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 159 for instruction: %v16f32_v16i1 = fptoui <16 x float> undef to <16 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 159 for instruction: %v16f64_v16i1 = fptoui <16 x double> undef to <16 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v16f32_v16i8 = call <16 x i8> @llvm.vp.fptoui.v16i8.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v16f64_v16i8 = call <16 x i8> @llvm.vp.fptoui.v16i8.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v16f32_v16i16 = call <16 x i16> @llvm.vp.fptoui.v16i16.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v16f64_v16i16 = call <16 x i16> @llvm.vp.fptoui.v16i16.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v16f32_v16i32 = call <16 x i32> @llvm.vp.fptoui.v16i32.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v16f64_v16i32 = call <16 x i32> @llvm.vp.fptoui.v16i32.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v16f32_v16i64 = call <16 x i64> @llvm.vp.fptoui.v16i64.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v16f64_v16i64 = call <16 x i64> @llvm.vp.fptoui.v16i64.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 159 for instruction: %vp_v16f32_v16i1 = call <16 x i1> @llvm.vp.fptoui.v16i1.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 159 for instruction: %vp_v16f64_v16i1 = call <16 x i1> @llvm.vp.fptoui.v16i1.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %v32f32_v32i8 = fptoui <32 x float> undef to <32 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %v32f64_v32i8 = fptoui <32 x double> undef to <32 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %v32f32_v32i16 = fptoui <32 x float> undef to <32 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %v32f64_v32i16 = fptoui <32 x double> undef to <32 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %v32f32_v32i32 = fptoui <32 x float> undef to <32 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %v32f64_v32i32 = fptoui <32 x double> undef to <32 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 62 for instruction: %v32f32_v32i64 = fptoui <32 x float> undef to <32 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 62 for instruction: %v32f64_v32i64 = fptoui <32 x double> undef to <32 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 319 for instruction: %v32f32_v32i1 = fptoui <32 x float> undef to <32 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 319 for instruction: %v32f64_v32i1 = fptoui <32 x double> undef to <32 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %vp_v32f32_v32i8 = call <32 x i8> @llvm.vp.fptoui.v32i8.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %vp_v32f64_v32i8 = call <32 x i8> @llvm.vp.fptoui.v32i8.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %vp_v32f32_v32i16 = call <32 x i16> @llvm.vp.fptoui.v32i16.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %vp_v32f64_v32i16 = call <32 x i16> @llvm.vp.fptoui.v32i16.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %vp_v32f32_v32i32 = call <32 x i32> @llvm.vp.fptoui.v32i32.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %vp_v32f64_v32i32 = call <32 x i32> @llvm.vp.fptoui.v32i32.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 62 for instruction: %vp_v32f32_v32i64 = call <32 x i64> @llvm.vp.fptoui.v32i64.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 62 for instruction: %vp_v32f64_v32i64 = call <32 x i64> @llvm.vp.fptoui.v32i64.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 319 for instruction: %vp_v32f32_v32i1 = call <32 x i1> @llvm.vp.fptoui.v32i1.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 319 for instruction: %vp_v32f64_v32i1 = call <32 x i1> @llvm.vp.fptoui.v32i1.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %v64f32_v64i8 = fptoui <64 x float> undef to <64 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %v64f64_v64i8 = fptoui <64 x double> undef to <64 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %v64f32_v64i16 = fptoui <64 x float> undef to <64 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %v64f64_v64i16 = fptoui <64 x double> undef to <64 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 254 for instruction: %v64f32_v64i32 = fptoui <64 x float> undef to <64 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 254 for instruction: %v64f64_v64i32 = fptoui <64 x double> undef to <64 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 124 for instruction: %v64f32_v64i64 = fptoui <64 x float> undef to <64 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 124 for instruction: %v64f64_v64i64 = fptoui <64 x double> undef to <64 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 639 for instruction: %v64f32_v64i1 = fptoui <64 x float> undef to <64 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 639 for instruction: %v64f64_v64i1 = fptoui <64 x double> undef to <64 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %vp_v64f32_v64i8 = call <64 x i8> @llvm.vp.fptoui.v64i8.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %vp_v64f64_v64i8 = call <64 x i8> @llvm.vp.fptoui.v64i8.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %vp_v64f32_v64i16 = call <64 x i16> @llvm.vp.fptoui.v64i16.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %vp_v64f64_v64i16 = call <64 x i16> @llvm.vp.fptoui.v64i16.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 254 for instruction: %vp_v64f32_v64i32 = call <64 x i32> @llvm.vp.fptoui.v64i32.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 254 for instruction: %vp_v64f64_v64i32 = call <64 x i32> @llvm.vp.fptoui.v64i32.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 124 for instruction: %vp_v64f32_v64i64 = call <64 x i64> @llvm.vp.fptoui.v64i64.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 124 for instruction: %vp_v64f64_v64i64 = call <64 x i64> @llvm.vp.fptoui.v64i64.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 639 for instruction: %vp_v64f32_v64i1 = call <64 x i1> @llvm.vp.fptoui.v64i1.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 639 for instruction: %vp_v64f64_v64i1 = call <64 x i1> @llvm.vp.fptoui.v64i1.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 511 for instruction: %v128f32_v128i8 = fptoui <128 x float> undef to <128 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 511 for instruction: %v128f64_v128i8 = fptoui <128 x double> undef to <128 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 510 for instruction: %v128f32_v128i16 = fptoui <128 x float> undef to <128 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 510 for instruction: %v128f64_v128i16 = fptoui <128 x double> undef to <128 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 508 for instruction: %v128f32_v128i32 = fptoui <128 x float> undef to <128 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 508 for instruction: %v128f64_v128i32 = fptoui <128 x double> undef to <128 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 248 for instruction: %v128f32_v128i64 = fptoui <128 x float> undef to <128 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 248 for instruction: %v128f64_v128i64 = fptoui <128 x double> undef to <128 x i64>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 1279 for instruction: %v128f32_v128i1 = fptoui <128 x float> undef to <128 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 1279 for instruction: %v128f64_v128i1 = fptoui <128 x double> undef to <128 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 511 for instruction: %vp_v128f32_v128i8 = call <128 x i8> @llvm.vp.fptoui.v128i8.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 511 for instruction: %vp_v128f64_v128i8 = call <128 x i8> @llvm.vp.fptoui.v128i8.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 510 for instruction: %vp_v128f32_v128i16 = call <128 x i16> @llvm.vp.fptoui.v128i16.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 510 for instruction: %vp_v128f64_v128i16 = call <128 x i16> @llvm.vp.fptoui.v128i16.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 508 for instruction: %vp_v128f32_v128i32 = call <128 x i32> @llvm.vp.fptoui.v128i32.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 508 for instruction: %vp_v128f64_v128i32 = call <128 x i32> @llvm.vp.fptoui.v128i32.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 248 for instruction: %vp_v128f32_v128i64 = call <128 x i64> @llvm.vp.fptoui.v128i64.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 248 for instruction: %vp_v128f64_v128i64 = call <128 x i64> @llvm.vp.fptoui.v128i64.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 1279 for instruction: %vp_v128f32_v128i1 = call <128 x i1> @llvm.vp.fptoui.v128i1.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 1279 for instruction: %vp_v128f64_v128i1 = call <128 x i1> @llvm.vp.fptoui.v128i1.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f32_nxv1i8 = fptoui <vscale x 1 x float> undef to <vscale x 1 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f64_nxv1i8 = fptoui <vscale x 1 x double> undef to <vscale x 1 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f32_nxv1i16 = fptoui <vscale x 1 x float> undef to <vscale x 1 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f64_nxv1i16 = fptoui <vscale x 1 x double> undef to <vscale x 1 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f32_nxv1i32 = fptoui <vscale x 1 x float> undef to <vscale x 1 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f64_nxv1i32 = fptoui <vscale x 1 x double> undef to <vscale x 1 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f32_nxv1i64 = fptoui <vscale x 1 x float> undef to <vscale x 1 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f64_nxv1i64 = fptoui <vscale x 1 x double> undef to <vscale x 1 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f32_nxv1i1 = fptoui <vscale x 1 x float> undef to <vscale x 1 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1f64_nxv1i1 = fptoui <vscale x 1 x double> undef to <vscale x 1 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f32_nxv1i8 = call <vscale x 1 x i8> @llvm.vp.fptoui.nxv1i8.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f64_nxv1i8 = call <vscale x 1 x i8> @llvm.vp.fptoui.nxv1i8.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f32_nxv1i16 = call <vscale x 1 x i16> @llvm.vp.fptoui.nxv1i16.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f64_nxv1i16 = call <vscale x 1 x i16> @llvm.vp.fptoui.nxv1i16.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f32_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.fptoui.nxv1i32.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f64_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.fptoui.nxv1i32.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f32_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.fptoui.nxv1i64.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f64_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.fptoui.nxv1i64.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f32_nxv1i1 = call <vscale x 1 x i1> @llvm.vp.fptoui.nxv1i1.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1f64_nxv1i1 = call <vscale x 1 x i1> @llvm.vp.fptoui.nxv1i1.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f32_nxv2i8 = fptoui <vscale x 2 x float> undef to <vscale x 2 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f64_nxv2i8 = fptoui <vscale x 2 x double> undef to <vscale x 2 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f32_nxv2i16 = fptoui <vscale x 2 x float> undef to <vscale x 2 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f64_nxv2i16 = fptoui <vscale x 2 x double> undef to <vscale x 2 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f32_nxv2i32 = fptoui <vscale x 2 x float> undef to <vscale x 2 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f64_nxv2i32 = fptoui <vscale x 2 x double> undef to <vscale x 2 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f32_nxv2i64 = fptoui <vscale x 2 x float> undef to <vscale x 2 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f64_nxv2i64 = fptoui <vscale x 2 x double> undef to <vscale x 2 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f32_nxv2i1 = fptoui <vscale x 2 x float> undef to <vscale x 2 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2f64_nxv2i1 = fptoui <vscale x 2 x double> undef to <vscale x 2 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f32_nxv2i8 = call <vscale x 2 x i8> @llvm.vp.fptoui.nxv2i8.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f64_nxv2i8 = call <vscale x 2 x i8> @llvm.vp.fptoui.nxv2i8.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f32_nxv2i16 = call <vscale x 2 x i16> @llvm.vp.fptoui.nxv2i16.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f64_nxv2i16 = call <vscale x 2 x i16> @llvm.vp.fptoui.nxv2i16.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f32_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.fptoui.nxv2i32.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f64_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.fptoui.nxv2i32.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f32_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.fptoui.nxv2i64.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f64_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.fptoui.nxv2i64.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f32_nxv2i1 = call <vscale x 2 x i1> @llvm.vp.fptoui.nxv2i1.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2f64_nxv2i1 = call <vscale x 2 x i1> @llvm.vp.fptoui.nxv2i1.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f32_nxv4i8 = fptoui <vscale x 4 x float> undef to <vscale x 4 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f64_nxv4i8 = fptoui <vscale x 4 x double> undef to <vscale x 4 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f32_nxv4i16 = fptoui <vscale x 4 x float> undef to <vscale x 4 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f64_nxv4i16 = fptoui <vscale x 4 x double> undef to <vscale x 4 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f32_nxv4i32 = fptoui <vscale x 4 x float> undef to <vscale x 4 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f64_nxv4i32 = fptoui <vscale x 4 x double> undef to <vscale x 4 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f32_nxv4i64 = fptoui <vscale x 4 x float> undef to <vscale x 4 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f64_nxv4i64 = fptoui <vscale x 4 x double> undef to <vscale x 4 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f32_nxv4i1 = fptoui <vscale x 4 x float> undef to <vscale x 4 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4f64_nxv4i1 = fptoui <vscale x 4 x double> undef to <vscale x 4 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f32_nxv4i8 = call <vscale x 4 x i8> @llvm.vp.fptoui.nxv4i8.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f64_nxv4i8 = call <vscale x 4 x i8> @llvm.vp.fptoui.nxv4i8.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f32_nxv4i16 = call <vscale x 4 x i16> @llvm.vp.fptoui.nxv4i16.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f64_nxv4i16 = call <vscale x 4 x i16> @llvm.vp.fptoui.nxv4i16.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f32_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.fptoui.nxv4i32.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f64_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.fptoui.nxv4i32.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f32_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.fptoui.nxv4i64.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f64_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.fptoui.nxv4i64.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f32_nxv4i1 = call <vscale x 4 x i1> @llvm.vp.fptoui.nxv4i1.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4f64_nxv4i1 = call <vscale x 4 x i1> @llvm.vp.fptoui.nxv4i1.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f32_nxv8i8 = fptoui <vscale x 8 x float> undef to <vscale x 8 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f64_nxv8i8 = fptoui <vscale x 8 x double> undef to <vscale x 8 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f32_nxv8i16 = fptoui <vscale x 8 x float> undef to <vscale x 8 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f64_nxv8i16 = fptoui <vscale x 8 x double> undef to <vscale x 8 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f32_nxv8i32 = fptoui <vscale x 8 x float> undef to <vscale x 8 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f64_nxv8i32 = fptoui <vscale x 8 x double> undef to <vscale x 8 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f32_nxv8i64 = fptoui <vscale x 8 x float> undef to <vscale x 8 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f64_nxv8i64 = fptoui <vscale x 8 x double> undef to <vscale x 8 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f32_nxv8i1 = fptoui <vscale x 8 x float> undef to <vscale x 8 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8f64_nxv8i1 = fptoui <vscale x 8 x double> undef to <vscale x 8 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f32_nxv8i8 = call <vscale x 8 x i8> @llvm.vp.fptoui.nxv8i8.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f64_nxv8i8 = call <vscale x 8 x i8> @llvm.vp.fptoui.nxv8i8.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f32_nxv8i16 = call <vscale x 8 x i16> @llvm.vp.fptoui.nxv8i16.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f64_nxv8i16 = call <vscale x 8 x i16> @llvm.vp.fptoui.nxv8i16.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f32_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.fptoui.nxv8i32.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f64_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.fptoui.nxv8i32.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f32_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.fptoui.nxv8i64.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f64_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.fptoui.nxv8i64.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f32_nxv8i1 = call <vscale x 8 x i1> @llvm.vp.fptoui.nxv8i1.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8f64_nxv8i1 = call <vscale x 8 x i1> @llvm.vp.fptoui.nxv8i1.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f32_nxv16i8 = fptoui <vscale x 16 x float> undef to <vscale x 16 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f64_nxv16i8 = fptoui <vscale x 16 x double> undef to <vscale x 16 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f32_nxv16i16 = fptoui <vscale x 16 x float> undef to <vscale x 16 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f64_nxv16i16 = fptoui <vscale x 16 x double> undef to <vscale x 16 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f32_nxv16i32 = fptoui <vscale x 16 x float> undef to <vscale x 16 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f64_nxv16i32 = fptoui <vscale x 16 x double> undef to <vscale x 16 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f32_nxv16i64 = fptoui <vscale x 16 x float> undef to <vscale x 16 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f64_nxv16i64 = fptoui <vscale x 16 x double> undef to <vscale x 16 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f32_nxv16i1 = fptoui <vscale x 16 x float> undef to <vscale x 16 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16f64_nxv16i1 = fptoui <vscale x 16 x double> undef to <vscale x 16 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f32_nxv16i8 = call <vscale x 16 x i8> @llvm.vp.fptoui.nxv16i8.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f64_nxv16i8 = call <vscale x 16 x i8> @llvm.vp.fptoui.nxv16i8.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f32_nxv16i16 = call <vscale x 16 x i16> @llvm.vp.fptoui.nxv16i16.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f64_nxv16i16 = call <vscale x 16 x i16> @llvm.vp.fptoui.nxv16i16.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f32_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.fptoui.nxv16i32.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f64_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.fptoui.nxv16i32.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f32_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.fptoui.nxv16i64.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f64_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.fptoui.nxv16i64.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f32_nxv16i1 = call <vscale x 16 x i1> @llvm.vp.fptoui.nxv16i1.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16f64_nxv16i1 = call <vscale x 16 x i1> @llvm.vp.fptoui.nxv16i1.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f32_nxv32i8 = fptoui <vscale x 32 x float> undef to <vscale x 32 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f64_nxv32i8 = fptoui <vscale x 32 x double> undef to <vscale x 32 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f32_nxv32i16 = fptoui <vscale x 32 x float> undef to <vscale x 32 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f64_nxv32i16 = fptoui <vscale x 32 x double> undef to <vscale x 32 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f32_nxv32i32 = fptoui <vscale x 32 x float> undef to <vscale x 32 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f64_nxv32i32 = fptoui <vscale x 32 x double> undef to <vscale x 32 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f32_nxv32i64 = fptoui <vscale x 32 x float> undef to <vscale x 32 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f64_nxv32i64 = fptoui <vscale x 32 x double> undef to <vscale x 32 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f32_nxv32i1 = fptoui <vscale x 32 x float> undef to <vscale x 32 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32f64_nxv32i1 = fptoui <vscale x 32 x double> undef to <vscale x 32 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f32_nxv32i8 = call <vscale x 32 x i8> @llvm.vp.fptoui.nxv32i8.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f64_nxv32i8 = call <vscale x 32 x i8> @llvm.vp.fptoui.nxv32i8.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f32_nxv32i16 = call <vscale x 32 x i16> @llvm.vp.fptoui.nxv32i16.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f64_nxv32i16 = call <vscale x 32 x i16> @llvm.vp.fptoui.nxv32i16.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f32_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.fptoui.nxv32i32.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f64_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.fptoui.nxv32i32.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f32_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.fptoui.nxv32i64.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f64_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.fptoui.nxv32i64.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f32_nxv32i1 = call <vscale x 32 x i1> @llvm.vp.fptoui.nxv32i1.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32f64_nxv32i1 = call <vscale x 32 x i1> @llvm.vp.fptoui.nxv32i1.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f32_nxv64i8 = fptoui <vscale x 64 x float> undef to <vscale x 64 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f64_nxv64i8 = fptoui <vscale x 64 x double> undef to <vscale x 64 x i8>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f32_nxv64i16 = fptoui <vscale x 64 x float> undef to <vscale x 64 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f64_nxv64i16 = fptoui <vscale x 64 x double> undef to <vscale x 64 x i16>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f32_nxv64i32 = fptoui <vscale x 64 x float> undef to <vscale x 64 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f64_nxv64i32 = fptoui <vscale x 64 x double> undef to <vscale x 64 x i32>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f32_nxv64i64 = fptoui <vscale x 64 x float> undef to <vscale x 64 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f64_nxv64i64 = fptoui <vscale x 64 x double> undef to <vscale x 64 x i64>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f32_nxv64i1 = fptoui <vscale x 64 x float> undef to <vscale x 64 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64f64_nxv64i1 = fptoui <vscale x 64 x double> undef to <vscale x 64 x i1>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f32_nxv64i8 = call <vscale x 64 x i8> @llvm.vp.fptoui.nxv64i8.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f64_nxv64i8 = call <vscale x 64 x i8> @llvm.vp.fptoui.nxv64i8.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f32_nxv64i16 = call <vscale x 64 x i16> @llvm.vp.fptoui.nxv64i16.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f64_nxv64i16 = call <vscale x 64 x i16> @llvm.vp.fptoui.nxv64i16.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f32_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.fptoui.nxv64i32.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f64_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.fptoui.nxv64i32.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f32_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.fptoui.nxv64i64.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f64_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.fptoui.nxv64i64.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f32_nxv64i1 = call <vscale x 64 x i1> @llvm.vp.fptoui.nxv64i1.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64f64_nxv64i1 = call <vscale x 64 x i1> @llvm.vp.fptoui.nxv64i1.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  %v2f32_v2i8 = fptoui <2 x float> undef to <2 x i8>
  %v2f64_v2i8 = fptoui <2 x double> undef to <2 x i8>
  %v2f32_v2i16 = fptoui <2 x float> undef to <2 x i16>
  %v2f64_v2i16 = fptoui <2 x double> undef to <2 x i16>
  %v2f32_v2i32 = fptoui <2 x float> undef to <2 x i32>
  %v2f64_v2i32 = fptoui <2 x double> undef to <2 x i32>
  %v2f32_v2i64 = fptoui <2 x float> undef to <2 x i64>
  %v2f64_v2i64 = fptoui <2 x double> undef to <2 x i64>
  %v2f32_v2i1 = fptoui <2 x float> undef to <2 x i1>
  %v2f64_v2i1 = fptoui <2 x double> undef to <2 x i1>

  %vp_v2f32_v2i8 = call <2 x i8> @llvm.vp.fptoui.v2i8.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
  %vp_v2f64_v2i8 = call <2 x i8> @llvm.vp.fptoui.v2i8.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
  %vp_v2f32_v2i16 = call <2 x i16> @llvm.vp.fptoui.v2i16.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
  %vp_v2f64_v2i16 = call <2 x i16> @llvm.vp.fptoui.v2i16.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
  %vp_v2f32_v2i32 = call <2 x i32> @llvm.vp.fptoui.v2i32.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
  %vp_v2f64_v2i32 = call <2 x i32> @llvm.vp.fptoui.v2i32.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
  %vp_v2f32_v2i64 = call <2 x i64> @llvm.vp.fptoui.v2i64.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
  %vp_v2f64_v2i64 = call <2 x i64> @llvm.vp.fptoui.v2i64.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
  %vp_v2f32_v2i1 = call <2 x i1> @llvm.vp.fptoui.v2i1.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
  %vp_v2f64_v2i1 = call <2 x i1> @llvm.vp.fptoui.v2i1.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)

  %v4f32_v4i8 = fptoui <4 x float> undef to <4 x i8>
  %v4f64_v4i8 = fptoui <4 x double> undef to <4 x i8>
  %v4f32_v4i16 = fptoui <4 x float> undef to <4 x i16>
  %v4f64_v4i16 = fptoui <4 x double> undef to <4 x i16>
  %v4f32_v4i32 = fptoui <4 x float> undef to <4 x i32>
  %v4f64_v4i32 = fptoui <4 x double> undef to <4 x i32>
  %v4f32_v4i64 = fptoui <4 x float> undef to <4 x i64>
  %v4f64_v4i64 = fptoui <4 x double> undef to <4 x i64>
  %v4f32_v4i1 = fptoui <4 x float> undef to <4 x i1>
  %v4f64_v4i1 = fptoui <4 x double> undef to <4 x i1>

  %vp_v4f32_v4i8 = call <4 x i8> @llvm.vp.fptoui.v4i8.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
  %vp_v4f64_v4i8 = call <4 x i8> @llvm.vp.fptoui.v4i8.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
  %vp_v4f32_v4i16 = call <4 x i16> @llvm.vp.fptoui.v4i16.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
  %vp_v4f64_v4i16 = call <4 x i16> @llvm.vp.fptoui.v4i16.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
  %vp_v4f32_v4i32 = call <4 x i32> @llvm.vp.fptoui.v4i32.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
  %vp_v4f64_v4i32 = call <4 x i32> @llvm.vp.fptoui.v4i32.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
  %vp_v4f32_v4i64 = call <4 x i64> @llvm.vp.fptoui.v4i64.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
  %vp_v4f64_v4i64 = call <4 x i64> @llvm.vp.fptoui.v4i64.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
  %vp_v4f32_v4i1 = call <4 x i1> @llvm.vp.fptoui.v4i1.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
  %vp_v4f64_v4i1 = call <4 x i1> @llvm.vp.fptoui.v4i1.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)

  %v8f32_v8i8 = fptoui <8 x float> undef to <8 x i8>
  %v8f64_v8i8 = fptoui <8 x double> undef to <8 x i8>
  %v8f32_v8i16 = fptoui <8 x float> undef to <8 x i16>
  %v8f64_v8i16 = fptoui <8 x double> undef to <8 x i16>
  %v8f32_v8i32 = fptoui <8 x float> undef to <8 x i32>
  %v8f64_v8i32 = fptoui <8 x double> undef to <8 x i32>
  %v8f32_v8i64 = fptoui <8 x float> undef to <8 x i64>
  %v8f64_v8i64 = fptoui <8 x double> undef to <8 x i64>
  %v8f32_v8i1 = fptoui <8 x float> undef to <8 x i1>
  %v8f64_v8i1 = fptoui <8 x double> undef to <8 x i1>

  %vp_v8f32_v8i8 = call <8 x i8> @llvm.vp.fptoui.v8i8.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
  %vp_v8f64_v8i8 = call <8 x i8> @llvm.vp.fptoui.v8i8.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
  %vp_v8f32_v8i16 = call <8 x i16> @llvm.vp.fptoui.v8i16.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
  %vp_v8f64_v8i16 = call <8 x i16> @llvm.vp.fptoui.v8i16.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
  %vp_v8f32_v8i32 = call <8 x i32> @llvm.vp.fptoui.v8i32.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
  %vp_v8f64_v8i32 = call <8 x i32> @llvm.vp.fptoui.v8i32.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
  %vp_v8f32_v8i64 = call <8 x i64> @llvm.vp.fptoui.v8i64.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
  %vp_v8f64_v8i64 = call <8 x i64> @llvm.vp.fptoui.v8i64.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
  %vp_v8f32_v8i1 = call <8 x i1> @llvm.vp.fptoui.v8i1.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
  %vp_v8f64_v8i1 = call <8 x i1> @llvm.vp.fptoui.v8i1.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)

  %v16f32_v16i8 = fptoui <16 x float> undef to <16 x i8>
  %v16f64_v16i8 = fptoui <16 x double> undef to <16 x i8>
  %v16f32_v16i16 = fptoui <16 x float> undef to <16 x i16>
  %v16f64_v16i16 = fptoui <16 x double> undef to <16 x i16>
  %v16f32_v16i32 = fptoui <16 x float> undef to <16 x i32>
  %v16f64_v16i32 = fptoui <16 x double> undef to <16 x i32>
  %v16f32_v16i64 = fptoui <16 x float> undef to <16 x i64>
  %v16f64_v16i64 = fptoui <16 x double> undef to <16 x i64>
  %v16f32_v16i1 = fptoui <16 x float> undef to <16 x i1>
  %v16f64_v16i1 = fptoui <16 x double> undef to <16 x i1>

  %vp_v16f32_v16i8 = call <16 x i8> @llvm.vp.fptoui.v16i8.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
  %vp_v16f64_v16i8 = call <16 x i8> @llvm.vp.fptoui.v16i8.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
  %vp_v16f32_v16i16 = call <16 x i16> @llvm.vp.fptoui.v16i16.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
  %vp_v16f64_v16i16 = call <16 x i16> @llvm.vp.fptoui.v16i16.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
  %vp_v16f32_v16i32 = call <16 x i32> @llvm.vp.fptoui.v16i32.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
  %vp_v16f64_v16i32 = call <16 x i32> @llvm.vp.fptoui.v16i32.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
  %vp_v16f32_v16i64 = call <16 x i64> @llvm.vp.fptoui.v16i64.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
  %vp_v16f64_v16i64 = call <16 x i64> @llvm.vp.fptoui.v16i64.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
  %vp_v16f32_v16i1 = call <16 x i1> @llvm.vp.fptoui.v16i1.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
  %vp_v16f64_v16i1 = call <16 x i1> @llvm.vp.fptoui.v16i1.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)

  %v32f32_v32i8 = fptoui <32 x float> undef to <32 x i8>
  %v32f64_v32i8 = fptoui <32 x double> undef to <32 x i8>
  %v32f32_v32i16 = fptoui <32 x float> undef to <32 x i16>
  %v32f64_v32i16 = fptoui <32 x double> undef to <32 x i16>
  %v32f32_v32i32 = fptoui <32 x float> undef to <32 x i32>
  %v32f64_v32i32 = fptoui <32 x double> undef to <32 x i32>
  %v32f32_v32i64 = fptoui <32 x float> undef to <32 x i64>
  %v32f64_v32i64 = fptoui <32 x double> undef to <32 x i64>
  %v32f32_v32i1 = fptoui <32 x float> undef to <32 x i1>
  %v32f64_v32i1 = fptoui <32 x double> undef to <32 x i1>

  %vp_v32f32_v32i8 = call <32 x i8> @llvm.vp.fptoui.v32i8.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
  %vp_v32f64_v32i8 = call <32 x i8> @llvm.vp.fptoui.v32i8.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
  %vp_v32f32_v32i16 = call <32 x i16> @llvm.vp.fptoui.v32i16.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
  %vp_v32f64_v32i16 = call <32 x i16> @llvm.vp.fptoui.v32i16.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
  %vp_v32f32_v32i32 = call <32 x i32> @llvm.vp.fptoui.v32i32.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
  %vp_v32f64_v32i32 = call <32 x i32> @llvm.vp.fptoui.v32i32.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
  %vp_v32f32_v32i64 = call <32 x i64> @llvm.vp.fptoui.v32i64.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
  %vp_v32f64_v32i64 = call <32 x i64> @llvm.vp.fptoui.v32i64.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)
  %vp_v32f32_v32i1 = call <32 x i1> @llvm.vp.fptoui.v32i1.v32f32(<32 x float> undef, <32 x i1> undef, i32 undef)
  %vp_v32f64_v32i1 = call <32 x i1> @llvm.vp.fptoui.v32i1.v32f64(<32 x double> undef, <32 x i1> undef, i32 undef)

  %v64f32_v64i8 = fptoui <64 x float> undef to <64 x i8>
  %v64f64_v64i8 = fptoui <64 x double> undef to <64 x i8>
  %v64f32_v64i16 = fptoui <64 x float> undef to <64 x i16>
  %v64f64_v64i16 = fptoui <64 x double> undef to <64 x i16>
  %v64f32_v64i32 = fptoui <64 x float> undef to <64 x i32>
  %v64f64_v64i32 = fptoui <64 x double> undef to <64 x i32>
  %v64f32_v64i64 = fptoui <64 x float> undef to <64 x i64>
  %v64f64_v64i64 = fptoui <64 x double> undef to <64 x i64>
  %v64f32_v64i1 = fptoui <64 x float> undef to <64 x i1>
  %v64f64_v64i1 = fptoui <64 x double> undef to <64 x i1>

  %vp_v64f32_v64i8 = call <64 x i8> @llvm.vp.fptoui.v64i8.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
  %vp_v64f64_v64i8 = call <64 x i8> @llvm.vp.fptoui.v64i8.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
  %vp_v64f32_v64i16 = call <64 x i16> @llvm.vp.fptoui.v64i16.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
  %vp_v64f64_v64i16 = call <64 x i16> @llvm.vp.fptoui.v64i16.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
  %vp_v64f32_v64i32 = call <64 x i32> @llvm.vp.fptoui.v64i32.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
  %vp_v64f64_v64i32 = call <64 x i32> @llvm.vp.fptoui.v64i32.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
  %vp_v64f32_v64i64 = call <64 x i64> @llvm.vp.fptoui.v64i64.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
  %vp_v64f64_v64i64 = call <64 x i64> @llvm.vp.fptoui.v64i64.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)
  %vp_v64f32_v64i1 = call <64 x i1> @llvm.vp.fptoui.v64i1.v64f32(<64 x float> undef, <64 x i1> undef, i32 undef)
  %vp_v64f64_v64i1 = call <64 x i1> @llvm.vp.fptoui.v64i1.v64f64(<64 x double> undef, <64 x i1> undef, i32 undef)

  %v128f32_v128i8 = fptoui <128 x float> undef to <128 x i8>
  %v128f64_v128i8 = fptoui <128 x double> undef to <128 x i8>
  %v128f32_v128i16 = fptoui <128 x float> undef to <128 x i16>
  %v128f64_v128i16 = fptoui <128 x double> undef to <128 x i16>
  %v128f32_v128i32 = fptoui <128 x float> undef to <128 x i32>
  %v128f64_v128i32 = fptoui <128 x double> undef to <128 x i32>
  %v128f32_v128i64 = fptoui <128 x float> undef to <128 x i64>
  %v128f64_v128i64 = fptoui <128 x double> undef to <128 x i64>
  %v128f32_v128i1 = fptoui <128 x float> undef to <128 x i1>
  %v128f64_v128i1 = fptoui <128 x double> undef to <128 x i1>

  %vp_v128f32_v128i8 = call <128 x i8> @llvm.vp.fptoui.v128i8.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
  %vp_v128f64_v128i8 = call <128 x i8> @llvm.vp.fptoui.v128i8.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
  %vp_v128f32_v128i16 = call <128 x i16> @llvm.vp.fptoui.v128i16.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
  %vp_v128f64_v128i16 = call <128 x i16> @llvm.vp.fptoui.v128i16.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
  %vp_v128f32_v128i32 = call <128 x i32> @llvm.vp.fptoui.v128i32.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
  %vp_v128f64_v128i32 = call <128 x i32> @llvm.vp.fptoui.v128i32.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
  %vp_v128f32_v128i64 = call <128 x i64> @llvm.vp.fptoui.v128i64.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
  %vp_v128f64_v128i64 = call <128 x i64> @llvm.vp.fptoui.v128i64.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)
  %vp_v128f32_v128i1 = call <128 x i1> @llvm.vp.fptoui.v128i1.v128f32(<128 x float> undef, <128 x i1> undef, i32 undef)
  %vp_v128f64_v128i1 = call <128 x i1> @llvm.vp.fptoui.v128i1.v128f64(<128 x double> undef, <128 x i1> undef, i32 undef)

  %nxv1f32_nxv1i8 = fptoui <vscale x 1 x float> undef to <vscale x 1 x i8>
  %nxv1f64_nxv1i8 = fptoui <vscale x 1 x double> undef to <vscale x 1 x i8>
  %nxv1f32_nxv1i16 = fptoui <vscale x 1 x float> undef to <vscale x 1 x i16>
  %nxv1f64_nxv1i16 = fptoui <vscale x 1 x double> undef to <vscale x 1 x i16>
  %nxv1f32_nxv1i32 = fptoui <vscale x 1 x float> undef to <vscale x 1 x i32>
  %nxv1f64_nxv1i32 = fptoui <vscale x 1 x double> undef to <vscale x 1 x i32>
  %nxv1f32_nxv1i64 = fptoui <vscale x 1 x float> undef to <vscale x 1 x i64>
  %nxv1f64_nxv1i64 = fptoui <vscale x 1 x double> undef to <vscale x 1 x i64>
  %nxv1f32_nxv1i1 = fptoui <vscale x 1 x float> undef to <vscale x 1 x i1>
  %nxv1f64_nxv1i1 = fptoui <vscale x 1 x double> undef to <vscale x 1 x i1>

  %vp_nxv1f32_nxv1i8 = call <vscale x 1 x i8> @llvm.vp.fptoui.nxv1i8.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1f64_nxv1i8 = call <vscale x 1 x i8> @llvm.vp.fptoui.nxv1i8.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1f32_nxv1i16 = call <vscale x 1 x i16> @llvm.vp.fptoui.nxv1i16.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1f64_nxv1i16 = call <vscale x 1 x i16> @llvm.vp.fptoui.nxv1i16.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1f32_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.fptoui.nxv1i32.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1f64_nxv1i32 = call <vscale x 1 x i32> @llvm.vp.fptoui.nxv1i32.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1f32_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.fptoui.nxv1i64.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1f64_nxv1i64 = call <vscale x 1 x i64> @llvm.vp.fptoui.nxv1i64.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1f32_nxv1i1 = call <vscale x 1 x i1> @llvm.vp.fptoui.nxv1i1.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1f64_nxv1i1 = call <vscale x 1 x i1> @llvm.vp.fptoui.nxv1i1.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)

  %nxv2f32_nxv2i8 = fptoui <vscale x 2 x float> undef to <vscale x 2 x i8>
  %nxv2f64_nxv2i8 = fptoui <vscale x 2 x double> undef to <vscale x 2 x i8>
  %nxv2f32_nxv2i16 = fptoui <vscale x 2 x float> undef to <vscale x 2 x i16>
  %nxv2f64_nxv2i16 = fptoui <vscale x 2 x double> undef to <vscale x 2 x i16>
  %nxv2f32_nxv2i32 = fptoui <vscale x 2 x float> undef to <vscale x 2 x i32>
  %nxv2f64_nxv2i32 = fptoui <vscale x 2 x double> undef to <vscale x 2 x i32>
  %nxv2f32_nxv2i64 = fptoui <vscale x 2 x float> undef to <vscale x 2 x i64>
  %nxv2f64_nxv2i64 = fptoui <vscale x 2 x double> undef to <vscale x 2 x i64>
  %nxv2f32_nxv2i1 = fptoui <vscale x 2 x float> undef to <vscale x 2 x i1>
  %nxv2f64_nxv2i1 = fptoui <vscale x 2 x double> undef to <vscale x 2 x i1>

  %vp_nxv2f32_nxv2i8 = call <vscale x 2 x i8> @llvm.vp.fptoui.nxv2i8.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2f64_nxv2i8 = call <vscale x 2 x i8> @llvm.vp.fptoui.nxv2i8.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2f32_nxv2i16 = call <vscale x 2 x i16> @llvm.vp.fptoui.nxv2i16.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2f64_nxv2i16 = call <vscale x 2 x i16> @llvm.vp.fptoui.nxv2i16.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2f32_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.fptoui.nxv2i32.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2f64_nxv2i32 = call <vscale x 2 x i32> @llvm.vp.fptoui.nxv2i32.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2f32_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.fptoui.nxv2i64.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2f64_nxv2i64 = call <vscale x 2 x i64> @llvm.vp.fptoui.nxv2i64.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2f32_nxv2i1 = call <vscale x 2 x i1> @llvm.vp.fptoui.nxv2i1.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2f64_nxv2i1 = call <vscale x 2 x i1> @llvm.vp.fptoui.nxv2i1.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)

  %nxv4f32_nxv4i8 = fptoui <vscale x 4 x float> undef to <vscale x 4 x i8>
  %nxv4f64_nxv4i8 = fptoui <vscale x 4 x double> undef to <vscale x 4 x i8>
  %nxv4f32_nxv4i16 = fptoui <vscale x 4 x float> undef to <vscale x 4 x i16>
  %nxv4f64_nxv4i16 = fptoui <vscale x 4 x double> undef to <vscale x 4 x i16>
  %nxv4f32_nxv4i32 = fptoui <vscale x 4 x float> undef to <vscale x 4 x i32>
  %nxv4f64_nxv4i32 = fptoui <vscale x 4 x double> undef to <vscale x 4 x i32>
  %nxv4f32_nxv4i64 = fptoui <vscale x 4 x float> undef to <vscale x 4 x i64>
  %nxv4f64_nxv4i64 = fptoui <vscale x 4 x double> undef to <vscale x 4 x i64>
  %nxv4f32_nxv4i1 = fptoui <vscale x 4 x float> undef to <vscale x 4 x i1>
  %nxv4f64_nxv4i1 = fptoui <vscale x 4 x double> undef to <vscale x 4 x i1>

  %vp_nxv4f32_nxv4i8 = call <vscale x 4 x i8> @llvm.vp.fptoui.nxv4i8.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4f64_nxv4i8 = call <vscale x 4 x i8> @llvm.vp.fptoui.nxv4i8.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4f32_nxv4i16 = call <vscale x 4 x i16> @llvm.vp.fptoui.nxv4i16.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4f64_nxv4i16 = call <vscale x 4 x i16> @llvm.vp.fptoui.nxv4i16.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4f32_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.fptoui.nxv4i32.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4f64_nxv4i32 = call <vscale x 4 x i32> @llvm.vp.fptoui.nxv4i32.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4f32_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.fptoui.nxv4i64.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4f64_nxv4i64 = call <vscale x 4 x i64> @llvm.vp.fptoui.nxv4i64.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4f32_nxv4i1 = call <vscale x 4 x i1> @llvm.vp.fptoui.nxv4i1.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4f64_nxv4i1 = call <vscale x 4 x i1> @llvm.vp.fptoui.nxv4i1.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)

  %nxv8f32_nxv8i8 = fptoui <vscale x 8 x float> undef to <vscale x 8 x i8>
  %nxv8f64_nxv8i8 = fptoui <vscale x 8 x double> undef to <vscale x 8 x i8>
  %nxv8f32_nxv8i16 = fptoui <vscale x 8 x float> undef to <vscale x 8 x i16>
  %nxv8f64_nxv8i16 = fptoui <vscale x 8 x double> undef to <vscale x 8 x i16>
  %nxv8f32_nxv8i32 = fptoui <vscale x 8 x float> undef to <vscale x 8 x i32>
  %nxv8f64_nxv8i32 = fptoui <vscale x 8 x double> undef to <vscale x 8 x i32>
  %nxv8f32_nxv8i64 = fptoui <vscale x 8 x float> undef to <vscale x 8 x i64>
  %nxv8f64_nxv8i64 = fptoui <vscale x 8 x double> undef to <vscale x 8 x i64>
  %nxv8f32_nxv8i1 = fptoui <vscale x 8 x float> undef to <vscale x 8 x i1>
  %nxv8f64_nxv8i1 = fptoui <vscale x 8 x double> undef to <vscale x 8 x i1>

  %vp_nxv8f32_nxv8i8 = call <vscale x 8 x i8> @llvm.vp.fptoui.nxv8i8.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8f64_nxv8i8 = call <vscale x 8 x i8> @llvm.vp.fptoui.nxv8i8.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8f32_nxv8i16 = call <vscale x 8 x i16> @llvm.vp.fptoui.nxv8i16.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8f64_nxv8i16 = call <vscale x 8 x i16> @llvm.vp.fptoui.nxv8i16.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8f32_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.fptoui.nxv8i32.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8f64_nxv8i32 = call <vscale x 8 x i32> @llvm.vp.fptoui.nxv8i32.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8f32_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.fptoui.nxv8i64.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8f64_nxv8i64 = call <vscale x 8 x i64> @llvm.vp.fptoui.nxv8i64.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8f32_nxv8i1 = call <vscale x 8 x i1> @llvm.vp.fptoui.nxv8i1.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8f64_nxv8i1 = call <vscale x 8 x i1> @llvm.vp.fptoui.nxv8i1.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)

  %nxv16f32_nxv16i8 = fptoui <vscale x 16 x float> undef to <vscale x 16 x i8>
  %nxv16f64_nxv16i8 = fptoui <vscale x 16 x double> undef to <vscale x 16 x i8>
  %nxv16f32_nxv16i16 = fptoui <vscale x 16 x float> undef to <vscale x 16 x i16>
  %nxv16f64_nxv16i16 = fptoui <vscale x 16 x double> undef to <vscale x 16 x i16>
  %nxv16f32_nxv16i32 = fptoui <vscale x 16 x float> undef to <vscale x 16 x i32>
  %nxv16f64_nxv16i32 = fptoui <vscale x 16 x double> undef to <vscale x 16 x i32>
  %nxv16f32_nxv16i64 = fptoui <vscale x 16 x float> undef to <vscale x 16 x i64>
  %nxv16f64_nxv16i64 = fptoui <vscale x 16 x double> undef to <vscale x 16 x i64>
  %nxv16f32_nxv16i1 = fptoui <vscale x 16 x float> undef to <vscale x 16 x i1>
  %nxv16f64_nxv16i1 = fptoui <vscale x 16 x double> undef to <vscale x 16 x i1>

  %vp_nxv16f32_nxv16i8 = call <vscale x 16 x i8> @llvm.vp.fptoui.nxv16i8.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16f64_nxv16i8 = call <vscale x 16 x i8> @llvm.vp.fptoui.nxv16i8.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16f32_nxv16i16 = call <vscale x 16 x i16> @llvm.vp.fptoui.nxv16i16.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16f64_nxv16i16 = call <vscale x 16 x i16> @llvm.vp.fptoui.nxv16i16.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16f32_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.fptoui.nxv16i32.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16f64_nxv16i32 = call <vscale x 16 x i32> @llvm.vp.fptoui.nxv16i32.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16f32_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.fptoui.nxv16i64.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16f64_nxv16i64 = call <vscale x 16 x i64> @llvm.vp.fptoui.nxv16i64.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16f32_nxv16i1 = call <vscale x 16 x i1> @llvm.vp.fptoui.nxv16i1.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16f64_nxv16i1 = call <vscale x 16 x i1> @llvm.vp.fptoui.nxv16i1.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)

  %nxv32f32_nxv32i8 = fptoui <vscale x 32 x float> undef to <vscale x 32 x i8>
  %nxv32f64_nxv32i8 = fptoui <vscale x 32 x double> undef to <vscale x 32 x i8>
  %nxv32f32_nxv32i16 = fptoui <vscale x 32 x float> undef to <vscale x 32 x i16>
  %nxv32f64_nxv32i16 = fptoui <vscale x 32 x double> undef to <vscale x 32 x i16>
  %nxv32f32_nxv32i32 = fptoui <vscale x 32 x float> undef to <vscale x 32 x i32>
  %nxv32f64_nxv32i32 = fptoui <vscale x 32 x double> undef to <vscale x 32 x i32>
  %nxv32f32_nxv32i64 = fptoui <vscale x 32 x float> undef to <vscale x 32 x i64>
  %nxv32f64_nxv32i64 = fptoui <vscale x 32 x double> undef to <vscale x 32 x i64>
  %nxv32f32_nxv32i1 = fptoui <vscale x 32 x float> undef to <vscale x 32 x i1>
  %nxv32f64_nxv32i1 = fptoui <vscale x 32 x double> undef to <vscale x 32 x i1>

  %vp_nxv32f32_nxv32i8 = call <vscale x 32 x i8> @llvm.vp.fptoui.nxv32i8.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32f64_nxv32i8 = call <vscale x 32 x i8> @llvm.vp.fptoui.nxv32i8.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32f32_nxv32i16 = call <vscale x 32 x i16> @llvm.vp.fptoui.nxv32i16.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32f64_nxv32i16 = call <vscale x 32 x i16> @llvm.vp.fptoui.nxv32i16.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32f32_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.fptoui.nxv32i32.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32f64_nxv32i32 = call <vscale x 32 x i32> @llvm.vp.fptoui.nxv32i32.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32f32_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.fptoui.nxv32i64.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32f64_nxv32i64 = call <vscale x 32 x i64> @llvm.vp.fptoui.nxv32i64.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32f32_nxv32i1 = call <vscale x 32 x i1> @llvm.vp.fptoui.nxv32i1.nxv32f32(<vscale x 32 x float> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32f64_nxv32i1 = call <vscale x 32 x i1> @llvm.vp.fptoui.nxv32i1.nxv32f64(<vscale x 32 x double> undef, <vscale x 32 x i1> undef, i32 undef)

  %nxv64f32_nxv64i8 = fptoui <vscale x 64 x float> undef to <vscale x 64 x i8>
  %nxv64f64_nxv64i8 = fptoui <vscale x 64 x double> undef to <vscale x 64 x i8>
  %nxv64f32_nxv64i16 = fptoui <vscale x 64 x float> undef to <vscale x 64 x i16>
  %nxv64f64_nxv64i16 = fptoui <vscale x 64 x double> undef to <vscale x 64 x i16>
  %nxv64f32_nxv64i32 = fptoui <vscale x 64 x float> undef to <vscale x 64 x i32>
  %nxv64f64_nxv64i32 = fptoui <vscale x 64 x double> undef to <vscale x 64 x i32>
  %nxv64f32_nxv64i64 = fptoui <vscale x 64 x float> undef to <vscale x 64 x i64>
  %nxv64f64_nxv64i64 = fptoui <vscale x 64 x double> undef to <vscale x 64 x i64>
  %nxv64f32_nxv64i1 = fptoui <vscale x 64 x float> undef to <vscale x 64 x i1>
  %nxv64f64_nxv64i1 = fptoui <vscale x 64 x double> undef to <vscale x 64 x i1>

  %vp_nxv64f32_nxv64i8 = call <vscale x 64 x i8> @llvm.vp.fptoui.nxv64i8.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64f64_nxv64i8 = call <vscale x 64 x i8> @llvm.vp.fptoui.nxv64i8.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64f32_nxv64i16 = call <vscale x 64 x i16> @llvm.vp.fptoui.nxv64i16.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64f64_nxv64i16 = call <vscale x 64 x i16> @llvm.vp.fptoui.nxv64i16.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64f32_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.fptoui.nxv64i32.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64f64_nxv64i32 = call <vscale x 64 x i32> @llvm.vp.fptoui.nxv64i32.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64f32_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.fptoui.nxv64i64.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64f64_nxv64i64 = call <vscale x 64 x i64> @llvm.vp.fptoui.nxv64i64.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64f32_nxv64i1 = call <vscale x 64 x i1> @llvm.vp.fptoui.nxv64i1.nxv64f32(<vscale x 64 x float> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64f64_nxv64i1 = call <vscale x 64 x i1> @llvm.vp.fptoui.nxv64i1.nxv64f64(<vscale x 64 x double> undef, <vscale x 64 x i1> undef, i32 undef)

  ret void
}

define void @sitofp() {
; ZVE64D-LABEL: 'sitofp'
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2i8_v2f32 = sitofp <2 x i8> undef to <2 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2i8_v2f64 = sitofp <2 x i8> undef to <2 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i16_v2f32 = sitofp <2 x i16> undef to <2 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2i16_v2f64 = sitofp <2 x i16> undef to <2 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i32_v2f32 = sitofp <2 x i32> undef to <2 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i32_v2f64 = sitofp <2 x i32> undef to <2 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i64_v2f32 = sitofp <2 x i64> undef to <2 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i64_v2f64 = sitofp <2 x i64> undef to <2 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2i1_v2f32 = sitofp <2 x i1> undef to <2 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2i1_v2f64 = sitofp <2 x i1> undef to <2 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2fi8_v2f32 = call <2 x float> @llvm.vp.sitofp.v2f32.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2fi8_v2f64 = call <2 x double> @llvm.vp.sitofp.v2f64.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2fi16_v2f32 = call <2 x float> @llvm.vp.sitofp.v2f32.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2fi16_v2f64 = call <2 x double> @llvm.vp.sitofp.v2f64.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2fi32_v2f32 = call <2 x float> @llvm.vp.sitofp.v2f32.v2i32(<2 x i32> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2fi32_v2f64 = call <2 x double> @llvm.vp.sitofp.v2f64.v2i32(<2 x i32> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2fi64_v2f32 = call <2 x float> @llvm.vp.sitofp.v2f32.v2i64(<2 x i64> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2fi64_v2f64 = call <2 x double> @llvm.vp.sitofp.v2f64.v2i64(<2 x i64> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2fi1_v2f32 = call <2 x float> @llvm.vp.sitofp.v2f32.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2fi1_v2f64 = call <2 x double> @llvm.vp.sitofp.v2f64.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i8_v4f32 = sitofp <4 x i8> undef to <4 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v4i8_v4f64 = sitofp <4 x i8> undef to <4 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4i16_v4f32 = sitofp <4 x i16> undef to <4 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v4i16_v4f64 = sitofp <4 x i16> undef to <4 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4i32_v4f32 = sitofp <4 x i32> undef to <4 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i32_v4f64 = sitofp <4 x i32> undef to <4 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4i64_v4f32 = sitofp <4 x i64> undef to <4 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i64_v4f64 = sitofp <4 x i64> undef to <4 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v4i1_v4f32 = sitofp <4 x i1> undef to <4 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v4i1_v4f64 = sitofp <4 x i1> undef to <4 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4fi8_v4f32 = call <4 x float> @llvm.vp.sitofp.v4f32.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v4fi8_v4f64 = call <4 x double> @llvm.vp.sitofp.v4f64.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v4fi16_v4f32 = call <4 x float> @llvm.vp.sitofp.v4f32.v4i16(<4 x i16> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v4fi16_v4f64 = call <4 x double> @llvm.vp.sitofp.v4f64.v4i16(<4 x i16> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v4fi32_v4f32 = call <4 x float> @llvm.vp.sitofp.v4f32.v4i32(<4 x i32> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4fi32_v4f64 = call <4 x double> @llvm.vp.sitofp.v4f64.v4i32(<4 x i32> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v4fi64_v4f32 = call <4 x float> @llvm.vp.sitofp.v4f32.v4i64(<4 x i64> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4fi64_v4f64 = call <4 x double> @llvm.vp.sitofp.v4f64.v4i64(<4 x i64> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v4fi1_v4f32 = call <4 x float> @llvm.vp.sitofp.v4f32.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v4fi1_v4f64 = call <4 x double> @llvm.vp.sitofp.v4f64.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v8i8_v8f32 = sitofp <8 x i8> undef to <8 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %v8i8_v8f64 = sitofp <8 x i8> undef to <8 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8i16_v8f32 = sitofp <8 x i16> undef to <8 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %v8i16_v8f64 = sitofp <8 x i16> undef to <8 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8i32_v8f32 = sitofp <8 x i32> undef to <8 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8i32_v8f64 = sitofp <8 x i32> undef to <8 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8i64_v8f32 = sitofp <8 x i64> undef to <8 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8i64_v8f64 = sitofp <8 x i64> undef to <8 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8i1_v8f32 = sitofp <8 x i1> undef to <8 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v8i1_v8f64 = sitofp <8 x i1> undef to <8 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v8fi8_v8f32 = call <8 x float> @llvm.vp.sitofp.v8f32.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_v8fi8_v8f64 = call <8 x double> @llvm.vp.sitofp.v8f64.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8fi16_v8f32 = call <8 x float> @llvm.vp.sitofp.v8f32.v8i16(<8 x i16> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_v8fi16_v8f64 = call <8 x double> @llvm.vp.sitofp.v8f64.v8i16(<8 x i16> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8fi32_v8f32 = call <8 x float> @llvm.vp.sitofp.v8f32.v8i32(<8 x i32> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v8fi32_v8f64 = call <8 x double> @llvm.vp.sitofp.v8f64.v8i32(<8 x i32> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8fi64_v8f32 = call <8 x float> @llvm.vp.sitofp.v8f32.v8i64(<8 x i64> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v8fi64_v8f64 = call <8 x double> @llvm.vp.sitofp.v8f64.v8i64(<8 x i64> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v8fi1_v8f32 = call <8 x float> @llvm.vp.sitofp.v8f32.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v8fi1_v8f64 = call <8 x double> @llvm.vp.sitofp.v8f64.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %v16i8_v16f32 = sitofp <16 x i8> undef to <16 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %v16i8_v16f64 = sitofp <16 x i8> undef to <16 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v16i16_v16f32 = sitofp <16 x i16> undef to <16 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %v16i16_v16f64 = sitofp <16 x i16> undef to <16 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v16i32_v16f32 = sitofp <16 x i32> undef to <16 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v16i32_v16f64 = sitofp <16 x i32> undef to <16 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v16i64_v16f32 = sitofp <16 x i64> undef to <16 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v16i64_v16f64 = sitofp <16 x i64> undef to <16 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v16i1_v16f32 = sitofp <16 x i1> undef to <16 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v16i1_v16f64 = sitofp <16 x i1> undef to <16 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_v16fi8_v16f32 = call <16 x float> @llvm.vp.sitofp.v16f32.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %vp_v16fi8_v16f64 = call <16 x double> @llvm.vp.sitofp.v16f64.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v16fi16_v16f32 = call <16 x float> @llvm.vp.sitofp.v16f32.v16i16(<16 x i16> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %vp_v16fi16_v16f64 = call <16 x double> @llvm.vp.sitofp.v16f64.v16i16(<16 x i16> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v16fi32_v16f32 = call <16 x float> @llvm.vp.sitofp.v16f32.v16i32(<16 x i32> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v16fi32_v16f64 = call <16 x double> @llvm.vp.sitofp.v16f64.v16i32(<16 x i32> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v16fi64_v16f32 = call <16 x float> @llvm.vp.sitofp.v16f32.v16i64(<16 x i64> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v16fi64_v16f64 = call <16 x double> @llvm.vp.sitofp.v16f64.v16i64(<16 x i64> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v16fi1_v16f32 = call <16 x float> @llvm.vp.sitofp.v16f32.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v16fi1_v16f64 = call <16 x double> @llvm.vp.sitofp.v16f64.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %v32i8_v32f32 = sitofp <32 x i8> undef to <32 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %v32i8_v32f64 = sitofp <32 x i8> undef to <32 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v32i16_v32f32 = sitofp <32 x i16> undef to <32 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %v32i16_v32f64 = sitofp <32 x i16> undef to <32 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v32i32_v32f32 = sitofp <32 x i32> undef to <32 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %v32i32_v32f64 = sitofp <32 x i32> undef to <32 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %v32i64_v32f32 = sitofp <32 x i64> undef to <32 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v32i64_v32f64 = sitofp <32 x i64> undef to <32 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v32i1_v32f32 = sitofp <32 x i1> undef to <32 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %v32i1_v32f64 = sitofp <32 x i1> undef to <32 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %vp_v32fi8_v32f32 = call <32 x float> @llvm.vp.sitofp.v32f32.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_v32fi8_v32f64 = call <32 x double> @llvm.vp.sitofp.v32f64.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v32fi16_v32f32 = call <32 x float> @llvm.vp.sitofp.v32f32.v32i16(<32 x i16> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_v32fi16_v32f64 = call <32 x double> @llvm.vp.sitofp.v32f64.v32i16(<32 x i16> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v32fi32_v32f32 = call <32 x float> @llvm.vp.sitofp.v32f32.v32i32(<32 x i32> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_v32fi32_v32f64 = call <32 x double> @llvm.vp.sitofp.v32f64.v32i32(<32 x i32> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %vp_v32fi64_v32f32 = call <32 x float> @llvm.vp.sitofp.v32f32.v32i64(<32 x i64> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v32fi64_v32f64 = call <32 x double> @llvm.vp.sitofp.v32f64.v32i64(<32 x i64> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v32fi1_v32f32 = call <32 x float> @llvm.vp.sitofp.v32f32.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %vp_v32fi1_v32f64 = call <32 x double> @llvm.vp.sitofp.v32f64.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %v64i8_v64f32 = sitofp <64 x i8> undef to <64 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %v64i8_v64f64 = sitofp <64 x i8> undef to <64 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %v64i16_v64f32 = sitofp <64 x i16> undef to <64 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %v64i16_v64f64 = sitofp <64 x i16> undef to <64 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v64i32_v64f32 = sitofp <64 x i32> undef to <64 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %v64i32_v64f64 = sitofp <64 x i32> undef to <64 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %v64i64_v64f32 = sitofp <64 x i64> undef to <64 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %v64i64_v64f64 = sitofp <64 x i64> undef to <64 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %v64i1_v64f32 = sitofp <64 x i1> undef to <64 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %v64i1_v64f64 = sitofp <64 x i1> undef to <64 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_v64fi8_v64f32 = call <64 x float> @llvm.vp.sitofp.v64f32.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %vp_v64fi8_v64f64 = call <64 x double> @llvm.vp.sitofp.v64f64.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_v64fi16_v64f32 = call <64 x float> @llvm.vp.sitofp.v64f32.v64i16(<64 x i16> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %vp_v64fi16_v64f64 = call <64 x double> @llvm.vp.sitofp.v64f64.v64i16(<64 x i16> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v64fi32_v64f32 = call <64 x float> @llvm.vp.sitofp.v64f32.v64i32(<64 x i32> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_v64fi32_v64f64 = call <64 x double> @llvm.vp.sitofp.v64f64.v64i32(<64 x i32> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %vp_v64fi64_v64f32 = call <64 x float> @llvm.vp.sitofp.v64f32.v64i64(<64 x i64> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %vp_v64fi64_v64f64 = call <64 x double> @llvm.vp.sitofp.v64f64.v64i64(<64 x i64> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %vp_v64fi1_v64f32 = call <64 x float> @llvm.vp.sitofp.v64f32.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %vp_v64fi1_v64f64 = call <64 x double> @llvm.vp.sitofp.v64f64.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %v128i8_v128f32 = sitofp <128 x i8> undef to <128 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 103 for instruction: %v128i8_v128f64 = sitofp <128 x i8> undef to <128 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %v128i16_v128f32 = sitofp <128 x i16> undef to <128 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 102 for instruction: %v128i16_v128f64 = sitofp <128 x i16> undef to <128 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %v128i32_v128f32 = sitofp <128 x i32> undef to <128 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %v128i32_v128f64 = sitofp <128 x i32> undef to <128 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %v128i64_v128f32 = sitofp <128 x i64> undef to <128 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %v128i64_v128f64 = sitofp <128 x i64> undef to <128 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %v128i1_v128f32 = sitofp <128 x i1> undef to <128 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 135 for instruction: %v128i1_v128f64 = sitofp <128 x i1> undef to <128 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %vp_v128fi8_v128f32 = call <128 x float> @llvm.vp.sitofp.v128f32.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 103 for instruction: %vp_v128fi8_v128f64 = call <128 x double> @llvm.vp.sitofp.v128f64.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_v128fi16_v128f32 = call <128 x float> @llvm.vp.sitofp.v128f32.v128i16(<128 x i16> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 102 for instruction: %vp_v128fi16_v128f64 = call <128 x double> @llvm.vp.sitofp.v128f64.v128i16(<128 x i16> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %vp_v128fi32_v128f32 = call <128 x float> @llvm.vp.sitofp.v128f32.v128i32(<128 x i32> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %vp_v128fi32_v128f64 = call <128 x double> @llvm.vp.sitofp.v128f64.v128i32(<128 x i32> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %vp_v128fi64_v128f32 = call <128 x float> @llvm.vp.sitofp.v128f32.v128i64(<128 x i64> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %vp_v128fi64_v128f64 = call <128 x double> @llvm.vp.sitofp.v128f64.v128i64(<128 x i64> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %vp_v128fi1_v128f32 = call <128 x float> @llvm.vp.sitofp.v128f32.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 135 for instruction: %vp_v128fi1_v128f64 = call <128 x double> @llvm.vp.sitofp.v128f64.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1i8_nxv1f32 = sitofp <vscale x 1 x i8> undef to <vscale x 1 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1i8_nxv1f64 = sitofp <vscale x 1 x i8> undef to <vscale x 1 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1i16_nxv1f32 = sitofp <vscale x 1 x i16> undef to <vscale x 1 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1i16_nxv1f64 = sitofp <vscale x 1 x i16> undef to <vscale x 1 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1i32_nxv1f32 = sitofp <vscale x 1 x i32> undef to <vscale x 1 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1i32_nxv1f64 = sitofp <vscale x 1 x i32> undef to <vscale x 1 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1i64_nxv1f32 = sitofp <vscale x 1 x i64> undef to <vscale x 1 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1i64_nxv1f64 = sitofp <vscale x 1 x i64> undef to <vscale x 1 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv1i1_nxv1f32 = sitofp <vscale x 1 x i1> undef to <vscale x 1 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv1i1_nxv1f64 = sitofp <vscale x 1 x i1> undef to <vscale x 1 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1fi8_nxv1f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1fi8_nxv1f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1fi16_nxv1f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1fi16_nxv1f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1fi32_nxv1f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1fi32_nxv1f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1fi64_nxv1f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1fi64_nxv1f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv1fi1_nxv1f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv1fi1_nxv1f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2i8_nxv2f32 = sitofp <vscale x 2 x i8> undef to <vscale x 2 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv2i8_nxv2f64 = sitofp <vscale x 2 x i8> undef to <vscale x 2 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv2i16_nxv2f32 = sitofp <vscale x 2 x i16> undef to <vscale x 2 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv2i16_nxv2f64 = sitofp <vscale x 2 x i16> undef to <vscale x 2 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv2i32_nxv2f32 = sitofp <vscale x 2 x i32> undef to <vscale x 2 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2i32_nxv2f64 = sitofp <vscale x 2 x i32> undef to <vscale x 2 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv2i64_nxv2f32 = sitofp <vscale x 2 x i64> undef to <vscale x 2 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2i64_nxv2f64 = sitofp <vscale x 2 x i64> undef to <vscale x 2 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv2i1_nxv2f32 = sitofp <vscale x 2 x i1> undef to <vscale x 2 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv2i1_nxv2f64 = sitofp <vscale x 2 x i1> undef to <vscale x 2 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2fi8_nxv2f32 = call <vscale x 2 x float> @llvm.vp.sitofp.nxv2f32.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv2fi8_nxv2f64 = call <vscale x 2 x double> @llvm.vp.sitofp.nxv2f64.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv2fi16_nxv2f32 = call <vscale x 2 x float> @llvm.vp.sitofp.nxv2f32.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv2fi16_nxv2f64 = call <vscale x 2 x double> @llvm.vp.sitofp.nxv2f64.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv2fi32_nxv2f32 = call <vscale x 2 x float> @llvm.vp.sitofp.nxv2f32.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2fi32_nxv2f64 = call <vscale x 2 x double> @llvm.vp.sitofp.nxv2f64.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv2fi64_nxv2f32 = call <vscale x 2 x float> @llvm.vp.sitofp.nxv2f32.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2fi64_nxv2f64 = call <vscale x 2 x double> @llvm.vp.sitofp.nxv2f64.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv2fi1_nxv2f32 = call <vscale x 2 x float> @llvm.vp.sitofp.nxv2f32.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv2fi1_nxv2f64 = call <vscale x 2 x double> @llvm.vp.sitofp.nxv2f64.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv4i8_nxv4f32 = sitofp <vscale x 4 x i8> undef to <vscale x 4 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %nxv4i8_nxv4f64 = sitofp <vscale x 4 x i8> undef to <vscale x 4 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4i16_nxv4f32 = sitofp <vscale x 4 x i16> undef to <vscale x 4 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %nxv4i16_nxv4f64 = sitofp <vscale x 4 x i16> undef to <vscale x 4 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4i32_nxv4f32 = sitofp <vscale x 4 x i32> undef to <vscale x 4 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv4i32_nxv4f64 = sitofp <vscale x 4 x i32> undef to <vscale x 4 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4i64_nxv4f32 = sitofp <vscale x 4 x i64> undef to <vscale x 4 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv4i64_nxv4f64 = sitofp <vscale x 4 x i64> undef to <vscale x 4 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv4i1_nxv4f32 = sitofp <vscale x 4 x i1> undef to <vscale x 4 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv4i1_nxv4f64 = sitofp <vscale x 4 x i1> undef to <vscale x 4 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv4fi8_nxv4f32 = call <vscale x 4 x float> @llvm.vp.sitofp.nxv4f32.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_nxv4fi8_nxv4f64 = call <vscale x 4 x double> @llvm.vp.sitofp.nxv4f64.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4fi16_nxv4f32 = call <vscale x 4 x float> @llvm.vp.sitofp.nxv4f32.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_nxv4fi16_nxv4f64 = call <vscale x 4 x double> @llvm.vp.sitofp.nxv4f64.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4fi32_nxv4f32 = call <vscale x 4 x float> @llvm.vp.sitofp.nxv4f32.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv4fi32_nxv4f64 = call <vscale x 4 x double> @llvm.vp.sitofp.nxv4f64.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4fi64_nxv4f32 = call <vscale x 4 x float> @llvm.vp.sitofp.nxv4f32.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv4fi64_nxv4f64 = call <vscale x 4 x double> @llvm.vp.sitofp.nxv4f64.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv4fi1_nxv4f32 = call <vscale x 4 x float> @llvm.vp.sitofp.nxv4f32.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv4fi1_nxv4f64 = call <vscale x 4 x double> @llvm.vp.sitofp.nxv4f64.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %nxv8i8_nxv8f32 = sitofp <vscale x 8 x i8> undef to <vscale x 8 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %nxv8i8_nxv8f64 = sitofp <vscale x 8 x i8> undef to <vscale x 8 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv8i16_nxv8f32 = sitofp <vscale x 8 x i16> undef to <vscale x 8 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %nxv8i16_nxv8f64 = sitofp <vscale x 8 x i16> undef to <vscale x 8 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv8i32_nxv8f32 = sitofp <vscale x 8 x i32> undef to <vscale x 8 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv8i32_nxv8f64 = sitofp <vscale x 8 x i32> undef to <vscale x 8 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv8i64_nxv8f32 = sitofp <vscale x 8 x i64> undef to <vscale x 8 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv8i64_nxv8f64 = sitofp <vscale x 8 x i64> undef to <vscale x 8 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv8i1_nxv8f32 = sitofp <vscale x 8 x i1> undef to <vscale x 8 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %nxv8i1_nxv8f64 = sitofp <vscale x 8 x i1> undef to <vscale x 8 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv8fi8_nxv8f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv8fi8_nxv8f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv8fi16_nxv8f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv8fi16_nxv8f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv8fi32_nxv8f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv8fi32_nxv8f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv8fi64_nxv8f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv8fi64_nxv8f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv8fi1_nxv8f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv8fi1_nxv8f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %nxv16i8_nxv16f32 = sitofp <vscale x 16 x i8> undef to <vscale x 16 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %nxv16i8_nxv16f64 = sitofp <vscale x 16 x i8> undef to <vscale x 16 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv16i16_nxv16f32 = sitofp <vscale x 16 x i16> undef to <vscale x 16 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %nxv16i16_nxv16f64 = sitofp <vscale x 16 x i16> undef to <vscale x 16 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv16i32_nxv16f32 = sitofp <vscale x 16 x i32> undef to <vscale x 16 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %nxv16i32_nxv16f64 = sitofp <vscale x 16 x i32> undef to <vscale x 16 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %nxv16i64_nxv16f32 = sitofp <vscale x 16 x i64> undef to <vscale x 16 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv16i64_nxv16f64 = sitofp <vscale x 16 x i64> undef to <vscale x 16 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %nxv16i1_nxv16f32 = sitofp <vscale x 16 x i1> undef to <vscale x 16 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %nxv16i1_nxv16f64 = sitofp <vscale x 16 x i1> undef to <vscale x 16 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %vp_nxv16fi8_nxv16f32 = call <vscale x 16 x float> @llvm.vp.sitofp.nxv16f32.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_nxv16fi8_nxv16f64 = call <vscale x 16 x double> @llvm.vp.sitofp.nxv16f64.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv16fi16_nxv16f32 = call <vscale x 16 x float> @llvm.vp.sitofp.nxv16f32.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_nxv16fi16_nxv16f64 = call <vscale x 16 x double> @llvm.vp.sitofp.nxv16f64.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv16fi32_nxv16f32 = call <vscale x 16 x float> @llvm.vp.sitofp.nxv16f32.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_nxv16fi32_nxv16f64 = call <vscale x 16 x double> @llvm.vp.sitofp.nxv16f64.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %vp_nxv16fi64_nxv16f32 = call <vscale x 16 x float> @llvm.vp.sitofp.nxv16f32.nxv16i64(<vscale x 16 x i64> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv16fi64_nxv16f64 = call <vscale x 16 x double> @llvm.vp.sitofp.nxv16f64.nxv16i64(<vscale x 16 x i64> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_nxv16fi1_nxv16f32 = call <vscale x 16 x float> @llvm.vp.sitofp.nxv16f32.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %vp_nxv16fi1_nxv16f64 = call <vscale x 16 x double> @llvm.vp.sitofp.nxv16f64.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %nxv32i8_nxv32f32 = sitofp <vscale x 32 x i8> undef to <vscale x 32 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %nxv32i8_nxv32f64 = sitofp <vscale x 32 x i8> undef to <vscale x 32 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %nxv32i16_nxv32f32 = sitofp <vscale x 32 x i16> undef to <vscale x 32 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %nxv32i16_nxv32f64 = sitofp <vscale x 32 x i16> undef to <vscale x 32 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv32i32_nxv32f32 = sitofp <vscale x 32 x i32> undef to <vscale x 32 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %nxv32i32_nxv32f64 = sitofp <vscale x 32 x i32> undef to <vscale x 32 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %nxv32i64_nxv32f32 = sitofp <vscale x 32 x i64> undef to <vscale x 32 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv32i64_nxv32f64 = sitofp <vscale x 32 x i64> undef to <vscale x 32 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %nxv32i1_nxv32f32 = sitofp <vscale x 32 x i1> undef to <vscale x 32 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %nxv32i1_nxv32f64 = sitofp <vscale x 32 x i1> undef to <vscale x 32 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_nxv32fi8_nxv32f32 = call <vscale x 32 x float> @llvm.vp.sitofp.nxv32f32.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %vp_nxv32fi8_nxv32f64 = call <vscale x 32 x double> @llvm.vp.sitofp.nxv32f64.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_nxv32fi16_nxv32f32 = call <vscale x 32 x float> @llvm.vp.sitofp.nxv32f32.nxv32i16(<vscale x 32 x i16> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %vp_nxv32fi16_nxv32f64 = call <vscale x 32 x double> @llvm.vp.sitofp.nxv32f64.nxv32i16(<vscale x 32 x i16> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv32fi32_nxv32f32 = call <vscale x 32 x float> @llvm.vp.sitofp.nxv32f32.nxv32i32(<vscale x 32 x i32> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_nxv32fi32_nxv32f64 = call <vscale x 32 x double> @llvm.vp.sitofp.nxv32f64.nxv32i32(<vscale x 32 x i32> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %vp_nxv32fi64_nxv32f32 = call <vscale x 32 x float> @llvm.vp.sitofp.nxv32f32.nxv32i64(<vscale x 32 x i64> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv32fi64_nxv32f64 = call <vscale x 32 x double> @llvm.vp.sitofp.nxv32f64.nxv32i64(<vscale x 32 x i64> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %vp_nxv32fi1_nxv32f32 = call <vscale x 32 x float> @llvm.vp.sitofp.nxv32f32.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %vp_nxv32fi1_nxv32f64 = call <vscale x 32 x double> @llvm.vp.sitofp.nxv32f64.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %nxv64i8_nxv64f32 = sitofp <vscale x 64 x i8> undef to <vscale x 64 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 103 for instruction: %nxv64i8_nxv64f64 = sitofp <vscale x 64 x i8> undef to <vscale x 64 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %nxv64i16_nxv64f32 = sitofp <vscale x 64 x i16> undef to <vscale x 64 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 102 for instruction: %nxv64i16_nxv64f64 = sitofp <vscale x 64 x i16> undef to <vscale x 64 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv64i32_nxv64f32 = sitofp <vscale x 64 x i32> undef to <vscale x 64 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %nxv64i32_nxv64f64 = sitofp <vscale x 64 x i32> undef to <vscale x 64 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %nxv64i64_nxv64f32 = sitofp <vscale x 64 x i64> undef to <vscale x 64 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv64i64_nxv64f64 = sitofp <vscale x 64 x i64> undef to <vscale x 64 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %nxv64i1_nxv64f32 = sitofp <vscale x 64 x i1> undef to <vscale x 64 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 135 for instruction: %nxv64i1_nxv64f64 = sitofp <vscale x 64 x i1> undef to <vscale x 64 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %vp_nxv64fi8_nxv64f32 = call <vscale x 64 x float> @llvm.vp.sitofp.nxv64f32.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 103 for instruction: %vp_nxv64fi8_nxv64f64 = call <vscale x 64 x double> @llvm.vp.sitofp.nxv64f64.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_nxv64fi16_nxv64f32 = call <vscale x 64 x float> @llvm.vp.sitofp.nxv64f32.nxv64i16(<vscale x 64 x i16> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 102 for instruction: %vp_nxv64fi16_nxv64f64 = call <vscale x 64 x double> @llvm.vp.sitofp.nxv64f64.nxv64i16(<vscale x 64 x i16> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv64fi32_nxv64f32 = call <vscale x 64 x float> @llvm.vp.sitofp.nxv64f32.nxv64i32(<vscale x 64 x i32> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %vp_nxv64fi32_nxv64f64 = call <vscale x 64 x double> @llvm.vp.sitofp.nxv64f64.nxv64i32(<vscale x 64 x i32> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %vp_nxv64fi64_nxv64f32 = call <vscale x 64 x float> @llvm.vp.sitofp.nxv64f32.nxv64i64(<vscale x 64 x i64> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv64fi64_nxv64f64 = call <vscale x 64 x double> @llvm.vp.sitofp.nxv64f64.nxv64i64(<vscale x 64 x i64> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %vp_nxv64fi1_nxv64f32 = call <vscale x 64 x float> @llvm.vp.sitofp.nxv64f32.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 135 for instruction: %vp_nxv64fi1_nxv64f64 = call <vscale x 64 x double> @llvm.vp.sitofp.nxv64f64.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; ZVE64X-LABEL: 'sitofp'
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2i8_v2f32 = sitofp <2 x i8> undef to <2 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2i8_v2f64 = sitofp <2 x i8> undef to <2 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2i16_v2f32 = sitofp <2 x i16> undef to <2 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2i16_v2f64 = sitofp <2 x i16> undef to <2 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2i32_v2f32 = sitofp <2 x i32> undef to <2 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2i32_v2f64 = sitofp <2 x i32> undef to <2 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2i64_v2f32 = sitofp <2 x i64> undef to <2 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2i64_v2f64 = sitofp <2 x i64> undef to <2 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2i1_v2f32 = sitofp <2 x i1> undef to <2 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2i1_v2f64 = sitofp <2 x i1> undef to <2 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2fi8_v2f32 = call <2 x float> @llvm.vp.sitofp.v2f32.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2fi8_v2f64 = call <2 x double> @llvm.vp.sitofp.v2f64.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2fi16_v2f32 = call <2 x float> @llvm.vp.sitofp.v2f32.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2fi16_v2f64 = call <2 x double> @llvm.vp.sitofp.v2f64.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2fi32_v2f32 = call <2 x float> @llvm.vp.sitofp.v2f32.v2i32(<2 x i32> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2fi32_v2f64 = call <2 x double> @llvm.vp.sitofp.v2f64.v2i32(<2 x i32> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2fi64_v2f32 = call <2 x float> @llvm.vp.sitofp.v2f32.v2i64(<2 x i64> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2fi64_v2f64 = call <2 x double> @llvm.vp.sitofp.v2f64.v2i64(<2 x i64> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2fi1_v2f32 = call <2 x float> @llvm.vp.sitofp.v2f32.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2fi1_v2f64 = call <2 x double> @llvm.vp.sitofp.v2f64.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4i8_v4f32 = sitofp <4 x i8> undef to <4 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4i8_v4f64 = sitofp <4 x i8> undef to <4 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4i16_v4f32 = sitofp <4 x i16> undef to <4 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4i16_v4f64 = sitofp <4 x i16> undef to <4 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4i32_v4f32 = sitofp <4 x i32> undef to <4 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4i32_v4f64 = sitofp <4 x i32> undef to <4 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4i64_v4f32 = sitofp <4 x i64> undef to <4 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4i64_v4f64 = sitofp <4 x i64> undef to <4 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4i1_v4f32 = sitofp <4 x i1> undef to <4 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4i1_v4f64 = sitofp <4 x i1> undef to <4 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v4fi8_v4f32 = call <4 x float> @llvm.vp.sitofp.v4f32.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v4fi8_v4f64 = call <4 x double> @llvm.vp.sitofp.v4f64.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v4fi16_v4f32 = call <4 x float> @llvm.vp.sitofp.v4f32.v4i16(<4 x i16> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v4fi16_v4f64 = call <4 x double> @llvm.vp.sitofp.v4f64.v4i16(<4 x i16> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v4fi32_v4f32 = call <4 x float> @llvm.vp.sitofp.v4f32.v4i32(<4 x i32> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v4fi32_v4f64 = call <4 x double> @llvm.vp.sitofp.v4f64.v4i32(<4 x i32> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v4fi64_v4f32 = call <4 x float> @llvm.vp.sitofp.v4f32.v4i64(<4 x i64> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v4fi64_v4f64 = call <4 x double> @llvm.vp.sitofp.v4f64.v4i64(<4 x i64> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v4fi1_v4f32 = call <4 x float> @llvm.vp.sitofp.v4f32.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v4fi1_v4f64 = call <4 x double> @llvm.vp.sitofp.v4f64.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v8i8_v8f32 = sitofp <8 x i8> undef to <8 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v8i8_v8f64 = sitofp <8 x i8> undef to <8 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v8i16_v8f32 = sitofp <8 x i16> undef to <8 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v8i16_v8f64 = sitofp <8 x i16> undef to <8 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v8i32_v8f32 = sitofp <8 x i32> undef to <8 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v8i32_v8f64 = sitofp <8 x i32> undef to <8 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v8i64_v8f32 = sitofp <8 x i64> undef to <8 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v8i64_v8f64 = sitofp <8 x i64> undef to <8 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v8i1_v8f32 = sitofp <8 x i1> undef to <8 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v8i1_v8f64 = sitofp <8 x i1> undef to <8 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v8fi8_v8f32 = call <8 x float> @llvm.vp.sitofp.v8f32.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v8fi8_v8f64 = call <8 x double> @llvm.vp.sitofp.v8f64.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v8fi16_v8f32 = call <8 x float> @llvm.vp.sitofp.v8f32.v8i16(<8 x i16> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v8fi16_v8f64 = call <8 x double> @llvm.vp.sitofp.v8f64.v8i16(<8 x i16> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v8fi32_v8f32 = call <8 x float> @llvm.vp.sitofp.v8f32.v8i32(<8 x i32> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v8fi32_v8f64 = call <8 x double> @llvm.vp.sitofp.v8f64.v8i32(<8 x i32> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v8fi64_v8f32 = call <8 x float> @llvm.vp.sitofp.v8f32.v8i64(<8 x i64> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v8fi64_v8f64 = call <8 x double> @llvm.vp.sitofp.v8f64.v8i64(<8 x i64> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v8fi1_v8f32 = call <8 x float> @llvm.vp.sitofp.v8f32.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v8fi1_v8f64 = call <8 x double> @llvm.vp.sitofp.v8f64.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v16i8_v16f32 = sitofp <16 x i8> undef to <16 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v16i8_v16f64 = sitofp <16 x i8> undef to <16 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v16i16_v16f32 = sitofp <16 x i16> undef to <16 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v16i16_v16f64 = sitofp <16 x i16> undef to <16 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v16i32_v16f32 = sitofp <16 x i32> undef to <16 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v16i32_v16f64 = sitofp <16 x i32> undef to <16 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v16i64_v16f32 = sitofp <16 x i64> undef to <16 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v16i64_v16f64 = sitofp <16 x i64> undef to <16 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v16i1_v16f32 = sitofp <16 x i1> undef to <16 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v16i1_v16f64 = sitofp <16 x i1> undef to <16 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v16fi8_v16f32 = call <16 x float> @llvm.vp.sitofp.v16f32.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v16fi8_v16f64 = call <16 x double> @llvm.vp.sitofp.v16f64.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v16fi16_v16f32 = call <16 x float> @llvm.vp.sitofp.v16f32.v16i16(<16 x i16> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v16fi16_v16f64 = call <16 x double> @llvm.vp.sitofp.v16f64.v16i16(<16 x i16> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v16fi32_v16f32 = call <16 x float> @llvm.vp.sitofp.v16f32.v16i32(<16 x i32> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v16fi32_v16f64 = call <16 x double> @llvm.vp.sitofp.v16f64.v16i32(<16 x i32> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v16fi64_v16f32 = call <16 x float> @llvm.vp.sitofp.v16f32.v16i64(<16 x i64> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v16fi64_v16f64 = call <16 x double> @llvm.vp.sitofp.v16f64.v16i64(<16 x i64> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v16fi1_v16f32 = call <16 x float> @llvm.vp.sitofp.v16f32.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v16fi1_v16f64 = call <16 x double> @llvm.vp.sitofp.v16f64.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v32i8_v32f32 = sitofp <32 x i8> undef to <32 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v32i8_v32f64 = sitofp <32 x i8> undef to <32 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v32i16_v32f32 = sitofp <32 x i16> undef to <32 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v32i16_v32f64 = sitofp <32 x i16> undef to <32 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v32i32_v32f32 = sitofp <32 x i32> undef to <32 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v32i32_v32f64 = sitofp <32 x i32> undef to <32 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 62 for instruction: %v32i64_v32f32 = sitofp <32 x i64> undef to <32 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 62 for instruction: %v32i64_v32f64 = sitofp <32 x i64> undef to <32 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v32i1_v32f32 = sitofp <32 x i1> undef to <32 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v32i1_v32f64 = sitofp <32 x i1> undef to <32 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v32fi8_v32f32 = call <32 x float> @llvm.vp.sitofp.v32f32.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v32fi8_v32f64 = call <32 x double> @llvm.vp.sitofp.v32f64.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v32fi16_v32f32 = call <32 x float> @llvm.vp.sitofp.v32f32.v32i16(<32 x i16> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v32fi16_v32f64 = call <32 x double> @llvm.vp.sitofp.v32f64.v32i16(<32 x i16> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v32fi32_v32f32 = call <32 x float> @llvm.vp.sitofp.v32f32.v32i32(<32 x i32> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v32fi32_v32f64 = call <32 x double> @llvm.vp.sitofp.v32f64.v32i32(<32 x i32> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 62 for instruction: %vp_v32fi64_v32f32 = call <32 x float> @llvm.vp.sitofp.v32f32.v32i64(<32 x i64> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 62 for instruction: %vp_v32fi64_v32f64 = call <32 x double> @llvm.vp.sitofp.v32f64.v32i64(<32 x i64> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v32fi1_v32f32 = call <32 x float> @llvm.vp.sitofp.v32f32.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v32fi1_v32f64 = call <32 x double> @llvm.vp.sitofp.v32f64.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %v64i8_v64f32 = sitofp <64 x i8> undef to <64 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %v64i8_v64f64 = sitofp <64 x i8> undef to <64 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %v64i16_v64f32 = sitofp <64 x i16> undef to <64 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %v64i16_v64f64 = sitofp <64 x i16> undef to <64 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 126 for instruction: %v64i32_v64f32 = sitofp <64 x i32> undef to <64 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 126 for instruction: %v64i32_v64f64 = sitofp <64 x i32> undef to <64 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 124 for instruction: %v64i64_v64f32 = sitofp <64 x i64> undef to <64 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 124 for instruction: %v64i64_v64f64 = sitofp <64 x i64> undef to <64 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %v64i1_v64f32 = sitofp <64 x i1> undef to <64 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %v64i1_v64f64 = sitofp <64 x i1> undef to <64 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %vp_v64fi8_v64f32 = call <64 x float> @llvm.vp.sitofp.v64f32.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %vp_v64fi8_v64f64 = call <64 x double> @llvm.vp.sitofp.v64f64.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %vp_v64fi16_v64f32 = call <64 x float> @llvm.vp.sitofp.v64f32.v64i16(<64 x i16> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %vp_v64fi16_v64f64 = call <64 x double> @llvm.vp.sitofp.v64f64.v64i16(<64 x i16> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 126 for instruction: %vp_v64fi32_v64f32 = call <64 x float> @llvm.vp.sitofp.v64f32.v64i32(<64 x i32> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 126 for instruction: %vp_v64fi32_v64f64 = call <64 x double> @llvm.vp.sitofp.v64f64.v64i32(<64 x i32> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 124 for instruction: %vp_v64fi64_v64f32 = call <64 x float> @llvm.vp.sitofp.v64f32.v64i64(<64 x i64> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 124 for instruction: %vp_v64fi64_v64f64 = call <64 x double> @llvm.vp.sitofp.v64f64.v64i64(<64 x i64> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %vp_v64fi1_v64f32 = call <64 x float> @llvm.vp.sitofp.v64f32.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %vp_v64fi1_v64f64 = call <64 x double> @llvm.vp.sitofp.v64f64.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %v128i8_v128f32 = sitofp <128 x i8> undef to <128 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %v128i8_v128f64 = sitofp <128 x i8> undef to <128 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 254 for instruction: %v128i16_v128f32 = sitofp <128 x i16> undef to <128 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 254 for instruction: %v128i16_v128f64 = sitofp <128 x i16> undef to <128 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 252 for instruction: %v128i32_v128f32 = sitofp <128 x i32> undef to <128 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 252 for instruction: %v128i32_v128f64 = sitofp <128 x i32> undef to <128 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 248 for instruction: %v128i64_v128f32 = sitofp <128 x i64> undef to <128 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 248 for instruction: %v128i64_v128f64 = sitofp <128 x i64> undef to <128 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %v128i1_v128f32 = sitofp <128 x i1> undef to <128 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %v128i1_v128f64 = sitofp <128 x i1> undef to <128 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %vp_v128fi8_v128f32 = call <128 x float> @llvm.vp.sitofp.v128f32.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %vp_v128fi8_v128f64 = call <128 x double> @llvm.vp.sitofp.v128f64.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 254 for instruction: %vp_v128fi16_v128f32 = call <128 x float> @llvm.vp.sitofp.v128f32.v128i16(<128 x i16> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 254 for instruction: %vp_v128fi16_v128f64 = call <128 x double> @llvm.vp.sitofp.v128f64.v128i16(<128 x i16> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 252 for instruction: %vp_v128fi32_v128f32 = call <128 x float> @llvm.vp.sitofp.v128f32.v128i32(<128 x i32> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 252 for instruction: %vp_v128fi32_v128f64 = call <128 x double> @llvm.vp.sitofp.v128f64.v128i32(<128 x i32> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 248 for instruction: %vp_v128fi64_v128f32 = call <128 x float> @llvm.vp.sitofp.v128f32.v128i64(<128 x i64> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 248 for instruction: %vp_v128fi64_v128f64 = call <128 x double> @llvm.vp.sitofp.v128f64.v128i64(<128 x i64> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %vp_v128fi1_v128f32 = call <128 x float> @llvm.vp.sitofp.v128f32.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %vp_v128fi1_v128f64 = call <128 x double> @llvm.vp.sitofp.v128f64.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1i8_nxv1f32 = sitofp <vscale x 1 x i8> undef to <vscale x 1 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1i8_nxv1f64 = sitofp <vscale x 1 x i8> undef to <vscale x 1 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1i16_nxv1f32 = sitofp <vscale x 1 x i16> undef to <vscale x 1 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1i16_nxv1f64 = sitofp <vscale x 1 x i16> undef to <vscale x 1 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1i32_nxv1f32 = sitofp <vscale x 1 x i32> undef to <vscale x 1 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1i32_nxv1f64 = sitofp <vscale x 1 x i32> undef to <vscale x 1 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1i64_nxv1f32 = sitofp <vscale x 1 x i64> undef to <vscale x 1 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1i64_nxv1f64 = sitofp <vscale x 1 x i64> undef to <vscale x 1 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1i1_nxv1f32 = sitofp <vscale x 1 x i1> undef to <vscale x 1 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1i1_nxv1f64 = sitofp <vscale x 1 x i1> undef to <vscale x 1 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1fi8_nxv1f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1fi8_nxv1f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1fi16_nxv1f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1fi16_nxv1f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1fi32_nxv1f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1fi32_nxv1f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1fi64_nxv1f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1fi64_nxv1f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1fi1_nxv1f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1fi1_nxv1f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2i8_nxv2f32 = sitofp <vscale x 2 x i8> undef to <vscale x 2 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2i8_nxv2f64 = sitofp <vscale x 2 x i8> undef to <vscale x 2 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2i16_nxv2f32 = sitofp <vscale x 2 x i16> undef to <vscale x 2 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2i16_nxv2f64 = sitofp <vscale x 2 x i16> undef to <vscale x 2 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2i32_nxv2f32 = sitofp <vscale x 2 x i32> undef to <vscale x 2 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2i32_nxv2f64 = sitofp <vscale x 2 x i32> undef to <vscale x 2 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2i64_nxv2f32 = sitofp <vscale x 2 x i64> undef to <vscale x 2 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2i64_nxv2f64 = sitofp <vscale x 2 x i64> undef to <vscale x 2 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2i1_nxv2f32 = sitofp <vscale x 2 x i1> undef to <vscale x 2 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2i1_nxv2f64 = sitofp <vscale x 2 x i1> undef to <vscale x 2 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2fi8_nxv2f32 = call <vscale x 2 x float> @llvm.vp.sitofp.nxv2f32.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2fi8_nxv2f64 = call <vscale x 2 x double> @llvm.vp.sitofp.nxv2f64.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2fi16_nxv2f32 = call <vscale x 2 x float> @llvm.vp.sitofp.nxv2f32.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2fi16_nxv2f64 = call <vscale x 2 x double> @llvm.vp.sitofp.nxv2f64.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2fi32_nxv2f32 = call <vscale x 2 x float> @llvm.vp.sitofp.nxv2f32.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2fi32_nxv2f64 = call <vscale x 2 x double> @llvm.vp.sitofp.nxv2f64.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2fi64_nxv2f32 = call <vscale x 2 x float> @llvm.vp.sitofp.nxv2f32.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2fi64_nxv2f64 = call <vscale x 2 x double> @llvm.vp.sitofp.nxv2f64.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2fi1_nxv2f32 = call <vscale x 2 x float> @llvm.vp.sitofp.nxv2f32.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2fi1_nxv2f64 = call <vscale x 2 x double> @llvm.vp.sitofp.nxv2f64.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4i8_nxv4f32 = sitofp <vscale x 4 x i8> undef to <vscale x 4 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4i8_nxv4f64 = sitofp <vscale x 4 x i8> undef to <vscale x 4 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4i16_nxv4f32 = sitofp <vscale x 4 x i16> undef to <vscale x 4 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4i16_nxv4f64 = sitofp <vscale x 4 x i16> undef to <vscale x 4 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4i32_nxv4f32 = sitofp <vscale x 4 x i32> undef to <vscale x 4 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4i32_nxv4f64 = sitofp <vscale x 4 x i32> undef to <vscale x 4 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4i64_nxv4f32 = sitofp <vscale x 4 x i64> undef to <vscale x 4 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4i64_nxv4f64 = sitofp <vscale x 4 x i64> undef to <vscale x 4 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4i1_nxv4f32 = sitofp <vscale x 4 x i1> undef to <vscale x 4 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4i1_nxv4f64 = sitofp <vscale x 4 x i1> undef to <vscale x 4 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4fi8_nxv4f32 = call <vscale x 4 x float> @llvm.vp.sitofp.nxv4f32.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4fi8_nxv4f64 = call <vscale x 4 x double> @llvm.vp.sitofp.nxv4f64.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4fi16_nxv4f32 = call <vscale x 4 x float> @llvm.vp.sitofp.nxv4f32.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4fi16_nxv4f64 = call <vscale x 4 x double> @llvm.vp.sitofp.nxv4f64.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4fi32_nxv4f32 = call <vscale x 4 x float> @llvm.vp.sitofp.nxv4f32.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4fi32_nxv4f64 = call <vscale x 4 x double> @llvm.vp.sitofp.nxv4f64.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4fi64_nxv4f32 = call <vscale x 4 x float> @llvm.vp.sitofp.nxv4f32.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4fi64_nxv4f64 = call <vscale x 4 x double> @llvm.vp.sitofp.nxv4f64.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4fi1_nxv4f32 = call <vscale x 4 x float> @llvm.vp.sitofp.nxv4f32.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4fi1_nxv4f64 = call <vscale x 4 x double> @llvm.vp.sitofp.nxv4f64.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8i8_nxv8f32 = sitofp <vscale x 8 x i8> undef to <vscale x 8 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8i8_nxv8f64 = sitofp <vscale x 8 x i8> undef to <vscale x 8 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8i16_nxv8f32 = sitofp <vscale x 8 x i16> undef to <vscale x 8 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8i16_nxv8f64 = sitofp <vscale x 8 x i16> undef to <vscale x 8 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8i32_nxv8f32 = sitofp <vscale x 8 x i32> undef to <vscale x 8 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8i32_nxv8f64 = sitofp <vscale x 8 x i32> undef to <vscale x 8 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8i64_nxv8f32 = sitofp <vscale x 8 x i64> undef to <vscale x 8 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8i64_nxv8f64 = sitofp <vscale x 8 x i64> undef to <vscale x 8 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8i1_nxv8f32 = sitofp <vscale x 8 x i1> undef to <vscale x 8 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8i1_nxv8f64 = sitofp <vscale x 8 x i1> undef to <vscale x 8 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8fi8_nxv8f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8fi8_nxv8f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8fi16_nxv8f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8fi16_nxv8f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8fi32_nxv8f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8fi32_nxv8f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8fi64_nxv8f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8fi64_nxv8f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8fi1_nxv8f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8fi1_nxv8f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16i8_nxv16f32 = sitofp <vscale x 16 x i8> undef to <vscale x 16 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16i8_nxv16f64 = sitofp <vscale x 16 x i8> undef to <vscale x 16 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16i16_nxv16f32 = sitofp <vscale x 16 x i16> undef to <vscale x 16 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16i16_nxv16f64 = sitofp <vscale x 16 x i16> undef to <vscale x 16 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16i32_nxv16f32 = sitofp <vscale x 16 x i32> undef to <vscale x 16 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16i32_nxv16f64 = sitofp <vscale x 16 x i32> undef to <vscale x 16 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16i64_nxv16f32 = sitofp <vscale x 16 x i64> undef to <vscale x 16 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16i64_nxv16f64 = sitofp <vscale x 16 x i64> undef to <vscale x 16 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16i1_nxv16f32 = sitofp <vscale x 16 x i1> undef to <vscale x 16 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16i1_nxv16f64 = sitofp <vscale x 16 x i1> undef to <vscale x 16 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16fi8_nxv16f32 = call <vscale x 16 x float> @llvm.vp.sitofp.nxv16f32.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16fi8_nxv16f64 = call <vscale x 16 x double> @llvm.vp.sitofp.nxv16f64.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16fi16_nxv16f32 = call <vscale x 16 x float> @llvm.vp.sitofp.nxv16f32.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16fi16_nxv16f64 = call <vscale x 16 x double> @llvm.vp.sitofp.nxv16f64.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16fi32_nxv16f32 = call <vscale x 16 x float> @llvm.vp.sitofp.nxv16f32.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16fi32_nxv16f64 = call <vscale x 16 x double> @llvm.vp.sitofp.nxv16f64.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16fi64_nxv16f32 = call <vscale x 16 x float> @llvm.vp.sitofp.nxv16f32.nxv16i64(<vscale x 16 x i64> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16fi64_nxv16f64 = call <vscale x 16 x double> @llvm.vp.sitofp.nxv16f64.nxv16i64(<vscale x 16 x i64> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16fi1_nxv16f32 = call <vscale x 16 x float> @llvm.vp.sitofp.nxv16f32.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16fi1_nxv16f64 = call <vscale x 16 x double> @llvm.vp.sitofp.nxv16f64.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32i8_nxv32f32 = sitofp <vscale x 32 x i8> undef to <vscale x 32 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32i8_nxv32f64 = sitofp <vscale x 32 x i8> undef to <vscale x 32 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32i16_nxv32f32 = sitofp <vscale x 32 x i16> undef to <vscale x 32 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32i16_nxv32f64 = sitofp <vscale x 32 x i16> undef to <vscale x 32 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32i32_nxv32f32 = sitofp <vscale x 32 x i32> undef to <vscale x 32 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32i32_nxv32f64 = sitofp <vscale x 32 x i32> undef to <vscale x 32 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32i64_nxv32f32 = sitofp <vscale x 32 x i64> undef to <vscale x 32 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32i64_nxv32f64 = sitofp <vscale x 32 x i64> undef to <vscale x 32 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32i1_nxv32f32 = sitofp <vscale x 32 x i1> undef to <vscale x 32 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32i1_nxv32f64 = sitofp <vscale x 32 x i1> undef to <vscale x 32 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32fi8_nxv32f32 = call <vscale x 32 x float> @llvm.vp.sitofp.nxv32f32.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32fi8_nxv32f64 = call <vscale x 32 x double> @llvm.vp.sitofp.nxv32f64.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32fi16_nxv32f32 = call <vscale x 32 x float> @llvm.vp.sitofp.nxv32f32.nxv32i16(<vscale x 32 x i16> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32fi16_nxv32f64 = call <vscale x 32 x double> @llvm.vp.sitofp.nxv32f64.nxv32i16(<vscale x 32 x i16> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32fi32_nxv32f32 = call <vscale x 32 x float> @llvm.vp.sitofp.nxv32f32.nxv32i32(<vscale x 32 x i32> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32fi32_nxv32f64 = call <vscale x 32 x double> @llvm.vp.sitofp.nxv32f64.nxv32i32(<vscale x 32 x i32> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32fi64_nxv32f32 = call <vscale x 32 x float> @llvm.vp.sitofp.nxv32f32.nxv32i64(<vscale x 32 x i64> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32fi64_nxv32f64 = call <vscale x 32 x double> @llvm.vp.sitofp.nxv32f64.nxv32i64(<vscale x 32 x i64> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32fi1_nxv32f32 = call <vscale x 32 x float> @llvm.vp.sitofp.nxv32f32.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32fi1_nxv32f64 = call <vscale x 32 x double> @llvm.vp.sitofp.nxv32f64.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64i8_nxv64f32 = sitofp <vscale x 64 x i8> undef to <vscale x 64 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64i8_nxv64f64 = sitofp <vscale x 64 x i8> undef to <vscale x 64 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64i16_nxv64f32 = sitofp <vscale x 64 x i16> undef to <vscale x 64 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64i16_nxv64f64 = sitofp <vscale x 64 x i16> undef to <vscale x 64 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64i32_nxv64f32 = sitofp <vscale x 64 x i32> undef to <vscale x 64 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64i32_nxv64f64 = sitofp <vscale x 64 x i32> undef to <vscale x 64 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64i64_nxv64f32 = sitofp <vscale x 64 x i64> undef to <vscale x 64 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64i64_nxv64f64 = sitofp <vscale x 64 x i64> undef to <vscale x 64 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64i1_nxv64f32 = sitofp <vscale x 64 x i1> undef to <vscale x 64 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64i1_nxv64f64 = sitofp <vscale x 64 x i1> undef to <vscale x 64 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64fi8_nxv64f32 = call <vscale x 64 x float> @llvm.vp.sitofp.nxv64f32.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64fi8_nxv64f64 = call <vscale x 64 x double> @llvm.vp.sitofp.nxv64f64.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64fi16_nxv64f32 = call <vscale x 64 x float> @llvm.vp.sitofp.nxv64f32.nxv64i16(<vscale x 64 x i16> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64fi16_nxv64f64 = call <vscale x 64 x double> @llvm.vp.sitofp.nxv64f64.nxv64i16(<vscale x 64 x i16> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64fi32_nxv64f32 = call <vscale x 64 x float> @llvm.vp.sitofp.nxv64f32.nxv64i32(<vscale x 64 x i32> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64fi32_nxv64f64 = call <vscale x 64 x double> @llvm.vp.sitofp.nxv64f64.nxv64i32(<vscale x 64 x i32> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64fi64_nxv64f32 = call <vscale x 64 x float> @llvm.vp.sitofp.nxv64f32.nxv64i64(<vscale x 64 x i64> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64fi64_nxv64f64 = call <vscale x 64 x double> @llvm.vp.sitofp.nxv64f64.nxv64i64(<vscale x 64 x i64> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64fi1_nxv64f32 = call <vscale x 64 x float> @llvm.vp.sitofp.nxv64f32.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64fi1_nxv64f64 = call <vscale x 64 x double> @llvm.vp.sitofp.nxv64f64.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  %v2i8_v2f32 = sitofp <2 x i8> undef to <2 x float>
  %v2i8_v2f64 = sitofp <2 x i8> undef to <2 x double>
  %v2i16_v2f32 = sitofp <2 x i16> undef to <2 x float>
  %v2i16_v2f64 = sitofp <2 x i16> undef to <2 x double>
  %v2i32_v2f32 = sitofp <2 x i32> undef to <2 x float>
  %v2i32_v2f64 = sitofp <2 x i32> undef to <2 x double>
  %v2i64_v2f32 = sitofp <2 x i64> undef to <2 x float>
  %v2i64_v2f64 = sitofp <2 x i64> undef to <2 x double>
  %v2i1_v2f32 = sitofp <2 x i1> undef to <2 x float>
  %v2i1_v2f64 = sitofp <2 x i1> undef to <2 x double>

  %vp_v2fi8_v2f32 = call <2 x float> @llvm.vp.sitofp.v2f32.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
  %vp_v2fi8_v2f64 = call <2 x double> @llvm.vp.sitofp.v2f64.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
  %vp_v2fi16_v2f32 = call <2 x float> @llvm.vp.sitofp.v2f32.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
  %vp_v2fi16_v2f64 = call <2 x double> @llvm.vp.sitofp.v2f64.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
  %vp_v2fi32_v2f32 = call <2 x float> @llvm.vp.sitofp.v2f32.v2i32(<2 x i32> undef, <2 x i1> undef, i32 undef)
  %vp_v2fi32_v2f64 = call <2 x double> @llvm.vp.sitofp.v2f64.v2i32(<2 x i32> undef, <2 x i1> undef, i32 undef)
  %vp_v2fi64_v2f32 = call <2 x float> @llvm.vp.sitofp.v2f32.v2i64(<2 x i64> undef, <2 x i1> undef, i32 undef)
  %vp_v2fi64_v2f64 = call <2 x double> @llvm.vp.sitofp.v2f64.v2i64(<2 x i64> undef, <2 x i1> undef, i32 undef)
  %vp_v2fi1_v2f32 = call <2 x float> @llvm.vp.sitofp.v2f32.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)
  %vp_v2fi1_v2f64 = call <2 x double> @llvm.vp.sitofp.v2f64.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)

  %v4i8_v4f32 = sitofp <4 x i8> undef to <4 x float>
  %v4i8_v4f64 = sitofp <4 x i8> undef to <4 x double>
  %v4i16_v4f32 = sitofp <4 x i16> undef to <4 x float>
  %v4i16_v4f64 = sitofp <4 x i16> undef to <4 x double>
  %v4i32_v4f32 = sitofp <4 x i32> undef to <4 x float>
  %v4i32_v4f64 = sitofp <4 x i32> undef to <4 x double>
  %v4i64_v4f32 = sitofp <4 x i64> undef to <4 x float>
  %v4i64_v4f64 = sitofp <4 x i64> undef to <4 x double>
  %v4i1_v4f32 = sitofp <4 x i1> undef to <4 x float>
  %v4i1_v4f64 = sitofp <4 x i1> undef to <4 x double>

  %vp_v4fi8_v4f32 = call <4 x float> @llvm.vp.sitofp.v4f32.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
  %vp_v4fi8_v4f64 = call <4 x double> @llvm.vp.sitofp.v4f64.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
  %vp_v4fi16_v4f32 = call <4 x float> @llvm.vp.sitofp.v4f32.v4i16(<4 x i16> undef, <4 x i1> undef, i32 undef)
  %vp_v4fi16_v4f64 = call <4 x double> @llvm.vp.sitofp.v4f64.v4i16(<4 x i16> undef, <4 x i1> undef, i32 undef)
  %vp_v4fi32_v4f32 = call <4 x float> @llvm.vp.sitofp.v4f32.v4i32(<4 x i32> undef, <4 x i1> undef, i32 undef)
  %vp_v4fi32_v4f64 = call <4 x double> @llvm.vp.sitofp.v4f64.v4i32(<4 x i32> undef, <4 x i1> undef, i32 undef)
  %vp_v4fi64_v4f32 = call <4 x float> @llvm.vp.sitofp.v4f32.v4i64(<4 x i64> undef, <4 x i1> undef, i32 undef)
  %vp_v4fi64_v4f64 = call <4 x double> @llvm.vp.sitofp.v4f64.v4i64(<4 x i64> undef, <4 x i1> undef, i32 undef)
  %vp_v4fi1_v4f32 = call <4 x float> @llvm.vp.sitofp.v4f32.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)
  %vp_v4fi1_v4f64 = call <4 x double> @llvm.vp.sitofp.v4f64.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)

  %v8i8_v8f32 = sitofp <8 x i8> undef to <8 x float>
  %v8i8_v8f64 = sitofp <8 x i8> undef to <8 x double>
  %v8i16_v8f32 = sitofp <8 x i16> undef to <8 x float>
  %v8i16_v8f64 = sitofp <8 x i16> undef to <8 x double>
  %v8i32_v8f32 = sitofp <8 x i32> undef to <8 x float>
  %v8i32_v8f64 = sitofp <8 x i32> undef to <8 x double>
  %v8i64_v8f32 = sitofp <8 x i64> undef to <8 x float>
  %v8i64_v8f64 = sitofp <8 x i64> undef to <8 x double>
  %v8i1_v8f32 = sitofp <8 x i1> undef to <8 x float>
  %v8i1_v8f64 = sitofp <8 x i1> undef to <8 x double>

  %vp_v8fi8_v8f32 = call <8 x float> @llvm.vp.sitofp.v8f32.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
  %vp_v8fi8_v8f64 = call <8 x double> @llvm.vp.sitofp.v8f64.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
  %vp_v8fi16_v8f32 = call <8 x float> @llvm.vp.sitofp.v8f32.v8i16(<8 x i16> undef, <8 x i1> undef, i32 undef)
  %vp_v8fi16_v8f64 = call <8 x double> @llvm.vp.sitofp.v8f64.v8i16(<8 x i16> undef, <8 x i1> undef, i32 undef)
  %vp_v8fi32_v8f32 = call <8 x float> @llvm.vp.sitofp.v8f32.v8i32(<8 x i32> undef, <8 x i1> undef, i32 undef)
  %vp_v8fi32_v8f64 = call <8 x double> @llvm.vp.sitofp.v8f64.v8i32(<8 x i32> undef, <8 x i1> undef, i32 undef)
  %vp_v8fi64_v8f32 = call <8 x float> @llvm.vp.sitofp.v8f32.v8i64(<8 x i64> undef, <8 x i1> undef, i32 undef)
  %vp_v8fi64_v8f64 = call <8 x double> @llvm.vp.sitofp.v8f64.v8i64(<8 x i64> undef, <8 x i1> undef, i32 undef)
  %vp_v8fi1_v8f32 = call <8 x float> @llvm.vp.sitofp.v8f32.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)
  %vp_v8fi1_v8f64 = call <8 x double> @llvm.vp.sitofp.v8f64.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)

  %v16i8_v16f32 = sitofp <16 x i8> undef to <16 x float>
  %v16i8_v16f64 = sitofp <16 x i8> undef to <16 x double>
  %v16i16_v16f32 = sitofp <16 x i16> undef to <16 x float>
  %v16i16_v16f64 = sitofp <16 x i16> undef to <16 x double>
  %v16i32_v16f32 = sitofp <16 x i32> undef to <16 x float>
  %v16i32_v16f64 = sitofp <16 x i32> undef to <16 x double>
  %v16i64_v16f32 = sitofp <16 x i64> undef to <16 x float>
  %v16i64_v16f64 = sitofp <16 x i64> undef to <16 x double>
  %v16i1_v16f32 = sitofp <16 x i1> undef to <16 x float>
  %v16i1_v16f64 = sitofp <16 x i1> undef to <16 x double>

  %vp_v16fi8_v16f32 = call <16 x float> @llvm.vp.sitofp.v16f32.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
  %vp_v16fi8_v16f64 = call <16 x double> @llvm.vp.sitofp.v16f64.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
  %vp_v16fi16_v16f32 = call <16 x float> @llvm.vp.sitofp.v16f32.v16i16(<16 x i16> undef, <16 x i1> undef, i32 undef)
  %vp_v16fi16_v16f64 = call <16 x double> @llvm.vp.sitofp.v16f64.v16i16(<16 x i16> undef, <16 x i1> undef, i32 undef)
  %vp_v16fi32_v16f32 = call <16 x float> @llvm.vp.sitofp.v16f32.v16i32(<16 x i32> undef, <16 x i1> undef, i32 undef)
  %vp_v16fi32_v16f64 = call <16 x double> @llvm.vp.sitofp.v16f64.v16i32(<16 x i32> undef, <16 x i1> undef, i32 undef)
  %vp_v16fi64_v16f32 = call <16 x float> @llvm.vp.sitofp.v16f32.v16i64(<16 x i64> undef, <16 x i1> undef, i32 undef)
  %vp_v16fi64_v16f64 = call <16 x double> @llvm.vp.sitofp.v16f64.v16i64(<16 x i64> undef, <16 x i1> undef, i32 undef)
  %vp_v16fi1_v16f32 = call <16 x float> @llvm.vp.sitofp.v16f32.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)
  %vp_v16fi1_v16f64 = call <16 x double> @llvm.vp.sitofp.v16f64.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)

  %v32i8_v32f32 = sitofp <32 x i8> undef to <32 x float>
  %v32i8_v32f64 = sitofp <32 x i8> undef to <32 x double>
  %v32i16_v32f32 = sitofp <32 x i16> undef to <32 x float>
  %v32i16_v32f64 = sitofp <32 x i16> undef to <32 x double>
  %v32i32_v32f32 = sitofp <32 x i32> undef to <32 x float>
  %v32i32_v32f64 = sitofp <32 x i32> undef to <32 x double>
  %v32i64_v32f32 = sitofp <32 x i64> undef to <32 x float>
  %v32i64_v32f64 = sitofp <32 x i64> undef to <32 x double>
  %v32i1_v32f32 = sitofp <32 x i1> undef to <32 x float>
  %v32i1_v32f64 = sitofp <32 x i1> undef to <32 x double>

  %vp_v32fi8_v32f32 = call <32 x float> @llvm.vp.sitofp.v32f32.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
  %vp_v32fi8_v32f64 = call <32 x double> @llvm.vp.sitofp.v32f64.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
  %vp_v32fi16_v32f32 = call <32 x float> @llvm.vp.sitofp.v32f32.v32i16(<32 x i16> undef, <32 x i1> undef, i32 undef)
  %vp_v32fi16_v32f64 = call <32 x double> @llvm.vp.sitofp.v32f64.v32i16(<32 x i16> undef, <32 x i1> undef, i32 undef)
  %vp_v32fi32_v32f32 = call <32 x float> @llvm.vp.sitofp.v32f32.v32i32(<32 x i32> undef, <32 x i1> undef, i32 undef)
  %vp_v32fi32_v32f64 = call <32 x double> @llvm.vp.sitofp.v32f64.v32i32(<32 x i32> undef, <32 x i1> undef, i32 undef)
  %vp_v32fi64_v32f32 = call <32 x float> @llvm.vp.sitofp.v32f32.v32i64(<32 x i64> undef, <32 x i1> undef, i32 undef)
  %vp_v32fi64_v32f64 = call <32 x double> @llvm.vp.sitofp.v32f64.v32i64(<32 x i64> undef, <32 x i1> undef, i32 undef)
  %vp_v32fi1_v32f32 = call <32 x float> @llvm.vp.sitofp.v32f32.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)
  %vp_v32fi1_v32f64 = call <32 x double> @llvm.vp.sitofp.v32f64.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)

  %v64i8_v64f32 = sitofp <64 x i8> undef to <64 x float>
  %v64i8_v64f64 = sitofp <64 x i8> undef to <64 x double>
  %v64i16_v64f32 = sitofp <64 x i16> undef to <64 x float>
  %v64i16_v64f64 = sitofp <64 x i16> undef to <64 x double>
  %v64i32_v64f32 = sitofp <64 x i32> undef to <64 x float>
  %v64i32_v64f64 = sitofp <64 x i32> undef to <64 x double>
  %v64i64_v64f32 = sitofp <64 x i64> undef to <64 x float>
  %v64i64_v64f64 = sitofp <64 x i64> undef to <64 x double>
  %v64i1_v64f32 = sitofp <64 x i1> undef to <64 x float>
  %v64i1_v64f64 = sitofp <64 x i1> undef to <64 x double>

  %vp_v64fi8_v64f32 = call <64 x float> @llvm.vp.sitofp.v64f32.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
  %vp_v64fi8_v64f64 = call <64 x double> @llvm.vp.sitofp.v64f64.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
  %vp_v64fi16_v64f32 = call <64 x float> @llvm.vp.sitofp.v64f32.v64i16(<64 x i16> undef, <64 x i1> undef, i32 undef)
  %vp_v64fi16_v64f64 = call <64 x double> @llvm.vp.sitofp.v64f64.v64i16(<64 x i16> undef, <64 x i1> undef, i32 undef)
  %vp_v64fi32_v64f32 = call <64 x float> @llvm.vp.sitofp.v64f32.v64i32(<64 x i32> undef, <64 x i1> undef, i32 undef)
  %vp_v64fi32_v64f64 = call <64 x double> @llvm.vp.sitofp.v64f64.v64i32(<64 x i32> undef, <64 x i1> undef, i32 undef)
  %vp_v64fi64_v64f32 = call <64 x float> @llvm.vp.sitofp.v64f32.v64i64(<64 x i64> undef, <64 x i1> undef, i32 undef)
  %vp_v64fi64_v64f64 = call <64 x double> @llvm.vp.sitofp.v64f64.v64i64(<64 x i64> undef, <64 x i1> undef, i32 undef)
  %vp_v64fi1_v64f32 = call <64 x float> @llvm.vp.sitofp.v64f32.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)
  %vp_v64fi1_v64f64 = call <64 x double> @llvm.vp.sitofp.v64f64.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)

  %v128i8_v128f32 = sitofp <128 x i8> undef to <128 x float>
  %v128i8_v128f64 = sitofp <128 x i8> undef to <128 x double>
  %v128i16_v128f32 = sitofp <128 x i16> undef to <128 x float>
  %v128i16_v128f64 = sitofp <128 x i16> undef to <128 x double>
  %v128i32_v128f32 = sitofp <128 x i32> undef to <128 x float>
  %v128i32_v128f64 = sitofp <128 x i32> undef to <128 x double>
  %v128i64_v128f32 = sitofp <128 x i64> undef to <128 x float>
  %v128i64_v128f64 = sitofp <128 x i64> undef to <128 x double>
  %v128i1_v128f32 = sitofp <128 x i1> undef to <128 x float>
  %v128i1_v128f64 = sitofp <128 x i1> undef to <128 x double>

  %vp_v128fi8_v128f32 = call <128 x float> @llvm.vp.sitofp.v128f32.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
  %vp_v128fi8_v128f64 = call <128 x double> @llvm.vp.sitofp.v128f64.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
  %vp_v128fi16_v128f32 = call <128 x float> @llvm.vp.sitofp.v128f32.v128i16(<128 x i16> undef, <128 x i1> undef, i32 undef)
  %vp_v128fi16_v128f64 = call <128 x double> @llvm.vp.sitofp.v128f64.v128i16(<128 x i16> undef, <128 x i1> undef, i32 undef)
  %vp_v128fi32_v128f32 = call <128 x float> @llvm.vp.sitofp.v128f32.v128i32(<128 x i32> undef, <128 x i1> undef, i32 undef)
  %vp_v128fi32_v128f64 = call <128 x double> @llvm.vp.sitofp.v128f64.v128i32(<128 x i32> undef, <128 x i1> undef, i32 undef)
  %vp_v128fi64_v128f32 = call <128 x float> @llvm.vp.sitofp.v128f32.v128i64(<128 x i64> undef, <128 x i1> undef, i32 undef)
  %vp_v128fi64_v128f64 = call <128 x double> @llvm.vp.sitofp.v128f64.v128i64(<128 x i64> undef, <128 x i1> undef, i32 undef)
  %vp_v128fi1_v128f32 = call <128 x float> @llvm.vp.sitofp.v128f32.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)
  %vp_v128fi1_v128f64 = call <128 x double> @llvm.vp.sitofp.v128f64.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)

  %nxv1i8_nxv1f32 = sitofp <vscale x 1 x i8> undef to <vscale x 1 x float>
  %nxv1i8_nxv1f64 = sitofp <vscale x 1 x i8> undef to <vscale x 1 x double>
  %nxv1i16_nxv1f32 = sitofp <vscale x 1 x i16> undef to <vscale x 1 x float>
  %nxv1i16_nxv1f64 = sitofp <vscale x 1 x i16> undef to <vscale x 1 x double>
  %nxv1i32_nxv1f32 = sitofp <vscale x 1 x i32> undef to <vscale x 1 x float>
  %nxv1i32_nxv1f64 = sitofp <vscale x 1 x i32> undef to <vscale x 1 x double>
  %nxv1i64_nxv1f32 = sitofp <vscale x 1 x i64> undef to <vscale x 1 x float>
  %nxv1i64_nxv1f64 = sitofp <vscale x 1 x i64> undef to <vscale x 1 x double>
  %nxv1i1_nxv1f32 = sitofp <vscale x 1 x i1> undef to <vscale x 1 x float>
  %nxv1i1_nxv1f64 = sitofp <vscale x 1 x i1> undef to <vscale x 1 x double>

  %vp_nxv1fi8_nxv1f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1fi8_nxv1f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1fi16_nxv1f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1fi16_nxv1f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1fi32_nxv1f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1fi32_nxv1f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1fi64_nxv1f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1fi64_nxv1f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1fi1_nxv1f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1fi1_nxv1f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)

  %nxv2i8_nxv2f32 = sitofp <vscale x 2 x i8> undef to <vscale x 2 x float>
  %nxv2i8_nxv2f64 = sitofp <vscale x 2 x i8> undef to <vscale x 2 x double>
  %nxv2i16_nxv2f32 = sitofp <vscale x 2 x i16> undef to <vscale x 2 x float>
  %nxv2i16_nxv2f64 = sitofp <vscale x 2 x i16> undef to <vscale x 2 x double>
  %nxv2i32_nxv2f32 = sitofp <vscale x 2 x i32> undef to <vscale x 2 x float>
  %nxv2i32_nxv2f64 = sitofp <vscale x 2 x i32> undef to <vscale x 2 x double>
  %nxv2i64_nxv2f32 = sitofp <vscale x 2 x i64> undef to <vscale x 2 x float>
  %nxv2i64_nxv2f64 = sitofp <vscale x 2 x i64> undef to <vscale x 2 x double>
  %nxv2i1_nxv2f32 = sitofp <vscale x 2 x i1> undef to <vscale x 2 x float>
  %nxv2i1_nxv2f64 = sitofp <vscale x 2 x i1> undef to <vscale x 2 x double>

  %vp_nxv2fi8_nxv2f32 = call <vscale x 2 x float> @llvm.vp.sitofp.nxv2f32.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2fi8_nxv2f64 = call <vscale x 2 x double> @llvm.vp.sitofp.nxv2f64.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2fi16_nxv2f32 = call <vscale x 2 x float> @llvm.vp.sitofp.nxv2f32.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2fi16_nxv2f64 = call <vscale x 2 x double> @llvm.vp.sitofp.nxv2f64.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2fi32_nxv2f32 = call <vscale x 2 x float> @llvm.vp.sitofp.nxv2f32.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2fi32_nxv2f64 = call <vscale x 2 x double> @llvm.vp.sitofp.nxv2f64.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2fi64_nxv2f32 = call <vscale x 2 x float> @llvm.vp.sitofp.nxv2f32.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2fi64_nxv2f64 = call <vscale x 2 x double> @llvm.vp.sitofp.nxv2f64.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2fi1_nxv2f32 = call <vscale x 2 x float> @llvm.vp.sitofp.nxv2f32.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2fi1_nxv2f64 = call <vscale x 2 x double> @llvm.vp.sitofp.nxv2f64.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)

  %nxv4i8_nxv4f32 = sitofp <vscale x 4 x i8> undef to <vscale x 4 x float>
  %nxv4i8_nxv4f64 = sitofp <vscale x 4 x i8> undef to <vscale x 4 x double>
  %nxv4i16_nxv4f32 = sitofp <vscale x 4 x i16> undef to <vscale x 4 x float>
  %nxv4i16_nxv4f64 = sitofp <vscale x 4 x i16> undef to <vscale x 4 x double>
  %nxv4i32_nxv4f32 = sitofp <vscale x 4 x i32> undef to <vscale x 4 x float>
  %nxv4i32_nxv4f64 = sitofp <vscale x 4 x i32> undef to <vscale x 4 x double>
  %nxv4i64_nxv4f32 = sitofp <vscale x 4 x i64> undef to <vscale x 4 x float>
  %nxv4i64_nxv4f64 = sitofp <vscale x 4 x i64> undef to <vscale x 4 x double>
  %nxv4i1_nxv4f32 = sitofp <vscale x 4 x i1> undef to <vscale x 4 x float>
  %nxv4i1_nxv4f64 = sitofp <vscale x 4 x i1> undef to <vscale x 4 x double>

  %vp_nxv4fi8_nxv4f32 = call <vscale x 4 x float> @llvm.vp.sitofp.nxv4f32.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4fi8_nxv4f64 = call <vscale x 4 x double> @llvm.vp.sitofp.nxv4f64.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4fi16_nxv4f32 = call <vscale x 4 x float> @llvm.vp.sitofp.nxv4f32.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4fi16_nxv4f64 = call <vscale x 4 x double> @llvm.vp.sitofp.nxv4f64.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4fi32_nxv4f32 = call <vscale x 4 x float> @llvm.vp.sitofp.nxv4f32.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4fi32_nxv4f64 = call <vscale x 4 x double> @llvm.vp.sitofp.nxv4f64.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4fi64_nxv4f32 = call <vscale x 4 x float> @llvm.vp.sitofp.nxv4f32.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4fi64_nxv4f64 = call <vscale x 4 x double> @llvm.vp.sitofp.nxv4f64.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4fi1_nxv4f32 = call <vscale x 4 x float> @llvm.vp.sitofp.nxv4f32.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4fi1_nxv4f64 = call <vscale x 4 x double> @llvm.vp.sitofp.nxv4f64.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)

  %nxv8i8_nxv8f32 = sitofp <vscale x 8 x i8> undef to <vscale x 8 x float>
  %nxv8i8_nxv8f64 = sitofp <vscale x 8 x i8> undef to <vscale x 8 x double>
  %nxv8i16_nxv8f32 = sitofp <vscale x 8 x i16> undef to <vscale x 8 x float>
  %nxv8i16_nxv8f64 = sitofp <vscale x 8 x i16> undef to <vscale x 8 x double>
  %nxv8i32_nxv8f32 = sitofp <vscale x 8 x i32> undef to <vscale x 8 x float>
  %nxv8i32_nxv8f64 = sitofp <vscale x 8 x i32> undef to <vscale x 8 x double>
  %nxv8i64_nxv8f32 = sitofp <vscale x 8 x i64> undef to <vscale x 8 x float>
  %nxv8i64_nxv8f64 = sitofp <vscale x 8 x i64> undef to <vscale x 8 x double>
  %nxv8i1_nxv8f32 = sitofp <vscale x 8 x i1> undef to <vscale x 8 x float>
  %nxv8i1_nxv8f64 = sitofp <vscale x 8 x i1> undef to <vscale x 8 x double>

  %vp_nxv8fi8_nxv8f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv8fi8_nxv8f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv8fi16_nxv8f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv8fi16_nxv8f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv8fi32_nxv8f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv8fi32_nxv8f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv8fi64_nxv8f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv8fi64_nxv8f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv8fi1_nxv8f32 = call <vscale x 1 x float> @llvm.vp.sitofp.nxv1f32.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv8fi1_nxv8f64 = call <vscale x 1 x double> @llvm.vp.sitofp.nxv1f64.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)

  %nxv16i8_nxv16f32 = sitofp <vscale x 16 x i8> undef to <vscale x 16 x float>
  %nxv16i8_nxv16f64 = sitofp <vscale x 16 x i8> undef to <vscale x 16 x double>
  %nxv16i16_nxv16f32 = sitofp <vscale x 16 x i16> undef to <vscale x 16 x float>
  %nxv16i16_nxv16f64 = sitofp <vscale x 16 x i16> undef to <vscale x 16 x double>
  %nxv16i32_nxv16f32 = sitofp <vscale x 16 x i32> undef to <vscale x 16 x float>
  %nxv16i32_nxv16f64 = sitofp <vscale x 16 x i32> undef to <vscale x 16 x double>
  %nxv16i64_nxv16f32 = sitofp <vscale x 16 x i64> undef to <vscale x 16 x float>
  %nxv16i64_nxv16f64 = sitofp <vscale x 16 x i64> undef to <vscale x 16 x double>
  %nxv16i1_nxv16f32 = sitofp <vscale x 16 x i1> undef to <vscale x 16 x float>
  %nxv16i1_nxv16f64 = sitofp <vscale x 16 x i1> undef to <vscale x 16 x double>

  %vp_nxv16fi8_nxv16f32 = call <vscale x 16 x float> @llvm.vp.sitofp.nxv16f32.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16fi8_nxv16f64 = call <vscale x 16 x double> @llvm.vp.sitofp.nxv16f64.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16fi16_nxv16f32 = call <vscale x 16 x float> @llvm.vp.sitofp.nxv16f32.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16fi16_nxv16f64 = call <vscale x 16 x double> @llvm.vp.sitofp.nxv16f64.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16fi32_nxv16f32 = call <vscale x 16 x float> @llvm.vp.sitofp.nxv16f32.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16fi32_nxv16f64 = call <vscale x 16 x double> @llvm.vp.sitofp.nxv16f64.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16fi64_nxv16f32 = call <vscale x 16 x float> @llvm.vp.sitofp.nxv16f32.nxv16i64(<vscale x 16 x i64> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16fi64_nxv16f64 = call <vscale x 16 x double> @llvm.vp.sitofp.nxv16f64.nxv16i64(<vscale x 16 x i64> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16fi1_nxv16f32 = call <vscale x 16 x float> @llvm.vp.sitofp.nxv16f32.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16fi1_nxv16f64 = call <vscale x 16 x double> @llvm.vp.sitofp.nxv16f64.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)

  %nxv32i8_nxv32f32 = sitofp <vscale x 32 x i8> undef to <vscale x 32 x float>
  %nxv32i8_nxv32f64 = sitofp <vscale x 32 x i8> undef to <vscale x 32 x double>
  %nxv32i16_nxv32f32 = sitofp <vscale x 32 x i16> undef to <vscale x 32 x float>
  %nxv32i16_nxv32f64 = sitofp <vscale x 32 x i16> undef to <vscale x 32 x double>
  %nxv32i32_nxv32f32 = sitofp <vscale x 32 x i32> undef to <vscale x 32 x float>
  %nxv32i32_nxv32f64 = sitofp <vscale x 32 x i32> undef to <vscale x 32 x double>
  %nxv32i64_nxv32f32 = sitofp <vscale x 32 x i64> undef to <vscale x 32 x float>
  %nxv32i64_nxv32f64 = sitofp <vscale x 32 x i64> undef to <vscale x 32 x double>
  %nxv32i1_nxv32f32 = sitofp <vscale x 32 x i1> undef to <vscale x 32 x float>
  %nxv32i1_nxv32f64 = sitofp <vscale x 32 x i1> undef to <vscale x 32 x double>

  %vp_nxv32fi8_nxv32f32 = call <vscale x 32 x float> @llvm.vp.sitofp.nxv32f32.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32fi8_nxv32f64 = call <vscale x 32 x double> @llvm.vp.sitofp.nxv32f64.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32fi16_nxv32f32 = call <vscale x 32 x float> @llvm.vp.sitofp.nxv32f32.nxv32i16(<vscale x 32 x i16> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32fi16_nxv32f64 = call <vscale x 32 x double> @llvm.vp.sitofp.nxv32f64.nxv32i16(<vscale x 32 x i16> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32fi32_nxv32f32 = call <vscale x 32 x float> @llvm.vp.sitofp.nxv32f32.nxv32i32(<vscale x 32 x i32> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32fi32_nxv32f64 = call <vscale x 32 x double> @llvm.vp.sitofp.nxv32f64.nxv32i32(<vscale x 32 x i32> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32fi64_nxv32f32 = call <vscale x 32 x float> @llvm.vp.sitofp.nxv32f32.nxv32i64(<vscale x 32 x i64> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32fi64_nxv32f64 = call <vscale x 32 x double> @llvm.vp.sitofp.nxv32f64.nxv32i64(<vscale x 32 x i64> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32fi1_nxv32f32 = call <vscale x 32 x float> @llvm.vp.sitofp.nxv32f32.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32fi1_nxv32f64 = call <vscale x 32 x double> @llvm.vp.sitofp.nxv32f64.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)

  %nxv64i8_nxv64f32 = sitofp <vscale x 64 x i8> undef to <vscale x 64 x float>
  %nxv64i8_nxv64f64 = sitofp <vscale x 64 x i8> undef to <vscale x 64 x double>
  %nxv64i16_nxv64f32 = sitofp <vscale x 64 x i16> undef to <vscale x 64 x float>
  %nxv64i16_nxv64f64 = sitofp <vscale x 64 x i16> undef to <vscale x 64 x double>
  %nxv64i32_nxv64f32 = sitofp <vscale x 64 x i32> undef to <vscale x 64 x float>
  %nxv64i32_nxv64f64 = sitofp <vscale x 64 x i32> undef to <vscale x 64 x double>
  %nxv64i64_nxv64f32 = sitofp <vscale x 64 x i64> undef to <vscale x 64 x float>
  %nxv64i64_nxv64f64 = sitofp <vscale x 64 x i64> undef to <vscale x 64 x double>
  %nxv64i1_nxv64f32 = sitofp <vscale x 64 x i1> undef to <vscale x 64 x float>
  %nxv64i1_nxv64f64 = sitofp <vscale x 64 x i1> undef to <vscale x 64 x double>

  %vp_nxv64fi8_nxv64f32 = call <vscale x 64 x float> @llvm.vp.sitofp.nxv64f32.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64fi8_nxv64f64 = call <vscale x 64 x double> @llvm.vp.sitofp.nxv64f64.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64fi16_nxv64f32 = call <vscale x 64 x float> @llvm.vp.sitofp.nxv64f32.nxv64i16(<vscale x 64 x i16> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64fi16_nxv64f64 = call <vscale x 64 x double> @llvm.vp.sitofp.nxv64f64.nxv64i16(<vscale x 64 x i16> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64fi32_nxv64f32 = call <vscale x 64 x float> @llvm.vp.sitofp.nxv64f32.nxv64i32(<vscale x 64 x i32> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64fi32_nxv64f64 = call <vscale x 64 x double> @llvm.vp.sitofp.nxv64f64.nxv64i32(<vscale x 64 x i32> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64fi64_nxv64f32 = call <vscale x 64 x float> @llvm.vp.sitofp.nxv64f32.nxv64i64(<vscale x 64 x i64> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64fi64_nxv64f64 = call <vscale x 64 x double> @llvm.vp.sitofp.nxv64f64.nxv64i64(<vscale x 64 x i64> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64fi1_nxv64f32 = call <vscale x 64 x float> @llvm.vp.sitofp.nxv64f32.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64fi1_nxv64f64 = call <vscale x 64 x double> @llvm.vp.sitofp.nxv64f64.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)

  ret void
}

define void @uitofp() {
; ZVE64D-LABEL: 'uitofp'
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2i8_v2f32 = uitofp <2 x i8> undef to <2 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2i8_v2f64 = uitofp <2 x i8> undef to <2 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i16_v2f32 = uitofp <2 x i16> undef to <2 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2i16_v2f64 = uitofp <2 x i16> undef to <2 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i32_v2f32 = uitofp <2 x i32> undef to <2 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i32_v2f64 = uitofp <2 x i32> undef to <2 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i64_v2f32 = uitofp <2 x i64> undef to <2 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v2i64_v2f64 = uitofp <2 x i64> undef to <2 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2i1_v2f32 = uitofp <2 x i1> undef to <2 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2i1_v2f64 = uitofp <2 x i1> undef to <2 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2fi8_v2f32 = call <2 x float> @llvm.vp.uitofp.v2f32.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2fi8_v2f64 = call <2 x double> @llvm.vp.uitofp.v2f64.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2fi16_v2f32 = call <2 x float> @llvm.vp.uitofp.v2f32.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v2fi16_v2f64 = call <2 x double> @llvm.vp.uitofp.v2f64.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2fi32_v2f32 = call <2 x float> @llvm.vp.uitofp.v2f32.v2i32(<2 x i32> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2fi32_v2f64 = call <2 x double> @llvm.vp.uitofp.v2f64.v2i32(<2 x i32> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2fi64_v2f32 = call <2 x float> @llvm.vp.uitofp.v2f32.v2i64(<2 x i64> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v2fi64_v2f64 = call <2 x double> @llvm.vp.uitofp.v2f64.v2i64(<2 x i64> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2fi1_v2f32 = call <2 x float> @llvm.vp.uitofp.v2f32.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2fi1_v2f64 = call <2 x double> @llvm.vp.uitofp.v2f64.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i8_v4f32 = uitofp <4 x i8> undef to <4 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v4i8_v4f64 = uitofp <4 x i8> undef to <4 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4i16_v4f32 = uitofp <4 x i16> undef to <4 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v4i16_v4f64 = uitofp <4 x i16> undef to <4 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4i32_v4f32 = uitofp <4 x i32> undef to <4 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i32_v4f64 = uitofp <4 x i32> undef to <4 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4i64_v4f32 = uitofp <4 x i64> undef to <4 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i64_v4f64 = uitofp <4 x i64> undef to <4 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v4i1_v4f32 = uitofp <4 x i1> undef to <4 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v4i1_v4f64 = uitofp <4 x i1> undef to <4 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4fi8_v4f32 = call <4 x float> @llvm.vp.uitofp.v4f32.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v4fi8_v4f64 = call <4 x double> @llvm.vp.uitofp.v4f64.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v4fi16_v4f32 = call <4 x float> @llvm.vp.uitofp.v4f32.v4i16(<4 x i16> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v4fi16_v4f64 = call <4 x double> @llvm.vp.uitofp.v4f64.v4i16(<4 x i16> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v4fi32_v4f32 = call <4 x float> @llvm.vp.uitofp.v4f32.v4i32(<4 x i32> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4fi32_v4f64 = call <4 x double> @llvm.vp.uitofp.v4f64.v4i32(<4 x i32> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_v4fi64_v4f32 = call <4 x float> @llvm.vp.uitofp.v4f32.v4i64(<4 x i64> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v4fi64_v4f64 = call <4 x double> @llvm.vp.uitofp.v4f64.v4i64(<4 x i64> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v4fi1_v4f32 = call <4 x float> @llvm.vp.uitofp.v4f32.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v4fi1_v4f64 = call <4 x double> @llvm.vp.uitofp.v4f64.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v8i8_v8f32 = uitofp <8 x i8> undef to <8 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %v8i8_v8f64 = uitofp <8 x i8> undef to <8 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8i16_v8f32 = uitofp <8 x i16> undef to <8 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %v8i16_v8f64 = uitofp <8 x i16> undef to <8 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8i32_v8f32 = uitofp <8 x i32> undef to <8 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8i32_v8f64 = uitofp <8 x i32> undef to <8 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v8i64_v8f32 = uitofp <8 x i64> undef to <8 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8i64_v8f64 = uitofp <8 x i64> undef to <8 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8i1_v8f32 = uitofp <8 x i1> undef to <8 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v8i1_v8f64 = uitofp <8 x i1> undef to <8 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v8fi8_v8f32 = call <8 x float> @llvm.vp.uitofp.v8f32.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_v8fi8_v8f64 = call <8 x double> @llvm.vp.uitofp.v8f64.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8fi16_v8f32 = call <8 x float> @llvm.vp.uitofp.v8f32.v8i16(<8 x i16> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_v8fi16_v8f64 = call <8 x double> @llvm.vp.uitofp.v8f64.v8i16(<8 x i16> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8fi32_v8f32 = call <8 x float> @llvm.vp.uitofp.v8f32.v8i32(<8 x i32> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v8fi32_v8f64 = call <8 x double> @llvm.vp.uitofp.v8f64.v8i32(<8 x i32> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_v8fi64_v8f32 = call <8 x float> @llvm.vp.uitofp.v8f32.v8i64(<8 x i64> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v8fi64_v8f64 = call <8 x double> @llvm.vp.uitofp.v8f64.v8i64(<8 x i64> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v8fi1_v8f32 = call <8 x float> @llvm.vp.uitofp.v8f32.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v8fi1_v8f64 = call <8 x double> @llvm.vp.uitofp.v8f64.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %v16i8_v16f32 = uitofp <16 x i8> undef to <16 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %v16i8_v16f64 = uitofp <16 x i8> undef to <16 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v16i16_v16f32 = uitofp <16 x i16> undef to <16 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %v16i16_v16f64 = uitofp <16 x i16> undef to <16 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v16i32_v16f32 = uitofp <16 x i32> undef to <16 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v16i32_v16f64 = uitofp <16 x i32> undef to <16 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v16i64_v16f32 = uitofp <16 x i64> undef to <16 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v16i64_v16f64 = uitofp <16 x i64> undef to <16 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v16i1_v16f32 = uitofp <16 x i1> undef to <16 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v16i1_v16f64 = uitofp <16 x i1> undef to <16 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_v16fi8_v16f32 = call <16 x float> @llvm.vp.uitofp.v16f32.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %vp_v16fi8_v16f64 = call <16 x double> @llvm.vp.uitofp.v16f64.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v16fi16_v16f32 = call <16 x float> @llvm.vp.uitofp.v16f32.v16i16(<16 x i16> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %vp_v16fi16_v16f64 = call <16 x double> @llvm.vp.uitofp.v16f64.v16i16(<16 x i16> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v16fi32_v16f32 = call <16 x float> @llvm.vp.uitofp.v16f32.v16i32(<16 x i32> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v16fi32_v16f64 = call <16 x double> @llvm.vp.uitofp.v16f64.v16i32(<16 x i32> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_v16fi64_v16f32 = call <16 x float> @llvm.vp.uitofp.v16f32.v16i64(<16 x i64> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v16fi64_v16f64 = call <16 x double> @llvm.vp.uitofp.v16f64.v16i64(<16 x i64> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v16fi1_v16f32 = call <16 x float> @llvm.vp.uitofp.v16f32.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v16fi1_v16f64 = call <16 x double> @llvm.vp.uitofp.v16f64.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %v32i8_v32f32 = uitofp <32 x i8> undef to <32 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %v32i8_v32f64 = uitofp <32 x i8> undef to <32 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v32i16_v32f32 = uitofp <32 x i16> undef to <32 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %v32i16_v32f64 = uitofp <32 x i16> undef to <32 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v32i32_v32f32 = uitofp <32 x i32> undef to <32 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %v32i32_v32f64 = uitofp <32 x i32> undef to <32 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %v32i64_v32f32 = uitofp <32 x i64> undef to <32 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v32i64_v32f64 = uitofp <32 x i64> undef to <32 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v32i1_v32f32 = uitofp <32 x i1> undef to <32 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %v32i1_v32f64 = uitofp <32 x i1> undef to <32 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %vp_v32fi8_v32f32 = call <32 x float> @llvm.vp.uitofp.v32f32.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_v32fi8_v32f64 = call <32 x double> @llvm.vp.uitofp.v32f64.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v32fi16_v32f32 = call <32 x float> @llvm.vp.uitofp.v32f32.v32i16(<32 x i16> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_v32fi16_v32f64 = call <32 x double> @llvm.vp.uitofp.v32f64.v32i16(<32 x i16> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_v32fi32_v32f32 = call <32 x float> @llvm.vp.uitofp.v32f32.v32i32(<32 x i32> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_v32fi32_v32f64 = call <32 x double> @llvm.vp.uitofp.v32f64.v32i32(<32 x i32> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %vp_v32fi64_v32f32 = call <32 x float> @llvm.vp.uitofp.v32f32.v32i64(<32 x i64> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v32fi64_v32f64 = call <32 x double> @llvm.vp.uitofp.v32f64.v32i64(<32 x i64> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v32fi1_v32f32 = call <32 x float> @llvm.vp.uitofp.v32f32.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %vp_v32fi1_v32f64 = call <32 x double> @llvm.vp.uitofp.v32f64.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %v64i8_v64f32 = uitofp <64 x i8> undef to <64 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %v64i8_v64f64 = uitofp <64 x i8> undef to <64 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %v64i16_v64f32 = uitofp <64 x i16> undef to <64 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %v64i16_v64f64 = uitofp <64 x i16> undef to <64 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %v64i32_v64f32 = uitofp <64 x i32> undef to <64 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %v64i32_v64f64 = uitofp <64 x i32> undef to <64 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %v64i64_v64f32 = uitofp <64 x i64> undef to <64 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %v64i64_v64f64 = uitofp <64 x i64> undef to <64 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %v64i1_v64f32 = uitofp <64 x i1> undef to <64 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %v64i1_v64f64 = uitofp <64 x i1> undef to <64 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_v64fi8_v64f32 = call <64 x float> @llvm.vp.uitofp.v64f32.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %vp_v64fi8_v64f64 = call <64 x double> @llvm.vp.uitofp.v64f64.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_v64fi16_v64f32 = call <64 x float> @llvm.vp.uitofp.v64f32.v64i16(<64 x i16> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %vp_v64fi16_v64f64 = call <64 x double> @llvm.vp.uitofp.v64f64.v64i16(<64 x i16> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_v64fi32_v64f32 = call <64 x float> @llvm.vp.uitofp.v64f32.v64i32(<64 x i32> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_v64fi32_v64f64 = call <64 x double> @llvm.vp.uitofp.v64f64.v64i32(<64 x i32> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %vp_v64fi64_v64f32 = call <64 x float> @llvm.vp.uitofp.v64f32.v64i64(<64 x i64> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %vp_v64fi64_v64f64 = call <64 x double> @llvm.vp.uitofp.v64f64.v64i64(<64 x i64> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %vp_v64fi1_v64f32 = call <64 x float> @llvm.vp.uitofp.v64f32.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %vp_v64fi1_v64f64 = call <64 x double> @llvm.vp.uitofp.v64f64.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %v128i8_v128f32 = uitofp <128 x i8> undef to <128 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 103 for instruction: %v128i8_v128f64 = uitofp <128 x i8> undef to <128 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %v128i16_v128f32 = uitofp <128 x i16> undef to <128 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 102 for instruction: %v128i16_v128f64 = uitofp <128 x i16> undef to <128 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %v128i32_v128f32 = uitofp <128 x i32> undef to <128 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %v128i32_v128f64 = uitofp <128 x i32> undef to <128 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %v128i64_v128f32 = uitofp <128 x i64> undef to <128 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %v128i64_v128f64 = uitofp <128 x i64> undef to <128 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %v128i1_v128f32 = uitofp <128 x i1> undef to <128 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 135 for instruction: %v128i1_v128f64 = uitofp <128 x i1> undef to <128 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %vp_v128fi8_v128f32 = call <128 x float> @llvm.vp.uitofp.v128f32.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 103 for instruction: %vp_v128fi8_v128f64 = call <128 x double> @llvm.vp.uitofp.v128f64.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_v128fi16_v128f32 = call <128 x float> @llvm.vp.uitofp.v128f32.v128i16(<128 x i16> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 102 for instruction: %vp_v128fi16_v128f64 = call <128 x double> @llvm.vp.uitofp.v128f64.v128i16(<128 x i16> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %vp_v128fi32_v128f32 = call <128 x float> @llvm.vp.uitofp.v128f32.v128i32(<128 x i32> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %vp_v128fi32_v128f64 = call <128 x double> @llvm.vp.uitofp.v128f64.v128i32(<128 x i32> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %vp_v128fi64_v128f32 = call <128 x float> @llvm.vp.uitofp.v128f32.v128i64(<128 x i64> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %vp_v128fi64_v128f64 = call <128 x double> @llvm.vp.uitofp.v128f64.v128i64(<128 x i64> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %vp_v128fi1_v128f32 = call <128 x float> @llvm.vp.uitofp.v128f32.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 135 for instruction: %vp_v128fi1_v128f64 = call <128 x double> @llvm.vp.uitofp.v128f64.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1i8_nxv1f32 = uitofp <vscale x 1 x i8> undef to <vscale x 1 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1i8_nxv1f64 = uitofp <vscale x 1 x i8> undef to <vscale x 1 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1i16_nxv1f32 = uitofp <vscale x 1 x i16> undef to <vscale x 1 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv1i16_nxv1f64 = uitofp <vscale x 1 x i16> undef to <vscale x 1 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1i32_nxv1f32 = uitofp <vscale x 1 x i32> undef to <vscale x 1 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1i32_nxv1f64 = uitofp <vscale x 1 x i32> undef to <vscale x 1 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1i64_nxv1f32 = uitofp <vscale x 1 x i64> undef to <vscale x 1 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv1i64_nxv1f64 = uitofp <vscale x 1 x i64> undef to <vscale x 1 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv1i1_nxv1f32 = uitofp <vscale x 1 x i1> undef to <vscale x 1 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv1i1_nxv1f64 = uitofp <vscale x 1 x i1> undef to <vscale x 1 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1fi8_nxv1f32 = call <vscale x 1 x float> @llvm.vp.uitofp.nxv1f32.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1fi8_nxv1f64 = call <vscale x 1 x double> @llvm.vp.uitofp.nxv1f64.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1fi16_nxv1f32 = call <vscale x 1 x float> @llvm.vp.uitofp.nxv1f32.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv1fi16_nxv1f64 = call <vscale x 1 x double> @llvm.vp.uitofp.nxv1f64.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1fi32_nxv1f32 = call <vscale x 1 x float> @llvm.vp.uitofp.nxv1f32.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1fi32_nxv1f64 = call <vscale x 1 x double> @llvm.vp.uitofp.nxv1f64.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1fi64_nxv1f32 = call <vscale x 1 x float> @llvm.vp.uitofp.nxv1f32.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv1fi64_nxv1f64 = call <vscale x 1 x double> @llvm.vp.uitofp.nxv1f64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv1fi1_nxv1f32 = call <vscale x 1 x float> @llvm.vp.uitofp.nxv1f32.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv1fi1_nxv1f64 = call <vscale x 1 x double> @llvm.vp.uitofp.nxv1f64.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2i8_nxv2f32 = uitofp <vscale x 2 x i8> undef to <vscale x 2 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv2i8_nxv2f64 = uitofp <vscale x 2 x i8> undef to <vscale x 2 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv2i16_nxv2f32 = uitofp <vscale x 2 x i16> undef to <vscale x 2 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv2i16_nxv2f64 = uitofp <vscale x 2 x i16> undef to <vscale x 2 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv2i32_nxv2f32 = uitofp <vscale x 2 x i32> undef to <vscale x 2 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2i32_nxv2f64 = uitofp <vscale x 2 x i32> undef to <vscale x 2 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %nxv2i64_nxv2f32 = uitofp <vscale x 2 x i64> undef to <vscale x 2 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv2i64_nxv2f64 = uitofp <vscale x 2 x i64> undef to <vscale x 2 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv2i1_nxv2f32 = uitofp <vscale x 2 x i1> undef to <vscale x 2 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv2i1_nxv2f64 = uitofp <vscale x 2 x i1> undef to <vscale x 2 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2fi8_nxv2f32 = call <vscale x 2 x float> @llvm.vp.uitofp.nxv2f32.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv2fi8_nxv2f64 = call <vscale x 2 x double> @llvm.vp.uitofp.nxv2f64.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv2fi16_nxv2f32 = call <vscale x 2 x float> @llvm.vp.uitofp.nxv2f32.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv2fi16_nxv2f64 = call <vscale x 2 x double> @llvm.vp.uitofp.nxv2f64.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv2fi32_nxv2f32 = call <vscale x 2 x float> @llvm.vp.uitofp.nxv2f32.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2fi32_nxv2f64 = call <vscale x 2 x double> @llvm.vp.uitofp.nxv2f64.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %vp_nxv2fi64_nxv2f32 = call <vscale x 2 x float> @llvm.vp.uitofp.nxv2f32.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv2fi64_nxv2f64 = call <vscale x 2 x double> @llvm.vp.uitofp.nxv2f64.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv2fi1_nxv2f32 = call <vscale x 2 x float> @llvm.vp.uitofp.nxv2f32.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv2fi1_nxv2f64 = call <vscale x 2 x double> @llvm.vp.uitofp.nxv2f64.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %nxv4i8_nxv4f32 = uitofp <vscale x 4 x i8> undef to <vscale x 4 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %nxv4i8_nxv4f64 = uitofp <vscale x 4 x i8> undef to <vscale x 4 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4i16_nxv4f32 = uitofp <vscale x 4 x i16> undef to <vscale x 4 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %nxv4i16_nxv4f64 = uitofp <vscale x 4 x i16> undef to <vscale x 4 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4i32_nxv4f32 = uitofp <vscale x 4 x i32> undef to <vscale x 4 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv4i32_nxv4f64 = uitofp <vscale x 4 x i32> undef to <vscale x 4 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv4i64_nxv4f32 = uitofp <vscale x 4 x i64> undef to <vscale x 4 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv4i64_nxv4f64 = uitofp <vscale x 4 x i64> undef to <vscale x 4 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv4i1_nxv4f32 = uitofp <vscale x 4 x i1> undef to <vscale x 4 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv4i1_nxv4f64 = uitofp <vscale x 4 x i1> undef to <vscale x 4 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_nxv4fi8_nxv4f32 = call <vscale x 4 x float> @llvm.vp.uitofp.nxv4f32.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_nxv4fi8_nxv4f64 = call <vscale x 4 x double> @llvm.vp.uitofp.nxv4f64.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4fi16_nxv4f32 = call <vscale x 4 x float> @llvm.vp.uitofp.nxv4f32.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_nxv4fi16_nxv4f64 = call <vscale x 4 x double> @llvm.vp.uitofp.nxv4f64.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4fi32_nxv4f32 = call <vscale x 4 x float> @llvm.vp.uitofp.nxv4f32.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv4fi32_nxv4f64 = call <vscale x 4 x double> @llvm.vp.uitofp.nxv4f64.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv4fi64_nxv4f32 = call <vscale x 4 x float> @llvm.vp.uitofp.nxv4f32.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv4fi64_nxv4f64 = call <vscale x 4 x double> @llvm.vp.uitofp.nxv4f64.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv4fi1_nxv4f32 = call <vscale x 4 x float> @llvm.vp.uitofp.nxv4f32.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv4fi1_nxv4f64 = call <vscale x 4 x double> @llvm.vp.uitofp.nxv4f64.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %nxv8i8_nxv8f32 = uitofp <vscale x 8 x i8> undef to <vscale x 8 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %nxv8i8_nxv8f64 = uitofp <vscale x 8 x i8> undef to <vscale x 8 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv8i16_nxv8f32 = uitofp <vscale x 8 x i16> undef to <vscale x 8 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %nxv8i16_nxv8f64 = uitofp <vscale x 8 x i16> undef to <vscale x 8 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv8i32_nxv8f32 = uitofp <vscale x 8 x i32> undef to <vscale x 8 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv8i32_nxv8f64 = uitofp <vscale x 8 x i32> undef to <vscale x 8 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv8i64_nxv8f32 = uitofp <vscale x 8 x i64> undef to <vscale x 8 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv8i64_nxv8f64 = uitofp <vscale x 8 x i64> undef to <vscale x 8 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv8i1_nxv8f32 = uitofp <vscale x 8 x i1> undef to <vscale x 8 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %nxv8i1_nxv8f64 = uitofp <vscale x 8 x i1> undef to <vscale x 8 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %vp_nxv8fi8_nxv8f32 = call <vscale x 8 x float> @llvm.vp.uitofp.nxv8f32.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %vp_nxv8fi8_nxv8f64 = call <vscale x 8 x double> @llvm.vp.uitofp.nxv8f64.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv8fi16_nxv8f32 = call <vscale x 8 x float> @llvm.vp.uitofp.nxv8f32.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %vp_nxv8fi16_nxv8f64 = call <vscale x 8 x double> @llvm.vp.uitofp.nxv8f64.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv8fi32_nxv8f32 = call <vscale x 8 x float> @llvm.vp.uitofp.nxv8f32.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv8fi32_nxv8f64 = call <vscale x 8 x double> @llvm.vp.uitofp.nxv8f64.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv8fi64_nxv8f32 = call <vscale x 8 x float> @llvm.vp.uitofp.nxv8f32.nxv8i64(<vscale x 8 x i64> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv8fi64_nxv8f64 = call <vscale x 8 x double> @llvm.vp.uitofp.nxv8f64.nxv8i64(<vscale x 8 x i64> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv8fi1_nxv8f32 = call <vscale x 8 x float> @llvm.vp.uitofp.nxv8f32.nxv8i1(<vscale x 8 x i1> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_nxv8fi1_nxv8f64 = call <vscale x 8 x double> @llvm.vp.uitofp.nxv8f64.nxv8i1(<vscale x 8 x i1> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %nxv16i8_nxv16f32 = uitofp <vscale x 16 x i8> undef to <vscale x 16 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %nxv16i8_nxv16f64 = uitofp <vscale x 16 x i8> undef to <vscale x 16 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv16i16_nxv16f32 = uitofp <vscale x 16 x i16> undef to <vscale x 16 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %nxv16i16_nxv16f64 = uitofp <vscale x 16 x i16> undef to <vscale x 16 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv16i32_nxv16f32 = uitofp <vscale x 16 x i32> undef to <vscale x 16 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %nxv16i32_nxv16f64 = uitofp <vscale x 16 x i32> undef to <vscale x 16 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %nxv16i64_nxv16f32 = uitofp <vscale x 16 x i64> undef to <vscale x 16 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv16i64_nxv16f64 = uitofp <vscale x 16 x i64> undef to <vscale x 16 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %nxv16i1_nxv16f32 = uitofp <vscale x 16 x i1> undef to <vscale x 16 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %nxv16i1_nxv16f64 = uitofp <vscale x 16 x i1> undef to <vscale x 16 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %vp_nxv16fi8_nxv16f32 = call <vscale x 16 x float> @llvm.vp.uitofp.nxv16f32.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_nxv16fi8_nxv16f64 = call <vscale x 16 x double> @llvm.vp.uitofp.nxv16f64.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv16fi16_nxv16f32 = call <vscale x 16 x float> @llvm.vp.uitofp.nxv16f32.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_nxv16fi16_nxv16f64 = call <vscale x 16 x double> @llvm.vp.uitofp.nxv16f64.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv16fi32_nxv16f32 = call <vscale x 16 x float> @llvm.vp.uitofp.nxv16f32.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_nxv16fi32_nxv16f64 = call <vscale x 16 x double> @llvm.vp.uitofp.nxv16f64.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %vp_nxv16fi64_nxv16f32 = call <vscale x 16 x float> @llvm.vp.uitofp.nxv16f32.nxv16i64(<vscale x 16 x i64> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv16fi64_nxv16f64 = call <vscale x 16 x double> @llvm.vp.uitofp.nxv16f64.nxv16i64(<vscale x 16 x i64> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %vp_nxv16fi1_nxv16f32 = call <vscale x 16 x float> @llvm.vp.uitofp.nxv16f32.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %vp_nxv16fi1_nxv16f64 = call <vscale x 16 x double> @llvm.vp.uitofp.nxv16f64.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %nxv32i8_nxv32f32 = uitofp <vscale x 32 x i8> undef to <vscale x 32 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %nxv32i8_nxv32f64 = uitofp <vscale x 32 x i8> undef to <vscale x 32 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %nxv32i16_nxv32f32 = uitofp <vscale x 32 x i16> undef to <vscale x 32 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %nxv32i16_nxv32f64 = uitofp <vscale x 32 x i16> undef to <vscale x 32 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %nxv32i32_nxv32f32 = uitofp <vscale x 32 x i32> undef to <vscale x 32 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %nxv32i32_nxv32f64 = uitofp <vscale x 32 x i32> undef to <vscale x 32 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %nxv32i64_nxv32f32 = uitofp <vscale x 32 x i64> undef to <vscale x 32 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv32i64_nxv32f64 = uitofp <vscale x 32 x i64> undef to <vscale x 32 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %nxv32i1_nxv32f32 = uitofp <vscale x 32 x i1> undef to <vscale x 32 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %nxv32i1_nxv32f64 = uitofp <vscale x 32 x i1> undef to <vscale x 32 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %vp_nxv32fi8_nxv32f32 = call <vscale x 32 x float> @llvm.vp.uitofp.nxv32f32.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %vp_nxv32fi8_nxv32f64 = call <vscale x 32 x double> @llvm.vp.uitofp.nxv32f64.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %vp_nxv32fi16_nxv32f32 = call <vscale x 32 x float> @llvm.vp.uitofp.nxv32f32.nxv32i16(<vscale x 32 x i16> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %vp_nxv32fi16_nxv32f64 = call <vscale x 32 x double> @llvm.vp.uitofp.nxv32f64.nxv32i16(<vscale x 32 x i16> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %vp_nxv32fi32_nxv32f32 = call <vscale x 32 x float> @llvm.vp.uitofp.nxv32f32.nxv32i32(<vscale x 32 x i32> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_nxv32fi32_nxv32f64 = call <vscale x 32 x double> @llvm.vp.uitofp.nxv32f64.nxv32i32(<vscale x 32 x i32> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %vp_nxv32fi64_nxv32f32 = call <vscale x 32 x float> @llvm.vp.uitofp.nxv32f32.nxv32i64(<vscale x 32 x i64> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv32fi64_nxv32f64 = call <vscale x 32 x double> @llvm.vp.uitofp.nxv32f64.nxv32i64(<vscale x 32 x i64> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %vp_nxv32fi1_nxv32f32 = call <vscale x 32 x float> @llvm.vp.uitofp.nxv32f32.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %vp_nxv32fi1_nxv32f64 = call <vscale x 32 x double> @llvm.vp.uitofp.nxv32f64.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %nxv64i8_nxv64f32 = uitofp <vscale x 64 x i8> undef to <vscale x 64 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 103 for instruction: %nxv64i8_nxv64f64 = uitofp <vscale x 64 x i8> undef to <vscale x 64 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %nxv64i16_nxv64f32 = uitofp <vscale x 64 x i16> undef to <vscale x 64 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 102 for instruction: %nxv64i16_nxv64f64 = uitofp <vscale x 64 x i16> undef to <vscale x 64 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %nxv64i32_nxv64f32 = uitofp <vscale x 64 x i32> undef to <vscale x 64 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %nxv64i32_nxv64f64 = uitofp <vscale x 64 x i32> undef to <vscale x 64 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %nxv64i64_nxv64f32 = uitofp <vscale x 64 x i64> undef to <vscale x 64 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %nxv64i64_nxv64f64 = uitofp <vscale x 64 x i64> undef to <vscale x 64 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %nxv64i1_nxv64f32 = uitofp <vscale x 64 x i1> undef to <vscale x 64 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 135 for instruction: %nxv64i1_nxv64f64 = uitofp <vscale x 64 x i1> undef to <vscale x 64 x double>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %vp_nxv64fi8_nxv64f32 = call <vscale x 64 x float> @llvm.vp.uitofp.nxv64f32.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 103 for instruction: %vp_nxv64fi8_nxv64f64 = call <vscale x 64 x double> @llvm.vp.uitofp.nxv64f64.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %vp_nxv64fi16_nxv64f32 = call <vscale x 64 x float> @llvm.vp.uitofp.nxv64f32.nxv64i16(<vscale x 64 x i16> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 102 for instruction: %vp_nxv64fi16_nxv64f64 = call <vscale x 64 x double> @llvm.vp.uitofp.nxv64f64.nxv64i16(<vscale x 64 x i16> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %vp_nxv64fi32_nxv64f32 = call <vscale x 64 x float> @llvm.vp.uitofp.nxv64f32.nxv64i32(<vscale x 64 x i32> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %vp_nxv64fi32_nxv64f64 = call <vscale x 64 x double> @llvm.vp.uitofp.nxv64f64.nxv64i32(<vscale x 64 x i32> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %vp_nxv64fi64_nxv64f32 = call <vscale x 64 x float> @llvm.vp.uitofp.nxv64f32.nxv64i64(<vscale x 64 x i64> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %vp_nxv64fi64_nxv64f64 = call <vscale x 64 x double> @llvm.vp.uitofp.nxv64f64.nxv64i64(<vscale x 64 x i64> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %vp_nxv64fi1_nxv64f32 = call <vscale x 64 x float> @llvm.vp.uitofp.nxv64f32.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 135 for instruction: %vp_nxv64fi1_nxv64f64 = call <vscale x 64 x double> @llvm.vp.uitofp.nxv64f64.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; ZVE64X-LABEL: 'uitofp'
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2i8_v2f32 = uitofp <2 x i8> undef to <2 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2i8_v2f64 = uitofp <2 x i8> undef to <2 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2i16_v2f32 = uitofp <2 x i16> undef to <2 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2i16_v2f64 = uitofp <2 x i16> undef to <2 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2i32_v2f32 = uitofp <2 x i32> undef to <2 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2i32_v2f64 = uitofp <2 x i32> undef to <2 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2i64_v2f32 = uitofp <2 x i64> undef to <2 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2i64_v2f64 = uitofp <2 x i64> undef to <2 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2i1_v2f32 = uitofp <2 x i1> undef to <2 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2i1_v2f64 = uitofp <2 x i1> undef to <2 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2fi8_v2f32 = call <2 x float> @llvm.vp.uitofp.v2f32.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2fi8_v2f64 = call <2 x double> @llvm.vp.uitofp.v2f64.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2fi16_v2f32 = call <2 x float> @llvm.vp.uitofp.v2f32.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2fi16_v2f64 = call <2 x double> @llvm.vp.uitofp.v2f64.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2fi32_v2f32 = call <2 x float> @llvm.vp.uitofp.v2f32.v2i32(<2 x i32> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2fi32_v2f64 = call <2 x double> @llvm.vp.uitofp.v2f64.v2i32(<2 x i32> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2fi64_v2f32 = call <2 x float> @llvm.vp.uitofp.v2f32.v2i64(<2 x i64> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2fi64_v2f64 = call <2 x double> @llvm.vp.uitofp.v2f64.v2i64(<2 x i64> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2fi1_v2f32 = call <2 x float> @llvm.vp.uitofp.v2f32.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %vp_v2fi1_v2f64 = call <2 x double> @llvm.vp.uitofp.v2f64.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4i8_v4f32 = uitofp <4 x i8> undef to <4 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4i8_v4f64 = uitofp <4 x i8> undef to <4 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4i16_v4f32 = uitofp <4 x i16> undef to <4 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4i16_v4f64 = uitofp <4 x i16> undef to <4 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4i32_v4f32 = uitofp <4 x i32> undef to <4 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4i32_v4f64 = uitofp <4 x i32> undef to <4 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4i64_v4f32 = uitofp <4 x i64> undef to <4 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4i64_v4f64 = uitofp <4 x i64> undef to <4 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4i1_v4f32 = uitofp <4 x i1> undef to <4 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4i1_v4f64 = uitofp <4 x i1> undef to <4 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v4fi8_v4f32 = call <4 x float> @llvm.vp.uitofp.v4f32.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v4fi8_v4f64 = call <4 x double> @llvm.vp.uitofp.v4f64.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v4fi16_v4f32 = call <4 x float> @llvm.vp.uitofp.v4f32.v4i16(<4 x i16> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v4fi16_v4f64 = call <4 x double> @llvm.vp.uitofp.v4f64.v4i16(<4 x i16> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v4fi32_v4f32 = call <4 x float> @llvm.vp.uitofp.v4f32.v4i32(<4 x i32> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v4fi32_v4f64 = call <4 x double> @llvm.vp.uitofp.v4f64.v4i32(<4 x i32> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v4fi64_v4f32 = call <4 x float> @llvm.vp.uitofp.v4f32.v4i64(<4 x i64> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v4fi64_v4f64 = call <4 x double> @llvm.vp.uitofp.v4f64.v4i64(<4 x i64> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v4fi1_v4f32 = call <4 x float> @llvm.vp.uitofp.v4f32.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %vp_v4fi1_v4f64 = call <4 x double> @llvm.vp.uitofp.v4f64.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v8i8_v8f32 = uitofp <8 x i8> undef to <8 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v8i8_v8f64 = uitofp <8 x i8> undef to <8 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v8i16_v8f32 = uitofp <8 x i16> undef to <8 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v8i16_v8f64 = uitofp <8 x i16> undef to <8 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v8i32_v8f32 = uitofp <8 x i32> undef to <8 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v8i32_v8f64 = uitofp <8 x i32> undef to <8 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v8i64_v8f32 = uitofp <8 x i64> undef to <8 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v8i64_v8f64 = uitofp <8 x i64> undef to <8 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v8i1_v8f32 = uitofp <8 x i1> undef to <8 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %v8i1_v8f64 = uitofp <8 x i1> undef to <8 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v8fi8_v8f32 = call <8 x float> @llvm.vp.uitofp.v8f32.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v8fi8_v8f64 = call <8 x double> @llvm.vp.uitofp.v8f64.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v8fi16_v8f32 = call <8 x float> @llvm.vp.uitofp.v8f32.v8i16(<8 x i16> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v8fi16_v8f64 = call <8 x double> @llvm.vp.uitofp.v8f64.v8i16(<8 x i16> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v8fi32_v8f32 = call <8 x float> @llvm.vp.uitofp.v8f32.v8i32(<8 x i32> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v8fi32_v8f64 = call <8 x double> @llvm.vp.uitofp.v8f64.v8i32(<8 x i32> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v8fi64_v8f32 = call <8 x float> @llvm.vp.uitofp.v8f32.v8i64(<8 x i64> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v8fi64_v8f64 = call <8 x double> @llvm.vp.uitofp.v8f64.v8i64(<8 x i64> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v8fi1_v8f32 = call <8 x float> @llvm.vp.uitofp.v8f32.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %vp_v8fi1_v8f64 = call <8 x double> @llvm.vp.uitofp.v8f64.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v16i8_v16f32 = uitofp <16 x i8> undef to <16 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v16i8_v16f64 = uitofp <16 x i8> undef to <16 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v16i16_v16f32 = uitofp <16 x i16> undef to <16 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v16i16_v16f64 = uitofp <16 x i16> undef to <16 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v16i32_v16f32 = uitofp <16 x i32> undef to <16 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v16i32_v16f64 = uitofp <16 x i32> undef to <16 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v16i64_v16f32 = uitofp <16 x i64> undef to <16 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v16i64_v16f64 = uitofp <16 x i64> undef to <16 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v16i1_v16f32 = uitofp <16 x i1> undef to <16 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %v16i1_v16f64 = uitofp <16 x i1> undef to <16 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v16fi8_v16f32 = call <16 x float> @llvm.vp.uitofp.v16f32.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v16fi8_v16f64 = call <16 x double> @llvm.vp.uitofp.v16f64.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v16fi16_v16f32 = call <16 x float> @llvm.vp.uitofp.v16f32.v16i16(<16 x i16> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v16fi16_v16f64 = call <16 x double> @llvm.vp.uitofp.v16f64.v16i16(<16 x i16> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v16fi32_v16f32 = call <16 x float> @llvm.vp.uitofp.v16f32.v16i32(<16 x i32> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v16fi32_v16f64 = call <16 x double> @llvm.vp.uitofp.v16f64.v16i32(<16 x i32> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v16fi64_v16f32 = call <16 x float> @llvm.vp.uitofp.v16f32.v16i64(<16 x i64> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v16fi64_v16f64 = call <16 x double> @llvm.vp.uitofp.v16f64.v16i64(<16 x i64> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v16fi1_v16f32 = call <16 x float> @llvm.vp.uitofp.v16f32.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %vp_v16fi1_v16f64 = call <16 x double> @llvm.vp.uitofp.v16f64.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v32i8_v32f32 = uitofp <32 x i8> undef to <32 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v32i8_v32f64 = uitofp <32 x i8> undef to <32 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v32i16_v32f32 = uitofp <32 x i16> undef to <32 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v32i16_v32f64 = uitofp <32 x i16> undef to <32 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v32i32_v32f32 = uitofp <32 x i32> undef to <32 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v32i32_v32f64 = uitofp <32 x i32> undef to <32 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 62 for instruction: %v32i64_v32f32 = uitofp <32 x i64> undef to <32 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 62 for instruction: %v32i64_v32f64 = uitofp <32 x i64> undef to <32 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v32i1_v32f32 = uitofp <32 x i1> undef to <32 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %v32i1_v32f64 = uitofp <32 x i1> undef to <32 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v32fi8_v32f32 = call <32 x float> @llvm.vp.uitofp.v32f32.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v32fi8_v32f64 = call <32 x double> @llvm.vp.uitofp.v32f64.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v32fi16_v32f32 = call <32 x float> @llvm.vp.uitofp.v32f32.v32i16(<32 x i16> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v32fi16_v32f64 = call <32 x double> @llvm.vp.uitofp.v32f64.v32i16(<32 x i16> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v32fi32_v32f32 = call <32 x float> @llvm.vp.uitofp.v32f32.v32i32(<32 x i32> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v32fi32_v32f64 = call <32 x double> @llvm.vp.uitofp.v32f64.v32i32(<32 x i32> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 62 for instruction: %vp_v32fi64_v32f32 = call <32 x float> @llvm.vp.uitofp.v32f32.v32i64(<32 x i64> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 62 for instruction: %vp_v32fi64_v32f64 = call <32 x double> @llvm.vp.uitofp.v32f64.v32i64(<32 x i64> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v32fi1_v32f32 = call <32 x float> @llvm.vp.uitofp.v32f32.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %vp_v32fi1_v32f64 = call <32 x double> @llvm.vp.uitofp.v32f64.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %v64i8_v64f32 = uitofp <64 x i8> undef to <64 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %v64i8_v64f64 = uitofp <64 x i8> undef to <64 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %v64i16_v64f32 = uitofp <64 x i16> undef to <64 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %v64i16_v64f64 = uitofp <64 x i16> undef to <64 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 126 for instruction: %v64i32_v64f32 = uitofp <64 x i32> undef to <64 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 126 for instruction: %v64i32_v64f64 = uitofp <64 x i32> undef to <64 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 124 for instruction: %v64i64_v64f32 = uitofp <64 x i64> undef to <64 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 124 for instruction: %v64i64_v64f64 = uitofp <64 x i64> undef to <64 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %v64i1_v64f32 = uitofp <64 x i1> undef to <64 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %v64i1_v64f64 = uitofp <64 x i1> undef to <64 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %vp_v64fi8_v64f32 = call <64 x float> @llvm.vp.uitofp.v64f32.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %vp_v64fi8_v64f64 = call <64 x double> @llvm.vp.uitofp.v64f64.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %vp_v64fi16_v64f32 = call <64 x float> @llvm.vp.uitofp.v64f32.v64i16(<64 x i16> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %vp_v64fi16_v64f64 = call <64 x double> @llvm.vp.uitofp.v64f64.v64i16(<64 x i16> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 126 for instruction: %vp_v64fi32_v64f32 = call <64 x float> @llvm.vp.uitofp.v64f32.v64i32(<64 x i32> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 126 for instruction: %vp_v64fi32_v64f64 = call <64 x double> @llvm.vp.uitofp.v64f64.v64i32(<64 x i32> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 124 for instruction: %vp_v64fi64_v64f32 = call <64 x float> @llvm.vp.uitofp.v64f32.v64i64(<64 x i64> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 124 for instruction: %vp_v64fi64_v64f64 = call <64 x double> @llvm.vp.uitofp.v64f64.v64i64(<64 x i64> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %vp_v64fi1_v64f32 = call <64 x float> @llvm.vp.uitofp.v64f32.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %vp_v64fi1_v64f64 = call <64 x double> @llvm.vp.uitofp.v64f64.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %v128i8_v128f32 = uitofp <128 x i8> undef to <128 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %v128i8_v128f64 = uitofp <128 x i8> undef to <128 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 254 for instruction: %v128i16_v128f32 = uitofp <128 x i16> undef to <128 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 254 for instruction: %v128i16_v128f64 = uitofp <128 x i16> undef to <128 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 252 for instruction: %v128i32_v128f32 = uitofp <128 x i32> undef to <128 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 252 for instruction: %v128i32_v128f64 = uitofp <128 x i32> undef to <128 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 248 for instruction: %v128i64_v128f32 = uitofp <128 x i64> undef to <128 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 248 for instruction: %v128i64_v128f64 = uitofp <128 x i64> undef to <128 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %v128i1_v128f32 = uitofp <128 x i1> undef to <128 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %v128i1_v128f64 = uitofp <128 x i1> undef to <128 x double>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %vp_v128fi8_v128f32 = call <128 x float> @llvm.vp.uitofp.v128f32.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %vp_v128fi8_v128f64 = call <128 x double> @llvm.vp.uitofp.v128f64.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 254 for instruction: %vp_v128fi16_v128f32 = call <128 x float> @llvm.vp.uitofp.v128f32.v128i16(<128 x i16> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 254 for instruction: %vp_v128fi16_v128f64 = call <128 x double> @llvm.vp.uitofp.v128f64.v128i16(<128 x i16> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 252 for instruction: %vp_v128fi32_v128f32 = call <128 x float> @llvm.vp.uitofp.v128f32.v128i32(<128 x i32> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 252 for instruction: %vp_v128fi32_v128f64 = call <128 x double> @llvm.vp.uitofp.v128f64.v128i32(<128 x i32> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 248 for instruction: %vp_v128fi64_v128f32 = call <128 x float> @llvm.vp.uitofp.v128f32.v128i64(<128 x i64> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 248 for instruction: %vp_v128fi64_v128f64 = call <128 x double> @llvm.vp.uitofp.v128f64.v128i64(<128 x i64> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %vp_v128fi1_v128f32 = call <128 x float> @llvm.vp.uitofp.v128f32.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 255 for instruction: %vp_v128fi1_v128f64 = call <128 x double> @llvm.vp.uitofp.v128f64.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1i8_nxv1f32 = uitofp <vscale x 1 x i8> undef to <vscale x 1 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1i8_nxv1f64 = uitofp <vscale x 1 x i8> undef to <vscale x 1 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1i16_nxv1f32 = uitofp <vscale x 1 x i16> undef to <vscale x 1 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1i16_nxv1f64 = uitofp <vscale x 1 x i16> undef to <vscale x 1 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1i32_nxv1f32 = uitofp <vscale x 1 x i32> undef to <vscale x 1 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1i32_nxv1f64 = uitofp <vscale x 1 x i32> undef to <vscale x 1 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1i64_nxv1f32 = uitofp <vscale x 1 x i64> undef to <vscale x 1 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1i64_nxv1f64 = uitofp <vscale x 1 x i64> undef to <vscale x 1 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1i1_nxv1f32 = uitofp <vscale x 1 x i1> undef to <vscale x 1 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv1i1_nxv1f64 = uitofp <vscale x 1 x i1> undef to <vscale x 1 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1fi8_nxv1f32 = call <vscale x 1 x float> @llvm.vp.uitofp.nxv1f32.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1fi8_nxv1f64 = call <vscale x 1 x double> @llvm.vp.uitofp.nxv1f64.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1fi16_nxv1f32 = call <vscale x 1 x float> @llvm.vp.uitofp.nxv1f32.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1fi16_nxv1f64 = call <vscale x 1 x double> @llvm.vp.uitofp.nxv1f64.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1fi32_nxv1f32 = call <vscale x 1 x float> @llvm.vp.uitofp.nxv1f32.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1fi32_nxv1f64 = call <vscale x 1 x double> @llvm.vp.uitofp.nxv1f64.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1fi64_nxv1f32 = call <vscale x 1 x float> @llvm.vp.uitofp.nxv1f32.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1fi64_nxv1f64 = call <vscale x 1 x double> @llvm.vp.uitofp.nxv1f64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1fi1_nxv1f32 = call <vscale x 1 x float> @llvm.vp.uitofp.nxv1f32.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv1fi1_nxv1f64 = call <vscale x 1 x double> @llvm.vp.uitofp.nxv1f64.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2i8_nxv2f32 = uitofp <vscale x 2 x i8> undef to <vscale x 2 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2i8_nxv2f64 = uitofp <vscale x 2 x i8> undef to <vscale x 2 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2i16_nxv2f32 = uitofp <vscale x 2 x i16> undef to <vscale x 2 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2i16_nxv2f64 = uitofp <vscale x 2 x i16> undef to <vscale x 2 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2i32_nxv2f32 = uitofp <vscale x 2 x i32> undef to <vscale x 2 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2i32_nxv2f64 = uitofp <vscale x 2 x i32> undef to <vscale x 2 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2i64_nxv2f32 = uitofp <vscale x 2 x i64> undef to <vscale x 2 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2i64_nxv2f64 = uitofp <vscale x 2 x i64> undef to <vscale x 2 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2i1_nxv2f32 = uitofp <vscale x 2 x i1> undef to <vscale x 2 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv2i1_nxv2f64 = uitofp <vscale x 2 x i1> undef to <vscale x 2 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2fi8_nxv2f32 = call <vscale x 2 x float> @llvm.vp.uitofp.nxv2f32.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2fi8_nxv2f64 = call <vscale x 2 x double> @llvm.vp.uitofp.nxv2f64.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2fi16_nxv2f32 = call <vscale x 2 x float> @llvm.vp.uitofp.nxv2f32.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2fi16_nxv2f64 = call <vscale x 2 x double> @llvm.vp.uitofp.nxv2f64.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2fi32_nxv2f32 = call <vscale x 2 x float> @llvm.vp.uitofp.nxv2f32.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2fi32_nxv2f64 = call <vscale x 2 x double> @llvm.vp.uitofp.nxv2f64.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2fi64_nxv2f32 = call <vscale x 2 x float> @llvm.vp.uitofp.nxv2f32.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2fi64_nxv2f64 = call <vscale x 2 x double> @llvm.vp.uitofp.nxv2f64.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2fi1_nxv2f32 = call <vscale x 2 x float> @llvm.vp.uitofp.nxv2f32.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv2fi1_nxv2f64 = call <vscale x 2 x double> @llvm.vp.uitofp.nxv2f64.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4i8_nxv4f32 = uitofp <vscale x 4 x i8> undef to <vscale x 4 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4i8_nxv4f64 = uitofp <vscale x 4 x i8> undef to <vscale x 4 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4i16_nxv4f32 = uitofp <vscale x 4 x i16> undef to <vscale x 4 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4i16_nxv4f64 = uitofp <vscale x 4 x i16> undef to <vscale x 4 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4i32_nxv4f32 = uitofp <vscale x 4 x i32> undef to <vscale x 4 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4i32_nxv4f64 = uitofp <vscale x 4 x i32> undef to <vscale x 4 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4i64_nxv4f32 = uitofp <vscale x 4 x i64> undef to <vscale x 4 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4i64_nxv4f64 = uitofp <vscale x 4 x i64> undef to <vscale x 4 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4i1_nxv4f32 = uitofp <vscale x 4 x i1> undef to <vscale x 4 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv4i1_nxv4f64 = uitofp <vscale x 4 x i1> undef to <vscale x 4 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4fi8_nxv4f32 = call <vscale x 4 x float> @llvm.vp.uitofp.nxv4f32.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4fi8_nxv4f64 = call <vscale x 4 x double> @llvm.vp.uitofp.nxv4f64.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4fi16_nxv4f32 = call <vscale x 4 x float> @llvm.vp.uitofp.nxv4f32.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4fi16_nxv4f64 = call <vscale x 4 x double> @llvm.vp.uitofp.nxv4f64.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4fi32_nxv4f32 = call <vscale x 4 x float> @llvm.vp.uitofp.nxv4f32.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4fi32_nxv4f64 = call <vscale x 4 x double> @llvm.vp.uitofp.nxv4f64.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4fi64_nxv4f32 = call <vscale x 4 x float> @llvm.vp.uitofp.nxv4f32.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4fi64_nxv4f64 = call <vscale x 4 x double> @llvm.vp.uitofp.nxv4f64.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4fi1_nxv4f32 = call <vscale x 4 x float> @llvm.vp.uitofp.nxv4f32.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv4fi1_nxv4f64 = call <vscale x 4 x double> @llvm.vp.uitofp.nxv4f64.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8i8_nxv8f32 = uitofp <vscale x 8 x i8> undef to <vscale x 8 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8i8_nxv8f64 = uitofp <vscale x 8 x i8> undef to <vscale x 8 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8i16_nxv8f32 = uitofp <vscale x 8 x i16> undef to <vscale x 8 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8i16_nxv8f64 = uitofp <vscale x 8 x i16> undef to <vscale x 8 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8i32_nxv8f32 = uitofp <vscale x 8 x i32> undef to <vscale x 8 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8i32_nxv8f64 = uitofp <vscale x 8 x i32> undef to <vscale x 8 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8i64_nxv8f32 = uitofp <vscale x 8 x i64> undef to <vscale x 8 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8i64_nxv8f64 = uitofp <vscale x 8 x i64> undef to <vscale x 8 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8i1_nxv8f32 = uitofp <vscale x 8 x i1> undef to <vscale x 8 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv8i1_nxv8f64 = uitofp <vscale x 8 x i1> undef to <vscale x 8 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8fi8_nxv8f32 = call <vscale x 8 x float> @llvm.vp.uitofp.nxv8f32.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8fi8_nxv8f64 = call <vscale x 8 x double> @llvm.vp.uitofp.nxv8f64.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8fi16_nxv8f32 = call <vscale x 8 x float> @llvm.vp.uitofp.nxv8f32.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8fi16_nxv8f64 = call <vscale x 8 x double> @llvm.vp.uitofp.nxv8f64.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8fi32_nxv8f32 = call <vscale x 8 x float> @llvm.vp.uitofp.nxv8f32.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8fi32_nxv8f64 = call <vscale x 8 x double> @llvm.vp.uitofp.nxv8f64.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8fi64_nxv8f32 = call <vscale x 8 x float> @llvm.vp.uitofp.nxv8f32.nxv8i64(<vscale x 8 x i64> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8fi64_nxv8f64 = call <vscale x 8 x double> @llvm.vp.uitofp.nxv8f64.nxv8i64(<vscale x 8 x i64> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8fi1_nxv8f32 = call <vscale x 8 x float> @llvm.vp.uitofp.nxv8f32.nxv8i1(<vscale x 8 x i1> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv8fi1_nxv8f64 = call <vscale x 8 x double> @llvm.vp.uitofp.nxv8f64.nxv8i1(<vscale x 8 x i1> undef, <vscale x 8 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16i8_nxv16f32 = uitofp <vscale x 16 x i8> undef to <vscale x 16 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16i8_nxv16f64 = uitofp <vscale x 16 x i8> undef to <vscale x 16 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16i16_nxv16f32 = uitofp <vscale x 16 x i16> undef to <vscale x 16 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16i16_nxv16f64 = uitofp <vscale x 16 x i16> undef to <vscale x 16 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16i32_nxv16f32 = uitofp <vscale x 16 x i32> undef to <vscale x 16 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16i32_nxv16f64 = uitofp <vscale x 16 x i32> undef to <vscale x 16 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16i64_nxv16f32 = uitofp <vscale x 16 x i64> undef to <vscale x 16 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16i64_nxv16f64 = uitofp <vscale x 16 x i64> undef to <vscale x 16 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16i1_nxv16f32 = uitofp <vscale x 16 x i1> undef to <vscale x 16 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv16i1_nxv16f64 = uitofp <vscale x 16 x i1> undef to <vscale x 16 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16fi8_nxv16f32 = call <vscale x 16 x float> @llvm.vp.uitofp.nxv16f32.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16fi8_nxv16f64 = call <vscale x 16 x double> @llvm.vp.uitofp.nxv16f64.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16fi16_nxv16f32 = call <vscale x 16 x float> @llvm.vp.uitofp.nxv16f32.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16fi16_nxv16f64 = call <vscale x 16 x double> @llvm.vp.uitofp.nxv16f64.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16fi32_nxv16f32 = call <vscale x 16 x float> @llvm.vp.uitofp.nxv16f32.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16fi32_nxv16f64 = call <vscale x 16 x double> @llvm.vp.uitofp.nxv16f64.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16fi64_nxv16f32 = call <vscale x 16 x float> @llvm.vp.uitofp.nxv16f32.nxv16i64(<vscale x 16 x i64> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16fi64_nxv16f64 = call <vscale x 16 x double> @llvm.vp.uitofp.nxv16f64.nxv16i64(<vscale x 16 x i64> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16fi1_nxv16f32 = call <vscale x 16 x float> @llvm.vp.uitofp.nxv16f32.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv16fi1_nxv16f64 = call <vscale x 16 x double> @llvm.vp.uitofp.nxv16f64.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32i8_nxv32f32 = uitofp <vscale x 32 x i8> undef to <vscale x 32 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32i8_nxv32f64 = uitofp <vscale x 32 x i8> undef to <vscale x 32 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32i16_nxv32f32 = uitofp <vscale x 32 x i16> undef to <vscale x 32 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32i16_nxv32f64 = uitofp <vscale x 32 x i16> undef to <vscale x 32 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32i32_nxv32f32 = uitofp <vscale x 32 x i32> undef to <vscale x 32 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32i32_nxv32f64 = uitofp <vscale x 32 x i32> undef to <vscale x 32 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32i64_nxv32f32 = uitofp <vscale x 32 x i64> undef to <vscale x 32 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32i64_nxv32f64 = uitofp <vscale x 32 x i64> undef to <vscale x 32 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32i1_nxv32f32 = uitofp <vscale x 32 x i1> undef to <vscale x 32 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv32i1_nxv32f64 = uitofp <vscale x 32 x i1> undef to <vscale x 32 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32fi8_nxv32f32 = call <vscale x 32 x float> @llvm.vp.uitofp.nxv32f32.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32fi8_nxv32f64 = call <vscale x 32 x double> @llvm.vp.uitofp.nxv32f64.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32fi16_nxv32f32 = call <vscale x 32 x float> @llvm.vp.uitofp.nxv32f32.nxv32i16(<vscale x 32 x i16> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32fi16_nxv32f64 = call <vscale x 32 x double> @llvm.vp.uitofp.nxv32f64.nxv32i16(<vscale x 32 x i16> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32fi32_nxv32f32 = call <vscale x 32 x float> @llvm.vp.uitofp.nxv32f32.nxv32i32(<vscale x 32 x i32> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32fi32_nxv32f64 = call <vscale x 32 x double> @llvm.vp.uitofp.nxv32f64.nxv32i32(<vscale x 32 x i32> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32fi64_nxv32f32 = call <vscale x 32 x float> @llvm.vp.uitofp.nxv32f32.nxv32i64(<vscale x 32 x i64> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32fi64_nxv32f64 = call <vscale x 32 x double> @llvm.vp.uitofp.nxv32f64.nxv32i64(<vscale x 32 x i64> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32fi1_nxv32f32 = call <vscale x 32 x float> @llvm.vp.uitofp.nxv32f32.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv32fi1_nxv32f64 = call <vscale x 32 x double> @llvm.vp.uitofp.nxv32f64.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64i8_nxv64f32 = uitofp <vscale x 64 x i8> undef to <vscale x 64 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64i8_nxv64f64 = uitofp <vscale x 64 x i8> undef to <vscale x 64 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64i16_nxv64f32 = uitofp <vscale x 64 x i16> undef to <vscale x 64 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64i16_nxv64f64 = uitofp <vscale x 64 x i16> undef to <vscale x 64 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64i32_nxv64f32 = uitofp <vscale x 64 x i32> undef to <vscale x 64 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64i32_nxv64f64 = uitofp <vscale x 64 x i32> undef to <vscale x 64 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64i64_nxv64f32 = uitofp <vscale x 64 x i64> undef to <vscale x 64 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64i64_nxv64f64 = uitofp <vscale x 64 x i64> undef to <vscale x 64 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64i1_nxv64f32 = uitofp <vscale x 64 x i1> undef to <vscale x 64 x float>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %nxv64i1_nxv64f64 = uitofp <vscale x 64 x i1> undef to <vscale x 64 x double>
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64fi8_nxv64f32 = call <vscale x 64 x float> @llvm.vp.uitofp.nxv64f32.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64fi8_nxv64f64 = call <vscale x 64 x double> @llvm.vp.uitofp.nxv64f64.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64fi16_nxv64f32 = call <vscale x 64 x float> @llvm.vp.uitofp.nxv64f32.nxv64i16(<vscale x 64 x i16> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64fi16_nxv64f64 = call <vscale x 64 x double> @llvm.vp.uitofp.nxv64f64.nxv64i16(<vscale x 64 x i16> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64fi32_nxv64f32 = call <vscale x 64 x float> @llvm.vp.uitofp.nxv64f32.nxv64i32(<vscale x 64 x i32> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64fi32_nxv64f64 = call <vscale x 64 x double> @llvm.vp.uitofp.nxv64f64.nxv64i32(<vscale x 64 x i32> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64fi64_nxv64f32 = call <vscale x 64 x float> @llvm.vp.uitofp.nxv64f32.nxv64i64(<vscale x 64 x i64> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64fi64_nxv64f64 = call <vscale x 64 x double> @llvm.vp.uitofp.nxv64f64.nxv64i64(<vscale x 64 x i64> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64fi1_nxv64f32 = call <vscale x 64 x float> @llvm.vp.uitofp.nxv64f32.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Invalid cost for instruction: %vp_nxv64fi1_nxv64f64 = call <vscale x 64 x double> @llvm.vp.uitofp.nxv64f64.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  %v2i8_v2f32 = uitofp <2 x i8> undef to <2 x float>
  %v2i8_v2f64 = uitofp <2 x i8> undef to <2 x double>
  %v2i16_v2f32 = uitofp <2 x i16> undef to <2 x float>
  %v2i16_v2f64 = uitofp <2 x i16> undef to <2 x double>
  %v2i32_v2f32 = uitofp <2 x i32> undef to <2 x float>
  %v2i32_v2f64 = uitofp <2 x i32> undef to <2 x double>
  %v2i64_v2f32 = uitofp <2 x i64> undef to <2 x float>
  %v2i64_v2f64 = uitofp <2 x i64> undef to <2 x double>
  %v2i1_v2f32 = uitofp <2 x i1> undef to <2 x float>
  %v2i1_v2f64 = uitofp <2 x i1> undef to <2 x double>

  %vp_v2fi8_v2f32 = call <2 x float> @llvm.vp.uitofp.v2f32.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
  %vp_v2fi8_v2f64 = call <2 x double> @llvm.vp.uitofp.v2f64.v2i8(<2 x i8> undef, <2 x i1> undef, i32 undef)
  %vp_v2fi16_v2f32 = call <2 x float> @llvm.vp.uitofp.v2f32.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
  %vp_v2fi16_v2f64 = call <2 x double> @llvm.vp.uitofp.v2f64.v2i16(<2 x i16> undef, <2 x i1> undef, i32 undef)
  %vp_v2fi32_v2f32 = call <2 x float> @llvm.vp.uitofp.v2f32.v2i32(<2 x i32> undef, <2 x i1> undef, i32 undef)
  %vp_v2fi32_v2f64 = call <2 x double> @llvm.vp.uitofp.v2f64.v2i32(<2 x i32> undef, <2 x i1> undef, i32 undef)
  %vp_v2fi64_v2f32 = call <2 x float> @llvm.vp.uitofp.v2f32.v2i64(<2 x i64> undef, <2 x i1> undef, i32 undef)
  %vp_v2fi64_v2f64 = call <2 x double> @llvm.vp.uitofp.v2f64.v2i64(<2 x i64> undef, <2 x i1> undef, i32 undef)
  %vp_v2fi1_v2f32 = call <2 x float> @llvm.vp.uitofp.v2f32.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)
  %vp_v2fi1_v2f64 = call <2 x double> @llvm.vp.uitofp.v2f64.v2i1(<2 x i1> undef, <2 x i1> undef, i32 undef)

  %v4i8_v4f32 = uitofp <4 x i8> undef to <4 x float>
  %v4i8_v4f64 = uitofp <4 x i8> undef to <4 x double>
  %v4i16_v4f32 = uitofp <4 x i16> undef to <4 x float>
  %v4i16_v4f64 = uitofp <4 x i16> undef to <4 x double>
  %v4i32_v4f32 = uitofp <4 x i32> undef to <4 x float>
  %v4i32_v4f64 = uitofp <4 x i32> undef to <4 x double>
  %v4i64_v4f32 = uitofp <4 x i64> undef to <4 x float>
  %v4i64_v4f64 = uitofp <4 x i64> undef to <4 x double>
  %v4i1_v4f32 = uitofp <4 x i1> undef to <4 x float>
  %v4i1_v4f64 = uitofp <4 x i1> undef to <4 x double>

  %vp_v4fi8_v4f32 = call <4 x float> @llvm.vp.uitofp.v4f32.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
  %vp_v4fi8_v4f64 = call <4 x double> @llvm.vp.uitofp.v4f64.v4i8(<4 x i8> undef, <4 x i1> undef, i32 undef)
  %vp_v4fi16_v4f32 = call <4 x float> @llvm.vp.uitofp.v4f32.v4i16(<4 x i16> undef, <4 x i1> undef, i32 undef)
  %vp_v4fi16_v4f64 = call <4 x double> @llvm.vp.uitofp.v4f64.v4i16(<4 x i16> undef, <4 x i1> undef, i32 undef)
  %vp_v4fi32_v4f32 = call <4 x float> @llvm.vp.uitofp.v4f32.v4i32(<4 x i32> undef, <4 x i1> undef, i32 undef)
  %vp_v4fi32_v4f64 = call <4 x double> @llvm.vp.uitofp.v4f64.v4i32(<4 x i32> undef, <4 x i1> undef, i32 undef)
  %vp_v4fi64_v4f32 = call <4 x float> @llvm.vp.uitofp.v4f32.v4i64(<4 x i64> undef, <4 x i1> undef, i32 undef)
  %vp_v4fi64_v4f64 = call <4 x double> @llvm.vp.uitofp.v4f64.v4i64(<4 x i64> undef, <4 x i1> undef, i32 undef)
  %vp_v4fi1_v4f32 = call <4 x float> @llvm.vp.uitofp.v4f32.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)
  %vp_v4fi1_v4f64 = call <4 x double> @llvm.vp.uitofp.v4f64.v4i1(<4 x i1> undef, <4 x i1> undef, i32 undef)

  %v8i8_v8f32 = uitofp <8 x i8> undef to <8 x float>
  %v8i8_v8f64 = uitofp <8 x i8> undef to <8 x double>
  %v8i16_v8f32 = uitofp <8 x i16> undef to <8 x float>
  %v8i16_v8f64 = uitofp <8 x i16> undef to <8 x double>
  %v8i32_v8f32 = uitofp <8 x i32> undef to <8 x float>
  %v8i32_v8f64 = uitofp <8 x i32> undef to <8 x double>
  %v8i64_v8f32 = uitofp <8 x i64> undef to <8 x float>
  %v8i64_v8f64 = uitofp <8 x i64> undef to <8 x double>
  %v8i1_v8f32 = uitofp <8 x i1> undef to <8 x float>
  %v8i1_v8f64 = uitofp <8 x i1> undef to <8 x double>

  %vp_v8fi8_v8f32 = call <8 x float> @llvm.vp.uitofp.v8f32.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
  %vp_v8fi8_v8f64 = call <8 x double> @llvm.vp.uitofp.v8f64.v8i8(<8 x i8> undef, <8 x i1> undef, i32 undef)
  %vp_v8fi16_v8f32 = call <8 x float> @llvm.vp.uitofp.v8f32.v8i16(<8 x i16> undef, <8 x i1> undef, i32 undef)
  %vp_v8fi16_v8f64 = call <8 x double> @llvm.vp.uitofp.v8f64.v8i16(<8 x i16> undef, <8 x i1> undef, i32 undef)
  %vp_v8fi32_v8f32 = call <8 x float> @llvm.vp.uitofp.v8f32.v8i32(<8 x i32> undef, <8 x i1> undef, i32 undef)
  %vp_v8fi32_v8f64 = call <8 x double> @llvm.vp.uitofp.v8f64.v8i32(<8 x i32> undef, <8 x i1> undef, i32 undef)
  %vp_v8fi64_v8f32 = call <8 x float> @llvm.vp.uitofp.v8f32.v8i64(<8 x i64> undef, <8 x i1> undef, i32 undef)
  %vp_v8fi64_v8f64 = call <8 x double> @llvm.vp.uitofp.v8f64.v8i64(<8 x i64> undef, <8 x i1> undef, i32 undef)
  %vp_v8fi1_v8f32 = call <8 x float> @llvm.vp.uitofp.v8f32.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)
  %vp_v8fi1_v8f64 = call <8 x double> @llvm.vp.uitofp.v8f64.v8i1(<8 x i1> undef, <8 x i1> undef, i32 undef)

  %v16i8_v16f32 = uitofp <16 x i8> undef to <16 x float>
  %v16i8_v16f64 = uitofp <16 x i8> undef to <16 x double>
  %v16i16_v16f32 = uitofp <16 x i16> undef to <16 x float>
  %v16i16_v16f64 = uitofp <16 x i16> undef to <16 x double>
  %v16i32_v16f32 = uitofp <16 x i32> undef to <16 x float>
  %v16i32_v16f64 = uitofp <16 x i32> undef to <16 x double>
  %v16i64_v16f32 = uitofp <16 x i64> undef to <16 x float>
  %v16i64_v16f64 = uitofp <16 x i64> undef to <16 x double>
  %v16i1_v16f32 = uitofp <16 x i1> undef to <16 x float>
  %v16i1_v16f64 = uitofp <16 x i1> undef to <16 x double>

  %vp_v16fi8_v16f32 = call <16 x float> @llvm.vp.uitofp.v16f32.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
  %vp_v16fi8_v16f64 = call <16 x double> @llvm.vp.uitofp.v16f64.v16i8(<16 x i8> undef, <16 x i1> undef, i32 undef)
  %vp_v16fi16_v16f32 = call <16 x float> @llvm.vp.uitofp.v16f32.v16i16(<16 x i16> undef, <16 x i1> undef, i32 undef)
  %vp_v16fi16_v16f64 = call <16 x double> @llvm.vp.uitofp.v16f64.v16i16(<16 x i16> undef, <16 x i1> undef, i32 undef)
  %vp_v16fi32_v16f32 = call <16 x float> @llvm.vp.uitofp.v16f32.v16i32(<16 x i32> undef, <16 x i1> undef, i32 undef)
  %vp_v16fi32_v16f64 = call <16 x double> @llvm.vp.uitofp.v16f64.v16i32(<16 x i32> undef, <16 x i1> undef, i32 undef)
  %vp_v16fi64_v16f32 = call <16 x float> @llvm.vp.uitofp.v16f32.v16i64(<16 x i64> undef, <16 x i1> undef, i32 undef)
  %vp_v16fi64_v16f64 = call <16 x double> @llvm.vp.uitofp.v16f64.v16i64(<16 x i64> undef, <16 x i1> undef, i32 undef)
  %vp_v16fi1_v16f32 = call <16 x float> @llvm.vp.uitofp.v16f32.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)
  %vp_v16fi1_v16f64 = call <16 x double> @llvm.vp.uitofp.v16f64.v16i1(<16 x i1> undef, <16 x i1> undef, i32 undef)

  %v32i8_v32f32 = uitofp <32 x i8> undef to <32 x float>
  %v32i8_v32f64 = uitofp <32 x i8> undef to <32 x double>
  %v32i16_v32f32 = uitofp <32 x i16> undef to <32 x float>
  %v32i16_v32f64 = uitofp <32 x i16> undef to <32 x double>
  %v32i32_v32f32 = uitofp <32 x i32> undef to <32 x float>
  %v32i32_v32f64 = uitofp <32 x i32> undef to <32 x double>
  %v32i64_v32f32 = uitofp <32 x i64> undef to <32 x float>
  %v32i64_v32f64 = uitofp <32 x i64> undef to <32 x double>
  %v32i1_v32f32 = uitofp <32 x i1> undef to <32 x float>
  %v32i1_v32f64 = uitofp <32 x i1> undef to <32 x double>

  %vp_v32fi8_v32f32 = call <32 x float> @llvm.vp.uitofp.v32f32.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
  %vp_v32fi8_v32f64 = call <32 x double> @llvm.vp.uitofp.v32f64.v32i8(<32 x i8> undef, <32 x i1> undef, i32 undef)
  %vp_v32fi16_v32f32 = call <32 x float> @llvm.vp.uitofp.v32f32.v32i16(<32 x i16> undef, <32 x i1> undef, i32 undef)
  %vp_v32fi16_v32f64 = call <32 x double> @llvm.vp.uitofp.v32f64.v32i16(<32 x i16> undef, <32 x i1> undef, i32 undef)
  %vp_v32fi32_v32f32 = call <32 x float> @llvm.vp.uitofp.v32f32.v32i32(<32 x i32> undef, <32 x i1> undef, i32 undef)
  %vp_v32fi32_v32f64 = call <32 x double> @llvm.vp.uitofp.v32f64.v32i32(<32 x i32> undef, <32 x i1> undef, i32 undef)
  %vp_v32fi64_v32f32 = call <32 x float> @llvm.vp.uitofp.v32f32.v32i64(<32 x i64> undef, <32 x i1> undef, i32 undef)
  %vp_v32fi64_v32f64 = call <32 x double> @llvm.vp.uitofp.v32f64.v32i64(<32 x i64> undef, <32 x i1> undef, i32 undef)
  %vp_v32fi1_v32f32 = call <32 x float> @llvm.vp.uitofp.v32f32.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)
  %vp_v32fi1_v32f64 = call <32 x double> @llvm.vp.uitofp.v32f64.v32i1(<32 x i1> undef, <32 x i1> undef, i32 undef)

  %v64i8_v64f32 = uitofp <64 x i8> undef to <64 x float>
  %v64i8_v64f64 = uitofp <64 x i8> undef to <64 x double>
  %v64i16_v64f32 = uitofp <64 x i16> undef to <64 x float>
  %v64i16_v64f64 = uitofp <64 x i16> undef to <64 x double>
  %v64i32_v64f32 = uitofp <64 x i32> undef to <64 x float>
  %v64i32_v64f64 = uitofp <64 x i32> undef to <64 x double>
  %v64i64_v64f32 = uitofp <64 x i64> undef to <64 x float>
  %v64i64_v64f64 = uitofp <64 x i64> undef to <64 x double>
  %v64i1_v64f32 = uitofp <64 x i1> undef to <64 x float>
  %v64i1_v64f64 = uitofp <64 x i1> undef to <64 x double>

  %vp_v64fi8_v64f32 = call <64 x float> @llvm.vp.uitofp.v64f32.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
  %vp_v64fi8_v64f64 = call <64 x double> @llvm.vp.uitofp.v64f64.v64i8(<64 x i8> undef, <64 x i1> undef, i32 undef)
  %vp_v64fi16_v64f32 = call <64 x float> @llvm.vp.uitofp.v64f32.v64i16(<64 x i16> undef, <64 x i1> undef, i32 undef)
  %vp_v64fi16_v64f64 = call <64 x double> @llvm.vp.uitofp.v64f64.v64i16(<64 x i16> undef, <64 x i1> undef, i32 undef)
  %vp_v64fi32_v64f32 = call <64 x float> @llvm.vp.uitofp.v64f32.v64i32(<64 x i32> undef, <64 x i1> undef, i32 undef)
  %vp_v64fi32_v64f64 = call <64 x double> @llvm.vp.uitofp.v64f64.v64i32(<64 x i32> undef, <64 x i1> undef, i32 undef)
  %vp_v64fi64_v64f32 = call <64 x float> @llvm.vp.uitofp.v64f32.v64i64(<64 x i64> undef, <64 x i1> undef, i32 undef)
  %vp_v64fi64_v64f64 = call <64 x double> @llvm.vp.uitofp.v64f64.v64i64(<64 x i64> undef, <64 x i1> undef, i32 undef)
  %vp_v64fi1_v64f32 = call <64 x float> @llvm.vp.uitofp.v64f32.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)
  %vp_v64fi1_v64f64 = call <64 x double> @llvm.vp.uitofp.v64f64.v64i1(<64 x i1> undef, <64 x i1> undef, i32 undef)

  %v128i8_v128f32 = uitofp <128 x i8> undef to <128 x float>
  %v128i8_v128f64 = uitofp <128 x i8> undef to <128 x double>
  %v128i16_v128f32 = uitofp <128 x i16> undef to <128 x float>
  %v128i16_v128f64 = uitofp <128 x i16> undef to <128 x double>
  %v128i32_v128f32 = uitofp <128 x i32> undef to <128 x float>
  %v128i32_v128f64 = uitofp <128 x i32> undef to <128 x double>
  %v128i64_v128f32 = uitofp <128 x i64> undef to <128 x float>
  %v128i64_v128f64 = uitofp <128 x i64> undef to <128 x double>
  %v128i1_v128f32 = uitofp <128 x i1> undef to <128 x float>
  %v128i1_v128f64 = uitofp <128 x i1> undef to <128 x double>

  %vp_v128fi8_v128f32 = call <128 x float> @llvm.vp.uitofp.v128f32.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
  %vp_v128fi8_v128f64 = call <128 x double> @llvm.vp.uitofp.v128f64.v128i8(<128 x i8> undef, <128 x i1> undef, i32 undef)
  %vp_v128fi16_v128f32 = call <128 x float> @llvm.vp.uitofp.v128f32.v128i16(<128 x i16> undef, <128 x i1> undef, i32 undef)
  %vp_v128fi16_v128f64 = call <128 x double> @llvm.vp.uitofp.v128f64.v128i16(<128 x i16> undef, <128 x i1> undef, i32 undef)
  %vp_v128fi32_v128f32 = call <128 x float> @llvm.vp.uitofp.v128f32.v128i32(<128 x i32> undef, <128 x i1> undef, i32 undef)
  %vp_v128fi32_v128f64 = call <128 x double> @llvm.vp.uitofp.v128f64.v128i32(<128 x i32> undef, <128 x i1> undef, i32 undef)
  %vp_v128fi64_v128f32 = call <128 x float> @llvm.vp.uitofp.v128f32.v128i64(<128 x i64> undef, <128 x i1> undef, i32 undef)
  %vp_v128fi64_v128f64 = call <128 x double> @llvm.vp.uitofp.v128f64.v128i64(<128 x i64> undef, <128 x i1> undef, i32 undef)
  %vp_v128fi1_v128f32 = call <128 x float> @llvm.vp.uitofp.v128f32.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)
  %vp_v128fi1_v128f64 = call <128 x double> @llvm.vp.uitofp.v128f64.v128i1(<128 x i1> undef, <128 x i1> undef, i32 undef)

  %nxv1i8_nxv1f32 = uitofp <vscale x 1 x i8> undef to <vscale x 1 x float>
  %nxv1i8_nxv1f64 = uitofp <vscale x 1 x i8> undef to <vscale x 1 x double>
  %nxv1i16_nxv1f32 = uitofp <vscale x 1 x i16> undef to <vscale x 1 x float>
  %nxv1i16_nxv1f64 = uitofp <vscale x 1 x i16> undef to <vscale x 1 x double>
  %nxv1i32_nxv1f32 = uitofp <vscale x 1 x i32> undef to <vscale x 1 x float>
  %nxv1i32_nxv1f64 = uitofp <vscale x 1 x i32> undef to <vscale x 1 x double>
  %nxv1i64_nxv1f32 = uitofp <vscale x 1 x i64> undef to <vscale x 1 x float>
  %nxv1i64_nxv1f64 = uitofp <vscale x 1 x i64> undef to <vscale x 1 x double>
  %nxv1i1_nxv1f32 = uitofp <vscale x 1 x i1> undef to <vscale x 1 x float>
  %nxv1i1_nxv1f64 = uitofp <vscale x 1 x i1> undef to <vscale x 1 x double>

  %vp_nxv1fi8_nxv1f32 = call <vscale x 1 x float> @llvm.vp.uitofp.nxv1f32.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1fi8_nxv1f64 = call <vscale x 1 x double> @llvm.vp.uitofp.nxv1f64.nxv1i8(<vscale x 1 x i8> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1fi16_nxv1f32 = call <vscale x 1 x float> @llvm.vp.uitofp.nxv1f32.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1fi16_nxv1f64 = call <vscale x 1 x double> @llvm.vp.uitofp.nxv1f64.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1fi32_nxv1f32 = call <vscale x 1 x float> @llvm.vp.uitofp.nxv1f32.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1fi32_nxv1f64 = call <vscale x 1 x double> @llvm.vp.uitofp.nxv1f64.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1fi64_nxv1f32 = call <vscale x 1 x float> @llvm.vp.uitofp.nxv1f32.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1fi64_nxv1f64 = call <vscale x 1 x double> @llvm.vp.uitofp.nxv1f64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1fi1_nxv1f32 = call <vscale x 1 x float> @llvm.vp.uitofp.nxv1f32.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)
  %vp_nxv1fi1_nxv1f64 = call <vscale x 1 x double> @llvm.vp.uitofp.nxv1f64.nxv1i1(<vscale x 1 x i1> undef, <vscale x 1 x i1> undef, i32 undef)

  %nxv2i8_nxv2f32 = uitofp <vscale x 2 x i8> undef to <vscale x 2 x float>
  %nxv2i8_nxv2f64 = uitofp <vscale x 2 x i8> undef to <vscale x 2 x double>
  %nxv2i16_nxv2f32 = uitofp <vscale x 2 x i16> undef to <vscale x 2 x float>
  %nxv2i16_nxv2f64 = uitofp <vscale x 2 x i16> undef to <vscale x 2 x double>
  %nxv2i32_nxv2f32 = uitofp <vscale x 2 x i32> undef to <vscale x 2 x float>
  %nxv2i32_nxv2f64 = uitofp <vscale x 2 x i32> undef to <vscale x 2 x double>
  %nxv2i64_nxv2f32 = uitofp <vscale x 2 x i64> undef to <vscale x 2 x float>
  %nxv2i64_nxv2f64 = uitofp <vscale x 2 x i64> undef to <vscale x 2 x double>
  %nxv2i1_nxv2f32 = uitofp <vscale x 2 x i1> undef to <vscale x 2 x float>
  %nxv2i1_nxv2f64 = uitofp <vscale x 2 x i1> undef to <vscale x 2 x double>

  %vp_nxv2fi8_nxv2f32 = call <vscale x 2 x float> @llvm.vp.uitofp.nxv2f32.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2fi8_nxv2f64 = call <vscale x 2 x double> @llvm.vp.uitofp.nxv2f64.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2fi16_nxv2f32 = call <vscale x 2 x float> @llvm.vp.uitofp.nxv2f32.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2fi16_nxv2f64 = call <vscale x 2 x double> @llvm.vp.uitofp.nxv2f64.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2fi32_nxv2f32 = call <vscale x 2 x float> @llvm.vp.uitofp.nxv2f32.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2fi32_nxv2f64 = call <vscale x 2 x double> @llvm.vp.uitofp.nxv2f64.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2fi64_nxv2f32 = call <vscale x 2 x float> @llvm.vp.uitofp.nxv2f32.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2fi64_nxv2f64 = call <vscale x 2 x double> @llvm.vp.uitofp.nxv2f64.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2fi1_nxv2f32 = call <vscale x 2 x float> @llvm.vp.uitofp.nxv2f32.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)
  %vp_nxv2fi1_nxv2f64 = call <vscale x 2 x double> @llvm.vp.uitofp.nxv2f64.nxv2i1(<vscale x 2 x i1> undef, <vscale x 2 x i1> undef, i32 undef)

  %nxv4i8_nxv4f32 = uitofp <vscale x 4 x i8> undef to <vscale x 4 x float>
  %nxv4i8_nxv4f64 = uitofp <vscale x 4 x i8> undef to <vscale x 4 x double>
  %nxv4i16_nxv4f32 = uitofp <vscale x 4 x i16> undef to <vscale x 4 x float>
  %nxv4i16_nxv4f64 = uitofp <vscale x 4 x i16> undef to <vscale x 4 x double>
  %nxv4i32_nxv4f32 = uitofp <vscale x 4 x i32> undef to <vscale x 4 x float>
  %nxv4i32_nxv4f64 = uitofp <vscale x 4 x i32> undef to <vscale x 4 x double>
  %nxv4i64_nxv4f32 = uitofp <vscale x 4 x i64> undef to <vscale x 4 x float>
  %nxv4i64_nxv4f64 = uitofp <vscale x 4 x i64> undef to <vscale x 4 x double>
  %nxv4i1_nxv4f32 = uitofp <vscale x 4 x i1> undef to <vscale x 4 x float>
  %nxv4i1_nxv4f64 = uitofp <vscale x 4 x i1> undef to <vscale x 4 x double>

  %vp_nxv4fi8_nxv4f32 = call <vscale x 4 x float> @llvm.vp.uitofp.nxv4f32.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4fi8_nxv4f64 = call <vscale x 4 x double> @llvm.vp.uitofp.nxv4f64.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4fi16_nxv4f32 = call <vscale x 4 x float> @llvm.vp.uitofp.nxv4f32.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4fi16_nxv4f64 = call <vscale x 4 x double> @llvm.vp.uitofp.nxv4f64.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4fi32_nxv4f32 = call <vscale x 4 x float> @llvm.vp.uitofp.nxv4f32.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4fi32_nxv4f64 = call <vscale x 4 x double> @llvm.vp.uitofp.nxv4f64.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4fi64_nxv4f32 = call <vscale x 4 x float> @llvm.vp.uitofp.nxv4f32.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4fi64_nxv4f64 = call <vscale x 4 x double> @llvm.vp.uitofp.nxv4f64.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4fi1_nxv4f32 = call <vscale x 4 x float> @llvm.vp.uitofp.nxv4f32.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)
  %vp_nxv4fi1_nxv4f64 = call <vscale x 4 x double> @llvm.vp.uitofp.nxv4f64.nxv4i1(<vscale x 4 x i1> undef, <vscale x 4 x i1> undef, i32 undef)

  %nxv8i8_nxv8f32 = uitofp <vscale x 8 x i8> undef to <vscale x 8 x float>
  %nxv8i8_nxv8f64 = uitofp <vscale x 8 x i8> undef to <vscale x 8 x double>
  %nxv8i16_nxv8f32 = uitofp <vscale x 8 x i16> undef to <vscale x 8 x float>
  %nxv8i16_nxv8f64 = uitofp <vscale x 8 x i16> undef to <vscale x 8 x double>
  %nxv8i32_nxv8f32 = uitofp <vscale x 8 x i32> undef to <vscale x 8 x float>
  %nxv8i32_nxv8f64 = uitofp <vscale x 8 x i32> undef to <vscale x 8 x double>
  %nxv8i64_nxv8f32 = uitofp <vscale x 8 x i64> undef to <vscale x 8 x float>
  %nxv8i64_nxv8f64 = uitofp <vscale x 8 x i64> undef to <vscale x 8 x double>
  %nxv8i1_nxv8f32 = uitofp <vscale x 8 x i1> undef to <vscale x 8 x float>
  %nxv8i1_nxv8f64 = uitofp <vscale x 8 x i1> undef to <vscale x 8 x double>

  %vp_nxv8fi8_nxv8f32 = call <vscale x 8 x float> @llvm.vp.uitofp.nxv8f32.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8fi8_nxv8f64 = call <vscale x 8 x double> @llvm.vp.uitofp.nxv8f64.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8fi16_nxv8f32 = call <vscale x 8 x float> @llvm.vp.uitofp.nxv8f32.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8fi16_nxv8f64 = call <vscale x 8 x double> @llvm.vp.uitofp.nxv8f64.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8fi32_nxv8f32 = call <vscale x 8 x float> @llvm.vp.uitofp.nxv8f32.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8fi32_nxv8f64 = call <vscale x 8 x double> @llvm.vp.uitofp.nxv8f64.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8fi64_nxv8f32 = call <vscale x 8 x float> @llvm.vp.uitofp.nxv8f32.nxv8i64(<vscale x 8 x i64> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8fi64_nxv8f64 = call <vscale x 8 x double> @llvm.vp.uitofp.nxv8f64.nxv8i64(<vscale x 8 x i64> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8fi1_nxv8f32 = call <vscale x 8 x float> @llvm.vp.uitofp.nxv8f32.nxv8i1(<vscale x 8 x i1> undef, <vscale x 8 x i1> undef, i32 undef)
  %vp_nxv8fi1_nxv8f64 = call <vscale x 8 x double> @llvm.vp.uitofp.nxv8f64.nxv8i1(<vscale x 8 x i1> undef, <vscale x 8 x i1> undef, i32 undef)

  %nxv16i8_nxv16f32 = uitofp <vscale x 16 x i8> undef to <vscale x 16 x float>
  %nxv16i8_nxv16f64 = uitofp <vscale x 16 x i8> undef to <vscale x 16 x double>
  %nxv16i16_nxv16f32 = uitofp <vscale x 16 x i16> undef to <vscale x 16 x float>
  %nxv16i16_nxv16f64 = uitofp <vscale x 16 x i16> undef to <vscale x 16 x double>
  %nxv16i32_nxv16f32 = uitofp <vscale x 16 x i32> undef to <vscale x 16 x float>
  %nxv16i32_nxv16f64 = uitofp <vscale x 16 x i32> undef to <vscale x 16 x double>
  %nxv16i64_nxv16f32 = uitofp <vscale x 16 x i64> undef to <vscale x 16 x float>
  %nxv16i64_nxv16f64 = uitofp <vscale x 16 x i64> undef to <vscale x 16 x double>
  %nxv16i1_nxv16f32 = uitofp <vscale x 16 x i1> undef to <vscale x 16 x float>
  %nxv16i1_nxv16f64 = uitofp <vscale x 16 x i1> undef to <vscale x 16 x double>

  %vp_nxv16fi8_nxv16f32 = call <vscale x 16 x float> @llvm.vp.uitofp.nxv16f32.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16fi8_nxv16f64 = call <vscale x 16 x double> @llvm.vp.uitofp.nxv16f64.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16fi16_nxv16f32 = call <vscale x 16 x float> @llvm.vp.uitofp.nxv16f32.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16fi16_nxv16f64 = call <vscale x 16 x double> @llvm.vp.uitofp.nxv16f64.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16fi32_nxv16f32 = call <vscale x 16 x float> @llvm.vp.uitofp.nxv16f32.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16fi32_nxv16f64 = call <vscale x 16 x double> @llvm.vp.uitofp.nxv16f64.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16fi64_nxv16f32 = call <vscale x 16 x float> @llvm.vp.uitofp.nxv16f32.nxv16i64(<vscale x 16 x i64> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16fi64_nxv16f64 = call <vscale x 16 x double> @llvm.vp.uitofp.nxv16f64.nxv16i64(<vscale x 16 x i64> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16fi1_nxv16f32 = call <vscale x 16 x float> @llvm.vp.uitofp.nxv16f32.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)
  %vp_nxv16fi1_nxv16f64 = call <vscale x 16 x double> @llvm.vp.uitofp.nxv16f64.nxv16i1(<vscale x 16 x i1> undef, <vscale x 16 x i1> undef, i32 undef)

  %nxv32i8_nxv32f32 = uitofp <vscale x 32 x i8> undef to <vscale x 32 x float>
  %nxv32i8_nxv32f64 = uitofp <vscale x 32 x i8> undef to <vscale x 32 x double>
  %nxv32i16_nxv32f32 = uitofp <vscale x 32 x i16> undef to <vscale x 32 x float>
  %nxv32i16_nxv32f64 = uitofp <vscale x 32 x i16> undef to <vscale x 32 x double>
  %nxv32i32_nxv32f32 = uitofp <vscale x 32 x i32> undef to <vscale x 32 x float>
  %nxv32i32_nxv32f64 = uitofp <vscale x 32 x i32> undef to <vscale x 32 x double>
  %nxv32i64_nxv32f32 = uitofp <vscale x 32 x i64> undef to <vscale x 32 x float>
  %nxv32i64_nxv32f64 = uitofp <vscale x 32 x i64> undef to <vscale x 32 x double>
  %nxv32i1_nxv32f32 = uitofp <vscale x 32 x i1> undef to <vscale x 32 x float>
  %nxv32i1_nxv32f64 = uitofp <vscale x 32 x i1> undef to <vscale x 32 x double>

  %vp_nxv32fi8_nxv32f32 = call <vscale x 32 x float> @llvm.vp.uitofp.nxv32f32.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32fi8_nxv32f64 = call <vscale x 32 x double> @llvm.vp.uitofp.nxv32f64.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32fi16_nxv32f32 = call <vscale x 32 x float> @llvm.vp.uitofp.nxv32f32.nxv32i16(<vscale x 32 x i16> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32fi16_nxv32f64 = call <vscale x 32 x double> @llvm.vp.uitofp.nxv32f64.nxv32i16(<vscale x 32 x i16> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32fi32_nxv32f32 = call <vscale x 32 x float> @llvm.vp.uitofp.nxv32f32.nxv32i32(<vscale x 32 x i32> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32fi32_nxv32f64 = call <vscale x 32 x double> @llvm.vp.uitofp.nxv32f64.nxv32i32(<vscale x 32 x i32> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32fi64_nxv32f32 = call <vscale x 32 x float> @llvm.vp.uitofp.nxv32f32.nxv32i64(<vscale x 32 x i64> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32fi64_nxv32f64 = call <vscale x 32 x double> @llvm.vp.uitofp.nxv32f64.nxv32i64(<vscale x 32 x i64> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32fi1_nxv32f32 = call <vscale x 32 x float> @llvm.vp.uitofp.nxv32f32.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)
  %vp_nxv32fi1_nxv32f64 = call <vscale x 32 x double> @llvm.vp.uitofp.nxv32f64.nxv32i1(<vscale x 32 x i1> undef, <vscale x 32 x i1> undef, i32 undef)

  %nxv64i8_nxv64f32 = uitofp <vscale x 64 x i8> undef to <vscale x 64 x float>
  %nxv64i8_nxv64f64 = uitofp <vscale x 64 x i8> undef to <vscale x 64 x double>
  %nxv64i16_nxv64f32 = uitofp <vscale x 64 x i16> undef to <vscale x 64 x float>
  %nxv64i16_nxv64f64 = uitofp <vscale x 64 x i16> undef to <vscale x 64 x double>
  %nxv64i32_nxv64f32 = uitofp <vscale x 64 x i32> undef to <vscale x 64 x float>
  %nxv64i32_nxv64f64 = uitofp <vscale x 64 x i32> undef to <vscale x 64 x double>
  %nxv64i64_nxv64f32 = uitofp <vscale x 64 x i64> undef to <vscale x 64 x float>
  %nxv64i64_nxv64f64 = uitofp <vscale x 64 x i64> undef to <vscale x 64 x double>
  %nxv64i1_nxv64f32 = uitofp <vscale x 64 x i1> undef to <vscale x 64 x float>
  %nxv64i1_nxv64f64 = uitofp <vscale x 64 x i1> undef to <vscale x 64 x double>

  %vp_nxv64fi8_nxv64f32 = call <vscale x 64 x float> @llvm.vp.uitofp.nxv64f32.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64fi8_nxv64f64 = call <vscale x 64 x double> @llvm.vp.uitofp.nxv64f64.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64fi16_nxv64f32 = call <vscale x 64 x float> @llvm.vp.uitofp.nxv64f32.nxv64i16(<vscale x 64 x i16> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64fi16_nxv64f64 = call <vscale x 64 x double> @llvm.vp.uitofp.nxv64f64.nxv64i16(<vscale x 64 x i16> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64fi32_nxv64f32 = call <vscale x 64 x float> @llvm.vp.uitofp.nxv64f32.nxv64i32(<vscale x 64 x i32> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64fi32_nxv64f64 = call <vscale x 64 x double> @llvm.vp.uitofp.nxv64f64.nxv64i32(<vscale x 64 x i32> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64fi64_nxv64f32 = call <vscale x 64 x float> @llvm.vp.uitofp.nxv64f32.nxv64i64(<vscale x 64 x i64> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64fi64_nxv64f64 = call <vscale x 64 x double> @llvm.vp.uitofp.nxv64f64.nxv64i64(<vscale x 64 x i64> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64fi1_nxv64f32 = call <vscale x 64 x float> @llvm.vp.uitofp.nxv64f32.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)
  %vp_nxv64fi1_nxv64f64 = call <vscale x 64 x double> @llvm.vp.uitofp.nxv64f64.nxv64i1(<vscale x 64 x i1> undef, <vscale x 64 x i1> undef, i32 undef)

  ret void
}

define void @oddvec_sizes() {
; ZVE64D-LABEL: 'oddvec_sizes'
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = sext <3 x i8> undef to <3 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %2 = sext <7 x i8> undef to <7 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %3 = sext <15 x i8> undef to <15 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = zext <3 x i8> undef to <3 x i16>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = zext <7 x i8> undef to <7 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %6 = zext <15 x i8> undef to <15 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %7 = trunc <3 x i32> undef to <3 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %8 = trunc <7 x i32> undef to <7 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %9 = trunc <15 x i32> undef to <15 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %10 = bitcast <3 x i32> undef to <3 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %11 = bitcast <7 x i32> undef to <7 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %12 = bitcast <15 x i32> undef to <15 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %13 = sitofp <3 x i32> undef to <3 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %14 = sitofp <7 x i32> undef to <7 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %15 = sitofp <15 x i32> undef to <15 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %16 = uitofp <3 x i32> undef to <3 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %17 = uitofp <7 x i32> undef to <7 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %18 = uitofp <15 x i32> undef to <15 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %19 = fptosi <3 x float> undef to <3 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %20 = fptosi <7 x float> undef to <7 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %21 = fptosi <15 x float> undef to <15 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %22 = fptoui <3 x float> undef to <3 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %23 = fptoui <7 x float> undef to <7 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %24 = fptoui <15 x float> undef to <15 x i32>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; ZVE64X-LABEL: 'oddvec_sizes'
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = sext <3 x i8> undef to <3 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %2 = sext <7 x i8> undef to <7 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %3 = sext <15 x i8> undef to <15 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = zext <3 x i8> undef to <3 x i16>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = zext <7 x i8> undef to <7 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %6 = zext <15 x i8> undef to <15 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %7 = trunc <3 x i32> undef to <3 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %8 = trunc <7 x i32> undef to <7 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %9 = trunc <15 x i32> undef to <15 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %10 = bitcast <3 x i32> undef to <3 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %11 = bitcast <7 x i32> undef to <7 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %12 = bitcast <15 x i32> undef to <15 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %13 = sitofp <3 x i32> undef to <3 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %14 = sitofp <7 x i32> undef to <7 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %15 = sitofp <15 x i32> undef to <15 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %16 = uitofp <3 x i32> undef to <3 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %17 = uitofp <7 x i32> undef to <7 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %18 = uitofp <15 x i32> undef to <15 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %19 = fptosi <3 x float> undef to <3 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %20 = fptosi <7 x float> undef to <7 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 73 for instruction: %21 = fptosi <15 x float> undef to <15 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %22 = fptoui <3 x float> undef to <3 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %23 = fptoui <7 x float> undef to <7 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 73 for instruction: %24 = fptoui <15 x float> undef to <15 x i32>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  sext <3 x i8> undef to <3 x i16>
  sext <7 x i8> undef to <7 x i32>
  sext <15 x i8> undef to <15 x i32>

  zext <3 x i8> undef to <3 x i16>
  zext <7 x i8> undef to <7 x i32>
  zext <15 x i8> undef to <15 x i32>

  trunc <3 x i32> undef to <3 x i8>
  trunc <7 x i32> undef to <7 x i8>
  trunc <15 x i32> undef to <15 x i8>

  bitcast <3 x i32> undef to <3 x float>
  bitcast <7 x i32> undef to <7 x float>
  bitcast <15 x i32> undef to <15 x float>

  sitofp <3 x i32> undef to <3 x float>
  sitofp <7 x i32> undef to <7 x float>
  sitofp <15 x i32> undef to <15 x float>

  uitofp <3 x i32> undef to <3 x float>
  uitofp <7 x i32> undef to <7 x float>
  uitofp <15 x i32> undef to <15 x float>

  fptosi <3 x float> undef to <3 x i32>
  fptosi <7 x float> undef to <7 x i32>
  fptosi <15 x float> undef to <15 x i32>

  fptoui <3 x float> undef to <3 x i32>
  fptoui <7 x float> undef to <7 x i32>
  fptoui <15 x float> undef to <15 x i32>

  ret void
}

; This test exercises cases which involve both vector widening and
; vector splitting.  We previously crashed on this case due to an
; infinite recursion between cast costing and scalarization costing.
define void @legalization_crash() {
; ZVE64D-LABEL: 'legalization_crash'
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16700 for instruction: %1 = bitcast <24 x i8> undef to <192 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %2 = trunc <192 x i8> undef to <192 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %3 = zext <192 x i1> undef to <192 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %4 = sext <192 x i1> undef to <192 x i8>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 948 for instruction: %5 = sitofp <192 x i1> undef to <192 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 948 for instruction: %6 = uitofp <192 x i1> undef to <192 x float>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16892 for instruction: %7 = fptosi <192 x float> undef to <192 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 16892 for instruction: %8 = fptoui <192 x float> undef to <192 x i1>
; ZVE64D-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; ZVE64X-LABEL: 'legalization_crash'
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 16700 for instruction: %1 = bitcast <24 x i8> undef to <192 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %2 = trunc <192 x i8> undef to <192 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %3 = zext <192 x i1> undef to <192 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %4 = sext <192 x i1> undef to <192 x i8>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %5 = sitofp <192 x i1> undef to <192 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %6 = uitofp <192 x i1> undef to <192 x float>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 16892 for instruction: %7 = fptosi <192 x float> undef to <192 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 16892 for instruction: %8 = fptoui <192 x float> undef to <192 x i1>
; ZVE64X-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
;
  bitcast <24 x i8> undef to <192 x i1>
  trunc <192 x i8> undef to <192 x i1>
  zext <192 x i1> undef to <192 x i8>
  sext <192 x i1> undef to <192 x i8>
  sitofp <192 x i1> undef to <192 x float>
  uitofp <192 x i1> undef to <192 x float>
  fptosi <192 x float> undef to <192 x i1>
  fptoui <192 x float> undef to <192 x i1>
  ret void
}