[llvm] [RISCV] Increase cost of vmv.x.s and vfmv.f.s (PR #214481)
Luke Lau via llvm-commits
llvm-commits at lists.llvm.org
Thu Aug 6 05:55:09 PDT 2026
https://github.com/lukel97 created https://github.com/llvm/llvm-project/pull/214481
A change landed that enabled SLP vectorization in more places, but it caused a 16% regression in 508.namd_r on the spacemit-k3: https://github.com/llvm/llvm-project/issues/214417
The root cause is that we're emitting two vfmv.f.s now which are generally slow due to the vector->scalar domain crossing: https://godbolt.org/z/dMo55zPMa
Reciprocal throughput of vmv.x.s/vfmv.f.s is 6 times higher than vmv.s.x/vfmv.s.f on the spacemit-k3 and 2.5 higher on the P870 according to https://camel-cdr.github.io/rvv-bench-results/
This PR increases the cost of a vmv.x.s/vfmv.f.s from 1->2, which is just enough to prevent unprofitable SLP vectorization and fixes the regression on 508.namd_r.
It also has the bonus side effect of preventing loop vectorization where we end up scalarizing libcalls, see llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-call-intrinsics.ll
Fixes #214417
>From 7831d2fa93b9b3e4aeaf48c4c4d806a1d0246b6e Mon Sep 17 00:00:00 2001
From: Luke Lau <luke at igalia.com>
Date: Thu, 6 Aug 2026 18:22:48 +0800
Subject: [PATCH 1/2] Precommit test
---
.../SLPVectorizer/RISCV/extract-cost.ll | 26 +++++++++++++++++++
1 file changed, 26 insertions(+)
create mode 100644 llvm/test/Transforms/SLPVectorizer/RISCV/extract-cost.ll
diff --git a/llvm/test/Transforms/SLPVectorizer/RISCV/extract-cost.ll b/llvm/test/Transforms/SLPVectorizer/RISCV/extract-cost.ll
new file mode 100644
index 0000000000000..afba2b7089441
--- /dev/null
+++ b/llvm/test/Transforms/SLPVectorizer/RISCV/extract-cost.ll
@@ -0,0 +1,26 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 6
+; RUN: opt < %s -S -p slp-vectorizer -mtriple riscv64 -mattr=+v | FileCheck %s
+
+; Don't vectorize as the extract elements are expensive enough to be unprofitable.
+
+define double @f(ptr %p, double %c) {
+; CHECK-LABEL: define double @f(
+; CHECK-SAME: ptr [[P:%.*]], double [[C:%.*]]) #[[ATTR0:[0-9]+]] {
+; CHECK-NEXT: [[TMP1:%.*]] = load <2 x double>, ptr [[P]], align 8
+; CHECK-NEXT: [[TMP2:%.*]] = fmul <2 x double> splat (double 2.000000e+00), [[TMP1]]
+; CHECK-NEXT: [[TMP3:%.*]] = fadd <2 x double> splat (double 3.000000e+00), [[TMP2]]
+; CHECK-NEXT: [[TMP4:%.*]] = extractelement <2 x double> [[TMP3]], i64 0
+; CHECK-NEXT: [[TMP5:%.*]] = extractelement <2 x double> [[TMP3]], i64 1
+; CHECK-NEXT: [[RES:%.*]] = call double @llvm.fmuladd.f64(double [[C]], double [[TMP5]], double [[TMP4]])
+; CHECK-NEXT: ret double [[RES]]
+;
+ %x0 = load double, ptr %p
+ %q = getelementptr double, ptr %p, i64 1
+ %x1 = load double, ptr %q
+ %m0 = fmul double 2.0, %x1
+ %s0 = fadd double 3.0, %m0
+ %m1 = fmul double 2.0, %x0
+ %s1 = fadd double 3.0, %m1
+ %res = call double @llvm.fmuladd.f64(double %c, double %s0, double %s1)
+ ret double %res
+}
>From c1503749ed67920888f0e6b5729e17978318c8d7 Mon Sep 17 00:00:00 2001
From: Luke Lau <luke at igalia.com>
Date: Thu, 6 Aug 2026 18:31:09 +0800
Subject: [PATCH 2/2] [RISCV] Increase cost of vmv.x.s and vfmv.f.s
A change landed that enabled SLP vectorization in more places, but it caused a 16% regression in 508.namd_r on the spacemit-k3: https://github.com/llvm/llvm-project/issues/214417
The root cause is that we're emitting two vfmv.f.s now which are generally slow due to the vector->scalar domain crossing: https://godbolt.org/z/dMo55zPMa
Reciprocal throughput of vmv.x.s/vfmv.f.s is 6 times higher than vmv.s.x/vfmv.s.f on the spacemit-k3 and 2.5 higher on the P870 according to https://camel-cdr.github.io/rvv-bench-results/
This PR increases the cost of a vmv.x.s/vfmv.f.s from 1->2, which is just enough to prevent unprofitable SLP vectorization and fixes the regression on 508.namd_r.
It also has the bonus side effect of preventing loop vectorization where we end up scalarizing libcalls, see llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-call-intrinsics.ll
Fixes #214417
---
.../Target/RISCV/RISCVTargetTransformInfo.cpp | 15 +-
.../test/Analysis/CostModel/RISCV/arith-fp.ll | 64 +-
.../test/Analysis/CostModel/RISCV/cast-sat.ll | 64 +-
llvm/test/Analysis/CostModel/RISCV/cast.ll | 478 ++---
llvm/test/Analysis/CostModel/RISCV/clmul.ll | 10 +-
.../Analysis/CostModel/RISCV/cttz_elts.ll | 16 +-
.../CostModel/RISCV/extract-last-active.ll | 86 +-
.../CostModel/RISCV/fixed-vector-gather.ll | 72 +-
.../CostModel/RISCV/fixed-vector-scatter.ll | 72 +-
.../Analysis/CostModel/RISCV/masked_ldst.ll | 8 +-
.../Analysis/CostModel/RISCV/reduce-add.ll | 64 +-
.../Analysis/CostModel/RISCV/reduce-and.ll | 64 +-
.../Analysis/CostModel/RISCV/reduce-fadd.ll | 252 +--
.../CostModel/RISCV/reduce-fmaximum.ll | 52 +-
.../CostModel/RISCV/reduce-fminimum.ll | 26 +-
.../Analysis/CostModel/RISCV/reduce-fmul.ll | 144 +-
.../Analysis/CostModel/RISCV/reduce-max.ll | 128 +-
.../Analysis/CostModel/RISCV/reduce-min.ll | 128 +-
.../Analysis/CostModel/RISCV/reduce-or.ll | 64 +-
.../CostModel/RISCV/reduce-scalable-fp.ll | 134 +-
.../CostModel/RISCV/reduce-scalable-int.ll | 228 +--
.../Analysis/CostModel/RISCV/reduce-xor.ll | 64 +-
.../RISCV/rvv-expandload-compressstore.ll | 64 +-
.../CostModel/RISCV/rvv-extractelement.ll | 1723 ++++++++---------
.../Analysis/CostModel/RISCV/rvv-shuffle.ll | 8 +-
.../CostModel/RISCV/shuffle-broadcast.ll | 28 +-
.../Analysis/CostModel/RISCV/vp-intrinsics.ll | 248 +--
.../RISCV/conditional-scalar-assignment.ll | 2 +-
.../RISCV/tail-folding-call-intrinsics.ll | 36 +-
.../LoopVectorize/RISCV/uniform-load-store.ll | 21 +-
.../SLPVectorizer/RISCV/complex-loads.ll | 123 +-
.../SLPVectorizer/RISCV/external.ll | 24 +-
.../SLPVectorizer/RISCV/extract-cost.ll | 12 +-
.../SLPVectorizer/RISCV/horizontal-list.ll | 2 +-
.../SLPVectorizer/RISCV/load-store.ll | 20 +-
.../reduced-value-repeated-and-vectorized.ll | 5 +-
.../SLPVectorizer/RISCV/reductions.ll | 134 +-
.../RISCV/reordered-interleaved-loads.ll | 87 +-
.../RISCV/runtime-strided-stores.ll | 198 +-
.../strided-loads-with-external-use-ptr.ll | 6 +-
.../SLPVectorizer/RISCV/vec3-base.ll | 34 +-
.../RISCV/fold-equivalent-reduction-cmp.ll | 4 +-
42 files changed, 2613 insertions(+), 2399 deletions(-)
diff --git a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
index 5ab8546722d4c..3165e89ec09e6 100644
--- a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
+++ b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
@@ -102,8 +102,11 @@ RISCVTTIImpl::getRISCVInstructionCost(ArrayRef<unsigned> OpCodes, MVT VT,
break;
}
case RISCV::VMV_X_S:
- case RISCV::VMV_S_X:
case RISCV::VFMV_F_S:
+ // Domain crossings from vector -> scalar are usually more expensive.
+ Cost += 2;
+ break;
+ case RISCV::VMV_S_X:
case RISCV::VFMV_S_F:
case RISCV::VMOR_MM:
case RISCV::VMXOR_MM:
@@ -2635,7 +2638,15 @@ InstructionCost RISCVTTIImpl::getVectorInstrCost(
// In RVV, we could use vslidedown + vmv.x.s to extract element from vector
// and vslideup + vmv.s.x to insert element to vector.
- unsigned BaseCost = 1;
+ unsigned MoveOpc;
+ if (LT.second.isFloatingPoint())
+ MoveOpc = Opcode == Instruction::InsertElement ? RISCV::VFMV_S_F
+ : RISCV::VFMV_F_S;
+ else
+ MoveOpc =
+ Opcode == Instruction::InsertElement ? RISCV::VMV_S_X : RISCV::VMV_X_S;
+ InstructionCost BaseCost =
+ getRISCVInstructionCost(MoveOpc, LT.second, CostKind);
// When insertelement we should add the index with 1 as the input of vslideup.
unsigned SlideCost = Opcode == Instruction::InsertElement ? 2 : 1;
diff --git a/llvm/test/Analysis/CostModel/RISCV/arith-fp.ll b/llvm/test/Analysis/CostModel/RISCV/arith-fp.ll
index 03bdd439ba2fc..5e26f5d327395 100644
--- a/llvm/test/Analysis/CostModel/RISCV/arith-fp.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/arith-fp.ll
@@ -1268,15 +1268,15 @@ define void @frem() {
; CHECK-NEXT: Cost Model: Invalid cost for instruction: %NXV2F64 = frem <vscale x 2 x double> poison, poison
; CHECK-NEXT: Cost Model: Invalid cost for instruction: %NXV4F64 = frem <vscale x 4 x double> poison, poison
; CHECK-NEXT: Cost Model: Invalid cost for instruction: %NXV8F64 = frem <vscale x 8 x double> poison, poison
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V1F32_VP = call <1 x float> @llvm.vp.frem.v1f32(<1 x float> poison, <1 x float> poison, <1 x i1> poison, i32 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %V2F32_VP = call <2 x float> @llvm.vp.frem.v2f32(<2 x float> poison, <2 x float> poison, <2 x i1> poison, i32 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 27 for instruction: %V4F32_VP = call <4 x float> @llvm.vp.frem.v4f32(<4 x float> poison, <4 x float> poison, <4 x i1> poison, i32 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 62 for instruction: %V8F32_VP = call <8 x float> @llvm.vp.frem.v8f32(<8 x float> poison, <8 x float> poison, <8 x i1> poison, i32 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 126 for instruction: %V16F32_VP = call <16 x float> @llvm.vp.frem.v16f32(<16 x float> poison, <16 x float> poison, <16 x i1> poison, i32 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V1F64_VP = call <1 x double> @llvm.vp.frem.v1f64(<1 x double> poison, <1 x double> poison, <1 x i1> poison, i32 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %V2F64_VP = call <2 x double> @llvm.vp.frem.v2f64(<2 x double> poison, <2 x double> poison, <2 x i1> poison, i32 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %V4F64_VP = call <4 x double> @llvm.vp.frem.v4f64(<4 x double> poison, <4 x double> poison, <4 x i1> poison, i32 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 62 for instruction: %V8F64_VP = call <8 x double> @llvm.vp.frem.v8f64(<8 x double> poison, <8 x double> poison, <8 x i1> poison, i32 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V1F32_VP = call <1 x float> @llvm.vp.frem.v1f32(<1 x float> poison, <1 x float> poison, <1 x i1> poison, i32 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V2F32_VP = call <2 x float> @llvm.vp.frem.v2f32(<2 x float> poison, <2 x float> poison, <2 x i1> poison, i32 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V4F32_VP = call <4 x float> @llvm.vp.frem.v4f32(<4 x float> poison, <4 x float> poison, <4 x i1> poison, i32 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 70 for instruction: %V8F32_VP = call <8 x float> @llvm.vp.frem.v8f32(<8 x float> poison, <8 x float> poison, <8 x i1> poison, i32 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 142 for instruction: %V16F32_VP = call <16 x float> @llvm.vp.frem.v16f32(<16 x float> poison, <16 x float> poison, <16 x i1> poison, i32 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V1F64_VP = call <1 x double> @llvm.vp.frem.v1f64(<1 x double> poison, <1 x double> poison, <1 x i1> poison, i32 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V2F64_VP = call <2 x double> @llvm.vp.frem.v2f64(<2 x double> poison, <2 x double> poison, <2 x i1> poison, i32 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 34 for instruction: %V4F64_VP = call <4 x double> @llvm.vp.frem.v4f64(<4 x double> poison, <4 x double> poison, <4 x i1> poison, i32 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 70 for instruction: %V8F64_VP = call <8 x double> @llvm.vp.frem.v8f64(<8 x double> poison, <8 x double> poison, <8 x i1> poison, i32 poison)
; CHECK-NEXT: Cost Model: Invalid cost for instruction: %NXV1F32_VP = call <vscale x 1 x float> @llvm.vp.frem.nxv1f32(<vscale x 1 x float> poison, <vscale x 1 x float> poison, <vscale x 1 x i1> poison, i32 poison)
; CHECK-NEXT: Cost Model: Invalid cost for instruction: %NXV2F32_VP = call <vscale x 2 x float> @llvm.vp.frem.nxv2f32(<vscale x 2 x float> poison, <vscale x 2 x float> poison, <vscale x 2 x i1> poison, i32 poison)
; CHECK-NEXT: Cost Model: Invalid cost for instruction: %NXV4F32_VP = call <vscale x 4 x float> @llvm.vp.frem.nxv4f32(<vscale x 4 x float> poison, <vscale x 4 x float> poison, <vscale x 4 x i1> poison, i32 poison)
@@ -1353,11 +1353,11 @@ define void @frem_bf16() {
; ZVFH-NEXT: Cost Model: Invalid cost for instruction: %NXV8BF16 = frem <vscale x 8 x bfloat> poison, poison
; ZVFH-NEXT: Cost Model: Invalid cost for instruction: %NXV16BF16 = frem <vscale x 16 x bfloat> poison, poison
; ZVFH-NEXT: Cost Model: Invalid cost for instruction: %NXV32BF16 = frem <vscale x 32 x bfloat> poison, poison
-; ZVFH-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V1BF16_VP = call <1 x bfloat> @llvm.vp.frem.v1bf16(<1 x bfloat> poison, <1 x bfloat> poison, <1 x i1> poison, i32 poison)
-; ZVFH-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %V2BF16_VP = call <2 x bfloat> @llvm.vp.frem.v2bf16(<2 x bfloat> poison, <2 x bfloat> poison, <2 x i1> poison, i32 poison)
-; ZVFH-NEXT: Cost Model: Found an estimated cost of 27 for instruction: %V4BF16_VP = call <4 x bfloat> @llvm.vp.frem.v4bf16(<4 x bfloat> poison, <4 x bfloat> poison, <4 x i1> poison, i32 poison)
-; ZVFH-NEXT: Cost Model: Found an estimated cost of 55 for instruction: %V8BF16_VP = call <8 x bfloat> @llvm.vp.frem.v8bf16(<8 x bfloat> poison, <8 x bfloat> poison, <8 x i1> poison, i32 poison)
-; ZVFH-NEXT: Cost Model: Found an estimated cost of 126 for instruction: %V16BF16_VP = call <16 x bfloat> @llvm.vp.frem.v16bf16(<16 x bfloat> poison, <16 x bfloat> poison, <16 x i1> poison, i32 poison)
+; ZVFH-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V1BF16_VP = call <1 x bfloat> @llvm.vp.frem.v1bf16(<1 x bfloat> poison, <1 x bfloat> poison, <1 x i1> poison, i32 poison)
+; ZVFH-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V2BF16_VP = call <2 x bfloat> @llvm.vp.frem.v2bf16(<2 x bfloat> poison, <2 x bfloat> poison, <2 x i1> poison, i32 poison)
+; ZVFH-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V4BF16_VP = call <4 x bfloat> @llvm.vp.frem.v4bf16(<4 x bfloat> poison, <4 x bfloat> poison, <4 x i1> poison, i32 poison)
+; ZVFH-NEXT: Cost Model: Found an estimated cost of 63 for instruction: %V8BF16_VP = call <8 x bfloat> @llvm.vp.frem.v8bf16(<8 x bfloat> poison, <8 x bfloat> poison, <8 x i1> poison, i32 poison)
+; ZVFH-NEXT: Cost Model: Found an estimated cost of 142 for instruction: %V16BF16_VP = call <16 x bfloat> @llvm.vp.frem.v16bf16(<16 x bfloat> poison, <16 x bfloat> poison, <16 x i1> poison, i32 poison)
; ZVFH-NEXT: Cost Model: Invalid cost for instruction: %NXV1BF16_VP = call <vscale x 1 x bfloat> @llvm.vp.frem.nxv1bf16(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x i1> poison, i32 poison)
; ZVFH-NEXT: Cost Model: Invalid cost for instruction: %NXV2BF16_VP = call <vscale x 2 x bfloat> @llvm.vp.frem.nxv2bf16(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x i1> poison, i32 poison)
; ZVFH-NEXT: Cost Model: Invalid cost for instruction: %NXV4BF16_VP = call <vscale x 4 x bfloat> @llvm.vp.frem.nxv4bf16(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x i1> poison, i32 poison)
@@ -1379,11 +1379,11 @@ define void @frem_bf16() {
; ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV8BF16 = frem <vscale x 8 x bfloat> poison, poison
; ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV16BF16 = frem <vscale x 16 x bfloat> poison, poison
; ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV32BF16 = frem <vscale x 32 x bfloat> poison, poison
-; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V1BF16_VP = call <1 x bfloat> @llvm.vp.frem.v1bf16(<1 x bfloat> poison, <1 x bfloat> poison, <1 x i1> poison, i32 poison)
-; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %V2BF16_VP = call <2 x bfloat> @llvm.vp.frem.v2bf16(<2 x bfloat> poison, <2 x bfloat> poison, <2 x i1> poison, i32 poison)
-; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 27 for instruction: %V4BF16_VP = call <4 x bfloat> @llvm.vp.frem.v4bf16(<4 x bfloat> poison, <4 x bfloat> poison, <4 x i1> poison, i32 poison)
-; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 55 for instruction: %V8BF16_VP = call <8 x bfloat> @llvm.vp.frem.v8bf16(<8 x bfloat> poison, <8 x bfloat> poison, <8 x i1> poison, i32 poison)
-; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 126 for instruction: %V16BF16_VP = call <16 x bfloat> @llvm.vp.frem.v16bf16(<16 x bfloat> poison, <16 x bfloat> poison, <16 x i1> poison, i32 poison)
+; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V1BF16_VP = call <1 x bfloat> @llvm.vp.frem.v1bf16(<1 x bfloat> poison, <1 x bfloat> poison, <1 x i1> poison, i32 poison)
+; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V2BF16_VP = call <2 x bfloat> @llvm.vp.frem.v2bf16(<2 x bfloat> poison, <2 x bfloat> poison, <2 x i1> poison, i32 poison)
+; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V4BF16_VP = call <4 x bfloat> @llvm.vp.frem.v4bf16(<4 x bfloat> poison, <4 x bfloat> poison, <4 x i1> poison, i32 poison)
+; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 63 for instruction: %V8BF16_VP = call <8 x bfloat> @llvm.vp.frem.v8bf16(<8 x bfloat> poison, <8 x bfloat> poison, <8 x i1> poison, i32 poison)
+; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 142 for instruction: %V16BF16_VP = call <16 x bfloat> @llvm.vp.frem.v16bf16(<16 x bfloat> poison, <16 x bfloat> poison, <16 x i1> poison, i32 poison)
; ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV1BF16_VP = call <vscale x 1 x bfloat> @llvm.vp.frem.nxv1bf16(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x i1> poison, i32 poison)
; ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV2BF16_VP = call <vscale x 2 x bfloat> @llvm.vp.frem.nxv2bf16(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x i1> poison, i32 poison)
; ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV4BF16_VP = call <vscale x 4 x bfloat> @llvm.vp.frem.nxv4bf16(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x i1> poison, i32 poison)
@@ -1463,11 +1463,11 @@ define void @frem_f16() {
; ZVFH-NEXT: Cost Model: Invalid cost for instruction: %NXV8F16 = frem <vscale x 8 x half> poison, poison
; ZVFH-NEXT: Cost Model: Invalid cost for instruction: %NXV16F16 = frem <vscale x 16 x half> poison, poison
; ZVFH-NEXT: Cost Model: Invalid cost for instruction: %NXV32F16 = frem <vscale x 32 x half> poison, poison
-; ZVFH-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V1F16_VP = call <1 x half> @llvm.vp.frem.v1f16(<1 x half> poison, <1 x half> poison, <1 x i1> poison, i32 poison)
-; ZVFH-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V2F16_VP = call <2 x half> @llvm.vp.frem.v2f16(<2 x half> poison, <2 x half> poison, <2 x i1> poison, i32 poison)
-; ZVFH-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V4F16_VP = call <4 x half> @llvm.vp.frem.v4f16(<4 x half> poison, <4 x half> poison, <4 x i1> poison, i32 poison)
-; ZVFH-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %V8F16_VP = call <8 x half> @llvm.vp.frem.v8f16(<8 x half> poison, <8 x half> poison, <8 x i1> poison, i32 poison)
-; ZVFH-NEXT: Cost Model: Found an estimated cost of 94 for instruction: %V16F16_VP = call <16 x half> @llvm.vp.frem.v16f16(<16 x half> poison, <16 x half> poison, <16 x i1> poison, i32 poison)
+; ZVFH-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V1F16_VP = call <1 x half> @llvm.vp.frem.v1f16(<1 x half> poison, <1 x half> poison, <1 x i1> poison, i32 poison)
+; ZVFH-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %V2F16_VP = call <2 x half> @llvm.vp.frem.v2f16(<2 x half> poison, <2 x half> poison, <2 x i1> poison, i32 poison)
+; ZVFH-NEXT: Cost Model: Found an estimated cost of 23 for instruction: %V4F16_VP = call <4 x half> @llvm.vp.frem.v4f16(<4 x half> poison, <4 x half> poison, <4 x i1> poison, i32 poison)
+; ZVFH-NEXT: Cost Model: Found an estimated cost of 47 for instruction: %V8F16_VP = call <8 x half> @llvm.vp.frem.v8f16(<8 x half> poison, <8 x half> poison, <8 x i1> poison, i32 poison)
+; ZVFH-NEXT: Cost Model: Found an estimated cost of 110 for instruction: %V16F16_VP = call <16 x half> @llvm.vp.frem.v16f16(<16 x half> poison, <16 x half> poison, <16 x i1> poison, i32 poison)
; ZVFH-NEXT: Cost Model: Invalid cost for instruction: %NXV1F16_VP = call <vscale x 1 x half> @llvm.vp.frem.nxv1f16(<vscale x 1 x half> poison, <vscale x 1 x half> poison, <vscale x 1 x i1> poison, i32 poison)
; ZVFH-NEXT: Cost Model: Invalid cost for instruction: %NXV2F16_VP = call <vscale x 2 x half> @llvm.vp.frem.nxv2f16(<vscale x 2 x half> poison, <vscale x 2 x half> poison, <vscale x 2 x i1> poison, i32 poison)
; ZVFH-NEXT: Cost Model: Invalid cost for instruction: %NXV4F16_VP = call <vscale x 4 x half> @llvm.vp.frem.nxv4f16(<vscale x 4 x half> poison, <vscale x 4 x half> poison, <vscale x 4 x i1> poison, i32 poison)
@@ -1489,11 +1489,11 @@ define void @frem_f16() {
; ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV8F16 = frem <vscale x 8 x half> poison, poison
; ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV16F16 = frem <vscale x 16 x half> poison, poison
; ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV32F16 = frem <vscale x 32 x half> poison, poison
-; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V1F16_VP = call <1 x half> @llvm.vp.frem.v1f16(<1 x half> poison, <1 x half> poison, <1 x i1> poison, i32 poison)
-; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V2F16_VP = call <2 x half> @llvm.vp.frem.v2f16(<2 x half> poison, <2 x half> poison, <2 x i1> poison, i32 poison)
-; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V4F16_VP = call <4 x half> @llvm.vp.frem.v4f16(<4 x half> poison, <4 x half> poison, <4 x i1> poison, i32 poison)
-; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %V8F16_VP = call <8 x half> @llvm.vp.frem.v8f16(<8 x half> poison, <8 x half> poison, <8 x i1> poison, i32 poison)
-; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 94 for instruction: %V16F16_VP = call <16 x half> @llvm.vp.frem.v16f16(<16 x half> poison, <16 x half> poison, <16 x i1> poison, i32 poison)
+; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V1F16_VP = call <1 x half> @llvm.vp.frem.v1f16(<1 x half> poison, <1 x half> poison, <1 x i1> poison, i32 poison)
+; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %V2F16_VP = call <2 x half> @llvm.vp.frem.v2f16(<2 x half> poison, <2 x half> poison, <2 x i1> poison, i32 poison)
+; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 23 for instruction: %V4F16_VP = call <4 x half> @llvm.vp.frem.v4f16(<4 x half> poison, <4 x half> poison, <4 x i1> poison, i32 poison)
+; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 47 for instruction: %V8F16_VP = call <8 x half> @llvm.vp.frem.v8f16(<8 x half> poison, <8 x half> poison, <8 x i1> poison, i32 poison)
+; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 110 for instruction: %V16F16_VP = call <16 x half> @llvm.vp.frem.v16f16(<16 x half> poison, <16 x half> poison, <16 x i1> poison, i32 poison)
; ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV1F16_VP = call <vscale x 1 x half> @llvm.vp.frem.nxv1f16(<vscale x 1 x half> poison, <vscale x 1 x half> poison, <vscale x 1 x i1> poison, i32 poison)
; ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV2F16_VP = call <vscale x 2 x half> @llvm.vp.frem.nxv2f16(<vscale x 2 x half> poison, <vscale x 2 x half> poison, <vscale x 2 x i1> poison, i32 poison)
; ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV4F16_VP = call <vscale x 4 x half> @llvm.vp.frem.nxv4f16(<vscale x 4 x half> poison, <vscale x 4 x half> poison, <vscale x 4 x i1> poison, i32 poison)
@@ -1929,7 +1929,7 @@ define void @fcopysign() {
define void @fcopysign_bf16() {
; ZVFH-LABEL: 'fcopysign_bf16'
; ZVFH-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %BF16 = call bfloat @llvm.copysign.bf16(bfloat poison, bfloat poison)
-; ZVFH-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V1BF16 = call <1 x bfloat> @llvm.copysign.v1bf16(<1 x bfloat> poison, <1 x bfloat> poison)
+; ZVFH-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V1BF16 = call <1 x bfloat> @llvm.copysign.v1bf16(<1 x bfloat> poison, <1 x bfloat> poison)
; ZVFH-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2BF16 = call <2 x bfloat> @llvm.copysign.v2bf16(<2 x bfloat> poison, <2 x bfloat> poison)
; ZVFH-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V4BF16 = call <4 x bfloat> @llvm.copysign.v4bf16(<4 x bfloat> poison, <4 x bfloat> poison)
; ZVFH-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V8BF16 = call <8 x bfloat> @llvm.copysign.v8bf16(<8 x bfloat> poison, <8 x bfloat> poison)
@@ -1945,7 +1945,7 @@ define void @fcopysign_bf16() {
;
; ZVFHMIN-LABEL: 'fcopysign_bf16'
; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %BF16 = call bfloat @llvm.copysign.bf16(bfloat poison, bfloat poison)
-; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V1BF16 = call <1 x bfloat> @llvm.copysign.v1bf16(<1 x bfloat> poison, <1 x bfloat> poison)
+; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V1BF16 = call <1 x bfloat> @llvm.copysign.v1bf16(<1 x bfloat> poison, <1 x bfloat> poison)
; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2BF16 = call <2 x bfloat> @llvm.copysign.v2bf16(<2 x bfloat> poison, <2 x bfloat> poison)
; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V4BF16 = call <4 x bfloat> @llvm.copysign.v4bf16(<4 x bfloat> poison, <4 x bfloat> poison)
; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V8BF16 = call <8 x bfloat> @llvm.copysign.v8bf16(<8 x bfloat> poison, <8 x bfloat> poison)
@@ -2013,7 +2013,7 @@ define void @fcopysign_f16() {
;
; ZVFHMIN-LABEL: 'fcopysign_f16'
; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F16 = call half @llvm.copysign.f16(half poison, half poison)
-; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V1F16 = call <1 x half> @llvm.copysign.v1f16(<1 x half> poison, <1 x half> poison)
+; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V1F16 = call <1 x half> @llvm.copysign.v1f16(<1 x half> poison, <1 x half> poison)
; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2F16 = call <2 x half> @llvm.copysign.v2f16(<2 x half> poison, <2 x half> poison)
; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V4F16 = call <4 x half> @llvm.copysign.v4f16(<4 x half> poison, <4 x half> poison)
; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V8F16 = call <8 x half> @llvm.copysign.v8f16(<8 x half> poison, <8 x half> poison)
diff --git a/llvm/test/Analysis/CostModel/RISCV/cast-sat.ll b/llvm/test/Analysis/CostModel/RISCV/cast-sat.ll
index ee70811bf9bd0..4a0112c2cb995 100644
--- a/llvm/test/Analysis/CostModel/RISCV/cast-sat.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/cast-sat.ll
@@ -7,45 +7,45 @@
define void @fptoui_sat() {
; RV64ZVE32F-LABEL: 'fptoui_sat'
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v1f32_v1i8 = call <1 x i8> @llvm.fptoui.sat.v1i8.v1f32(<1 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v1f64_v1i8 = call <1 x i8> @llvm.fptoui.sat.v1i8.v1f64(<1 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v1f64_v1i8 = call <1 x i8> @llvm.fptoui.sat.v1i8.v1f64(<1 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v1f32_v1i16 = call <1 x i16> @llvm.fptoui.sat.v1i16.v1f32(<1 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v1f64_v1i16 = call <1 x i16> @llvm.fptoui.sat.v1i16.v1f64(<1 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v1f64_v1i16 = call <1 x i16> @llvm.fptoui.sat.v1i16.v1f64(<1 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v1f32_v1i32 = call <1 x i32> @llvm.fptoui.sat.v1i32.v1f32(<1 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v1f64_v1i32 = call <1 x i32> @llvm.fptoui.sat.v1i32.v1f64(<1 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v1f64_v1i32 = call <1 x i32> @llvm.fptoui.sat.v1i32.v1f64(<1 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v1f32_v1i64 = call <1 x i64> @llvm.fptoui.sat.v1i64.v1f32(<1 x float> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v1f64_v1i64 = call <1 x i64> @llvm.fptoui.sat.v1i64.v1f64(<1 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v1f32_v1i1 = call <1 x i1> @llvm.fptoui.sat.v1i1.v1f32(<1 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %v1f64_v1i1 = call <1 x i1> @llvm.fptoui.sat.v1i1.v1f64(<1 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %v1f64_v1i1 = call <1 x i1> @llvm.fptoui.sat.v1i1.v1f64(<1 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v2f32_v2i8 = call <2 x i8> @llvm.fptoui.sat.v2i8.v2f32(<2 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %v2f64_v2i8 = call <2 x i8> @llvm.fptoui.sat.v2i8.v2f64(<2 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %v2f64_v2i8 = call <2 x i8> @llvm.fptoui.sat.v2i8.v2f64(<2 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v2f32_v2i16 = call <2 x i16> @llvm.fptoui.sat.v2i16.v2f32(<2 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %v2f64_v2i16 = call <2 x i16> @llvm.fptoui.sat.v2i16.v2f64(<2 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %v2f64_v2i16 = call <2 x i16> @llvm.fptoui.sat.v2i16.v2f64(<2 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v2f32_v2i32 = call <2 x i32> @llvm.fptoui.sat.v2i32.v2f32(<2 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %v2f64_v2i32 = call <2 x i32> @llvm.fptoui.sat.v2i32.v2f64(<2 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %v2f64_v2i32 = call <2 x i32> @llvm.fptoui.sat.v2i32.v2f64(<2 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %v2f32_v2i64 = call <2 x i64> @llvm.fptoui.sat.v2i64.v2f32(<2 x float> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v2f64_v2i64 = call <2 x i64> @llvm.fptoui.sat.v2i64.v2f64(<2 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v2f32_v2i1 = call <2 x i1> @llvm.fptoui.sat.v2i1.v2f32(<2 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %v2f64_v2i1 = call <2 x i1> @llvm.fptoui.sat.v2i1.v2f64(<2 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %v2f64_v2i1 = call <2 x i1> @llvm.fptoui.sat.v2i1.v2f64(<2 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v4f32_v4i8 = call <4 x i8> @llvm.fptoui.sat.v4i8.v4f32(<4 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %v4f64_v4i8 = call <4 x i8> @llvm.fptoui.sat.v4i8.v4f64(<4 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %v4f64_v4i8 = call <4 x i8> @llvm.fptoui.sat.v4i8.v4f64(<4 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v4f32_v4i16 = call <4 x i16> @llvm.fptoui.sat.v4i16.v4f32(<4 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %v4f64_v4i16 = call <4 x i16> @llvm.fptoui.sat.v4i16.v4f64(<4 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %v4f64_v4i16 = call <4 x i16> @llvm.fptoui.sat.v4i16.v4f64(<4 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v4f32_v4i32 = call <4 x i32> @llvm.fptoui.sat.v4i32.v4f32(<4 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %v4f64_v4i32 = call <4 x i32> @llvm.fptoui.sat.v4i32.v4f64(<4 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %v4f64_v4i32 = call <4 x i32> @llvm.fptoui.sat.v4i32.v4f64(<4 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %v4f32_v4i64 = call <4 x i64> @llvm.fptoui.sat.v4i64.v4f32(<4 x float> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %v4f64_v4i64 = call <4 x i64> @llvm.fptoui.sat.v4i64.v4f64(<4 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v4f32_v4i1 = call <4 x i1> @llvm.fptoui.sat.v4i1.v4f32(<4 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 46 for instruction: %v4f64_v4i1 = call <4 x i1> @llvm.fptoui.sat.v4i1.v4f64(<4 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 50 for instruction: %v4f64_v4i1 = call <4 x i1> @llvm.fptoui.sat.v4i1.v4f64(<4 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %v8f32_v8i8 = call <8 x i8> @llvm.fptoui.sat.v8i8.v8f32(<8 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 40 for instruction: %v8f64_v8i8 = call <8 x i8> @llvm.fptoui.sat.v8i8.v8f64(<8 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 48 for instruction: %v8f64_v8i8 = call <8 x i8> @llvm.fptoui.sat.v8i8.v8f64(<8 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %v8f32_v8i16 = call <8 x i16> @llvm.fptoui.sat.v8i16.v8f32(<8 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 40 for instruction: %v8f64_v8i16 = call <8 x i16> @llvm.fptoui.sat.v8i16.v8f64(<8 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 48 for instruction: %v8f64_v8i16 = call <8 x i16> @llvm.fptoui.sat.v8i16.v8f64(<8 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v8f32_v8i32 = call <8 x i32> @llvm.fptoui.sat.v8i32.v8f32(<8 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 41 for instruction: %v8f64_v8i32 = call <8 x i32> @llvm.fptoui.sat.v8i32.v8f64(<8 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 49 for instruction: %v8f64_v8i32 = call <8 x i32> @llvm.fptoui.sat.v8i32.v8f64(<8 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %v8f32_v8i64 = call <8 x i64> @llvm.fptoui.sat.v8i64.v8f32(<8 x float> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %v8f64_v8i64 = call <8 x i64> @llvm.fptoui.sat.v8i64.v8f64(<8 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %v8f32_v8i1 = call <8 x i1> @llvm.fptoui.sat.v8i1.v8f32(<8 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 90 for instruction: %v8f64_v8i1 = call <8 x i1> @llvm.fptoui.sat.v8i1.v8f64(<8 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 98 for instruction: %v8f64_v8i1 = call <8 x i1> @llvm.fptoui.sat.v8i1.v8f64(<8 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %nxv1f32_nxv1i8 = call <vscale x 1 x i8> @llvm.fptoui.sat.nxv1i8.nxv1f32(<vscale x 1 x float> poison)
; RV64ZVE32F-NEXT: Cost Model: Invalid cost for instruction: %nxv1f64_nxv1i8 = call <vscale x 1 x i8> @llvm.fptoui.sat.nxv1i8.nxv1f64(<vscale x 1 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv1f32_nxv1i16 = call <vscale x 1 x i16> @llvm.fptoui.sat.nxv1i16.nxv1f32(<vscale x 1 x float> poison)
@@ -296,45 +296,45 @@ define void @fptoui_sat() {
define void @fptosi_sat() {
; RV64ZVE32F-LABEL: 'fptosi_sat'
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v1f32_v1i8 = call <1 x i8> @llvm.fptosi.sat.v1i8.v1f32(<1 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v1f64_v1i8 = call <1 x i8> @llvm.fptosi.sat.v1i8.v1f64(<1 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v1f64_v1i8 = call <1 x i8> @llvm.fptosi.sat.v1i8.v1f64(<1 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v1f32_v1i16 = call <1 x i16> @llvm.fptosi.sat.v1i16.v1f32(<1 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v1f64_v1i16 = call <1 x i16> @llvm.fptosi.sat.v1i16.v1f64(<1 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v1f64_v1i16 = call <1 x i16> @llvm.fptosi.sat.v1i16.v1f64(<1 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v1f32_v1i32 = call <1 x i32> @llvm.fptosi.sat.v1i32.v1f32(<1 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v1f64_v1i32 = call <1 x i32> @llvm.fptosi.sat.v1i32.v1f64(<1 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v1f64_v1i32 = call <1 x i32> @llvm.fptosi.sat.v1i32.v1f64(<1 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v1f32_v1i64 = call <1 x i64> @llvm.fptosi.sat.v1i64.v1f32(<1 x float> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v1f64_v1i64 = call <1 x i64> @llvm.fptosi.sat.v1i64.v1f64(<1 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v1f32_v1i1 = call <1 x i1> @llvm.fptosi.sat.v1i1.v1f32(<1 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %v1f64_v1i1 = call <1 x i1> @llvm.fptosi.sat.v1i1.v1f64(<1 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %v1f64_v1i1 = call <1 x i1> @llvm.fptosi.sat.v1i1.v1f64(<1 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v2f32_v2i8 = call <2 x i8> @llvm.fptosi.sat.v2i8.v2f32(<2 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %v2f64_v2i8 = call <2 x i8> @llvm.fptosi.sat.v2i8.v2f64(<2 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %v2f64_v2i8 = call <2 x i8> @llvm.fptosi.sat.v2i8.v2f64(<2 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v2f32_v2i16 = call <2 x i16> @llvm.fptosi.sat.v2i16.v2f32(<2 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %v2f64_v2i16 = call <2 x i16> @llvm.fptosi.sat.v2i16.v2f64(<2 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %v2f64_v2i16 = call <2 x i16> @llvm.fptosi.sat.v2i16.v2f64(<2 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v2f32_v2i32 = call <2 x i32> @llvm.fptosi.sat.v2i32.v2f32(<2 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %v2f64_v2i32 = call <2 x i32> @llvm.fptosi.sat.v2i32.v2f64(<2 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %v2f64_v2i32 = call <2 x i32> @llvm.fptosi.sat.v2i32.v2f64(<2 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %v2f32_v2i64 = call <2 x i64> @llvm.fptosi.sat.v2i64.v2f32(<2 x float> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v2f64_v2i64 = call <2 x i64> @llvm.fptosi.sat.v2i64.v2f64(<2 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v2f32_v2i1 = call <2 x i1> @llvm.fptosi.sat.v2i1.v2f32(<2 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %v2f64_v2i1 = call <2 x i1> @llvm.fptosi.sat.v2i1.v2f64(<2 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %v2f64_v2i1 = call <2 x i1> @llvm.fptosi.sat.v2i1.v2f64(<2 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v4f32_v4i8 = call <4 x i8> @llvm.fptosi.sat.v4i8.v4f32(<4 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %v4f64_v4i8 = call <4 x i8> @llvm.fptosi.sat.v4i8.v4f64(<4 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %v4f64_v4i8 = call <4 x i8> @llvm.fptosi.sat.v4i8.v4f64(<4 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v4f32_v4i16 = call <4 x i16> @llvm.fptosi.sat.v4i16.v4f32(<4 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %v4f64_v4i16 = call <4 x i16> @llvm.fptosi.sat.v4i16.v4f64(<4 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %v4f64_v4i16 = call <4 x i16> @llvm.fptosi.sat.v4i16.v4f64(<4 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v4f32_v4i32 = call <4 x i32> @llvm.fptosi.sat.v4i32.v4f32(<4 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %v4f64_v4i32 = call <4 x i32> @llvm.fptosi.sat.v4i32.v4f64(<4 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %v4f64_v4i32 = call <4 x i32> @llvm.fptosi.sat.v4i32.v4f64(<4 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %v4f32_v4i64 = call <4 x i64> @llvm.fptosi.sat.v4i64.v4f32(<4 x float> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %v4f64_v4i64 = call <4 x i64> @llvm.fptosi.sat.v4i64.v4f64(<4 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v4f32_v4i1 = call <4 x i1> @llvm.fptosi.sat.v4i1.v4f32(<4 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 46 for instruction: %v4f64_v4i1 = call <4 x i1> @llvm.fptosi.sat.v4i1.v4f64(<4 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 50 for instruction: %v4f64_v4i1 = call <4 x i1> @llvm.fptosi.sat.v4i1.v4f64(<4 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %v8f32_v8i8 = call <8 x i8> @llvm.fptosi.sat.v8i8.v8f32(<8 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 40 for instruction: %v8f64_v8i8 = call <8 x i8> @llvm.fptosi.sat.v8i8.v8f64(<8 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 48 for instruction: %v8f64_v8i8 = call <8 x i8> @llvm.fptosi.sat.v8i8.v8f64(<8 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %v8f32_v8i16 = call <8 x i16> @llvm.fptosi.sat.v8i16.v8f32(<8 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 40 for instruction: %v8f64_v8i16 = call <8 x i16> @llvm.fptosi.sat.v8i16.v8f64(<8 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 48 for instruction: %v8f64_v8i16 = call <8 x i16> @llvm.fptosi.sat.v8i16.v8f64(<8 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v8f32_v8i32 = call <8 x i32> @llvm.fptosi.sat.v8i32.v8f32(<8 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 41 for instruction: %v8f64_v8i32 = call <8 x i32> @llvm.fptosi.sat.v8i32.v8f64(<8 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 49 for instruction: %v8f64_v8i32 = call <8 x i32> @llvm.fptosi.sat.v8i32.v8f64(<8 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %v8f32_v8i64 = call <8 x i64> @llvm.fptosi.sat.v8i64.v8f32(<8 x float> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %v8f64_v8i64 = call <8 x i64> @llvm.fptosi.sat.v8i64.v8f64(<8 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %v8f32_v8i1 = call <8 x i1> @llvm.fptosi.sat.v8i1.v8f32(<8 x float> poison)
-; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 90 for instruction: %v8f64_v8i1 = call <8 x i1> @llvm.fptosi.sat.v8i1.v8f64(<8 x double> poison)
+; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 98 for instruction: %v8f64_v8i1 = call <8 x i1> @llvm.fptosi.sat.v8i1.v8f64(<8 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %nxv1f32_nxv1i8 = call <vscale x 1 x i8> @llvm.fptosi.sat.nxv1i8.nxv1f32(<vscale x 1 x float> poison)
; RV64ZVE32F-NEXT: Cost Model: Invalid cost for instruction: %nxv1f64_nxv1i8 = call <vscale x 1 x i8> @llvm.fptosi.sat.nxv1i8.nxv1f64(<vscale x 1 x double> poison)
; RV64ZVE32F-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv1f32_nxv1i16 = call <vscale x 1 x i16> @llvm.fptosi.sat.nxv1i16.nxv1f32(<vscale x 1 x float> poison)
diff --git a/llvm/test/Analysis/CostModel/RISCV/cast.ll b/llvm/test/Analysis/CostModel/RISCV/cast.ll
index 6dacd59f07fde..bfbc8648433f6 100644
--- a/llvm/test/Analysis/CostModel/RISCV/cast.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/cast.ll
@@ -2886,146 +2886,146 @@ define void @fptosi() {
; 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 9 for instruction: %v2f32_v2i8 = fptosi <2 x float> undef to <2 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v2f64_v2i8 = fptosi <2 x double> undef to <2 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v2f32_v2i16 = fptosi <2 x float> undef to <2 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v2f64_v2i16 = fptosi <2 x double> undef to <2 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v2f32_v2i32 = fptosi <2 x float> undef to <2 x i32>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 9 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 21 for instruction: %v2f32_v2i1 = fptosi <2 x float> undef to <2 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %v2f64_v2i1 = fptosi <2 x double> undef to <2 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 9 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 9 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 9 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 9 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 9 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 9 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 21 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 21 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 19 for instruction: %v4f32_v4i8 = fptosi <4 x float> undef to <4 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %v4f64_v4i8 = fptosi <4 x double> undef to <4 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %v4f32_v4i16 = fptosi <4 x float> undef to <4 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %v4f64_v4i16 = fptosi <4 x double> undef to <4 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %v4f32_v4i32 = fptosi <4 x float> undef to <4 x i32>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 19 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 43 for instruction: %v4f32_v4i1 = fptosi <4 x float> undef to <4 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 43 for instruction: %v4f64_v4i1 = fptosi <4 x double> undef to <4 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 19 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 19 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 19 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 19 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 19 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 19 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 43 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 43 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 39 for instruction: %v8f32_v8i8 = fptosi <8 x float> undef to <8 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %v8f64_v8i8 = fptosi <8 x double> undef to <8 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %v8f32_v8i16 = fptosi <8 x float> undef to <8 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %v8f64_v8i16 = fptosi <8 x double> undef to <8 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %v8f32_v8i32 = fptosi <8 x float> undef to <8 x i32>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 39 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 87 for instruction: %v8f32_v8i1 = fptosi <8 x float> undef to <8 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 87 for instruction: %v8f64_v8i1 = fptosi <8 x double> undef to <8 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 39 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 39 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 39 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 39 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 39 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 39 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 87 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 87 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 79 for instruction: %v16f32_v16i8 = fptosi <16 x float> undef to <16 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 79 for instruction: %v16f64_v16i8 = fptosi <16 x double> undef to <16 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 79 for instruction: %v16f32_v16i16 = fptosi <16 x float> undef to <16 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 79 for instruction: %v16f64_v16i16 = fptosi <16 x double> undef to <16 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 79 for instruction: %v16f32_v16i32 = fptosi <16 x float> undef to <16 x i32>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 79 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 175 for instruction: %v16f32_v16i1 = fptosi <16 x float> undef to <16 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 175 for instruction: %v16f64_v16i1 = fptosi <16 x double> undef to <16 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 79 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 79 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 79 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 79 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 79 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 79 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 175 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 175 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 159 for instruction: %v32f32_v32i8 = fptosi <32 x float> undef to <32 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 159 for instruction: %v32f64_v32i8 = fptosi <32 x double> undef to <32 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 159 for instruction: %v32f32_v32i16 = fptosi <32 x float> undef to <32 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 159 for instruction: %v32f64_v32i16 = fptosi <32 x double> undef to <32 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 159 for instruction: %v32f32_v32i32 = fptosi <32 x float> undef to <32 x i32>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 159 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 351 for instruction: %v32f32_v32i1 = fptosi <32 x float> undef to <32 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 351 for instruction: %v32f64_v32i1 = fptosi <32 x double> undef to <32 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 159 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 159 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 159 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 159 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 159 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 159 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 351 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 351 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 319 for instruction: %v64f32_v64i8 = fptosi <64 x float> undef to <64 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 319 for instruction: %v64f64_v64i8 = fptosi <64 x double> undef to <64 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 319 for instruction: %v64f32_v64i16 = fptosi <64 x float> undef to <64 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 319 for instruction: %v64f64_v64i16 = fptosi <64 x double> undef to <64 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 318 for instruction: %v64f32_v64i32 = fptosi <64 x float> undef to <64 x i32>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 318 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 703 for instruction: %v64f32_v64i1 = fptosi <64 x float> undef to <64 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 703 for instruction: %v64f64_v64i1 = fptosi <64 x double> undef to <64 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 319 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 319 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 319 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 319 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 318 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 318 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 703 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 703 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 639 for instruction: %v128f32_v128i8 = fptosi <128 x float> undef to <128 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 639 for instruction: %v128f64_v128i8 = fptosi <128 x double> undef to <128 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 638 for instruction: %v128f32_v128i16 = fptosi <128 x float> undef to <128 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 638 for instruction: %v128f64_v128i16 = fptosi <128 x double> undef to <128 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 636 for instruction: %v128f32_v128i32 = fptosi <128 x float> undef to <128 x i32>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 636 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 1407 for instruction: %v128f32_v128i1 = fptosi <128 x float> undef to <128 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 1407 for instruction: %v128f64_v128i1 = fptosi <128 x double> undef to <128 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 639 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 639 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 638 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 638 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 636 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 636 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: Found an estimated cost of 1407 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 1407 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>
@@ -3764,146 +3764,146 @@ define void @fptoui() {
; 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 9 for instruction: %v2f32_v2i8 = fptoui <2 x float> undef to <2 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v2f64_v2i8 = fptoui <2 x double> undef to <2 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v2f32_v2i16 = fptoui <2 x float> undef to <2 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v2f64_v2i16 = fptoui <2 x double> undef to <2 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v2f32_v2i32 = fptoui <2 x float> undef to <2 x i32>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 9 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 21 for instruction: %v2f32_v2i1 = fptoui <2 x float> undef to <2 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %v2f64_v2i1 = fptoui <2 x double> undef to <2 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 9 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 9 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 9 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 9 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 9 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 9 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 21 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 21 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 19 for instruction: %v4f32_v4i8 = fptoui <4 x float> undef to <4 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %v4f64_v4i8 = fptoui <4 x double> undef to <4 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %v4f32_v4i16 = fptoui <4 x float> undef to <4 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %v4f64_v4i16 = fptoui <4 x double> undef to <4 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %v4f32_v4i32 = fptoui <4 x float> undef to <4 x i32>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 19 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 43 for instruction: %v4f32_v4i1 = fptoui <4 x float> undef to <4 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 43 for instruction: %v4f64_v4i1 = fptoui <4 x double> undef to <4 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 19 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 19 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 19 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 19 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 19 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 19 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 43 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 43 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 39 for instruction: %v8f32_v8i8 = fptoui <8 x float> undef to <8 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %v8f64_v8i8 = fptoui <8 x double> undef to <8 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %v8f32_v8i16 = fptoui <8 x float> undef to <8 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %v8f64_v8i16 = fptoui <8 x double> undef to <8 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %v8f32_v8i32 = fptoui <8 x float> undef to <8 x i32>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 39 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 87 for instruction: %v8f32_v8i1 = fptoui <8 x float> undef to <8 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 87 for instruction: %v8f64_v8i1 = fptoui <8 x double> undef to <8 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 39 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 39 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 39 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 39 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 39 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 39 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 87 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 87 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 79 for instruction: %v16f32_v16i8 = fptoui <16 x float> undef to <16 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 79 for instruction: %v16f64_v16i8 = fptoui <16 x double> undef to <16 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 79 for instruction: %v16f32_v16i16 = fptoui <16 x float> undef to <16 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 79 for instruction: %v16f64_v16i16 = fptoui <16 x double> undef to <16 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 79 for instruction: %v16f32_v16i32 = fptoui <16 x float> undef to <16 x i32>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 79 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 175 for instruction: %v16f32_v16i1 = fptoui <16 x float> undef to <16 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 175 for instruction: %v16f64_v16i1 = fptoui <16 x double> undef to <16 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 79 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 79 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 79 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 79 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 79 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 79 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 175 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 175 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 159 for instruction: %v32f32_v32i8 = fptoui <32 x float> undef to <32 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 159 for instruction: %v32f64_v32i8 = fptoui <32 x double> undef to <32 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 159 for instruction: %v32f32_v32i16 = fptoui <32 x float> undef to <32 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 159 for instruction: %v32f64_v32i16 = fptoui <32 x double> undef to <32 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 159 for instruction: %v32f32_v32i32 = fptoui <32 x float> undef to <32 x i32>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 159 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 351 for instruction: %v32f32_v32i1 = fptoui <32 x float> undef to <32 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 351 for instruction: %v32f64_v32i1 = fptoui <32 x double> undef to <32 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 159 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 159 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 159 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 159 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 159 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 159 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 351 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 351 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 319 for instruction: %v64f32_v64i8 = fptoui <64 x float> undef to <64 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 319 for instruction: %v64f64_v64i8 = fptoui <64 x double> undef to <64 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 319 for instruction: %v64f32_v64i16 = fptoui <64 x float> undef to <64 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 319 for instruction: %v64f64_v64i16 = fptoui <64 x double> undef to <64 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 318 for instruction: %v64f32_v64i32 = fptoui <64 x float> undef to <64 x i32>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 318 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 703 for instruction: %v64f32_v64i1 = fptoui <64 x float> undef to <64 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 703 for instruction: %v64f64_v64i1 = fptoui <64 x double> undef to <64 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 319 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 319 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 319 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 319 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 318 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 318 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 703 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 703 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 639 for instruction: %v128f32_v128i8 = fptoui <128 x float> undef to <128 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 639 for instruction: %v128f64_v128i8 = fptoui <128 x double> undef to <128 x i8>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 638 for instruction: %v128f32_v128i16 = fptoui <128 x float> undef to <128 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 638 for instruction: %v128f64_v128i16 = fptoui <128 x double> undef to <128 x i16>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 636 for instruction: %v128f32_v128i32 = fptoui <128 x float> undef to <128 x i32>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 636 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 1407 for instruction: %v128f32_v128i1 = fptoui <128 x float> undef to <128 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 1407 for instruction: %v128f64_v128i1 = fptoui <128 x double> undef to <128 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 639 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 639 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 638 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 638 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 636 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 636 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: Found an estimated cost of 1407 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 1407 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>
@@ -6160,12 +6160,12 @@ define void @oddvec_sizes() {
; 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 16 for instruction: %19 = fptosi <3 x float> undef to <3 x i32>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 40 for instruction: %20 = fptosi <7 x float> undef to <7 x i32>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %21 = fptosi <15 x float> undef to <15 x i32>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %22 = fptoui <3 x float> undef to <3 x i32>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 40 for instruction: %23 = fptoui <7 x float> undef to <7 x i32>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 88 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>
@@ -6208,25 +6208,25 @@ define void @oddvec_sizes() {
; 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 16892 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 1140 for instruction: %5 = sitofp <192 x i1> undef to <192 x float>
+; ZVE64D-NEXT: Cost Model: Found an estimated cost of 1140 for instruction: %6 = uitofp <192 x i1> undef to <192 x float>
+; ZVE64D-NEXT: Cost Model: Found an estimated cost of 17084 for instruction: %7 = fptosi <192 x float> undef to <192 x i1>
+; ZVE64D-NEXT: Cost Model: Found an estimated cost of 17084 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 16892 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 17084 for instruction: %7 = fptosi <192 x float> undef to <192 x i1>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 17084 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>
@@ -6243,7 +6243,7 @@ define void @legalization_crash() {
; Test that types that need to be split go through BasicTTIImpl.
define void @BitInt_crash() {
; ZVE64X-LABEL: 'BitInt_crash'
-; ZVE64X-NEXT: Cost Model: Found an estimated cost of 2043 for instruction: %1 = bitcast <16 x i64> poison to <512 x i2>
+; ZVE64X-NEXT: Cost Model: Found an estimated cost of 2555 for instruction: %1 = bitcast <16 x i64> poison to <512 x i2>
; ZVE64X-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
bitcast <16 x i64> poison to <512 x i2>
diff --git a/llvm/test/Analysis/CostModel/RISCV/clmul.ll b/llvm/test/Analysis/CostModel/RISCV/clmul.ll
index a0e87b2dbeb99..6b1918f19dcb7 100644
--- a/llvm/test/Analysis/CostModel/RISCV/clmul.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/clmul.ll
@@ -48,11 +48,11 @@ define void @clmul() {
; RV32ZVBC-NEXT: Cost Model: Found costs of RThru:0 CodeSize:1 Lat:1 SizeLat:1 for: ret void
;
; RV64ZVBC-LABEL: 'clmul'
-; RV64ZVBC-NEXT: Cost Model: Found costs of 6 for: %call_i128 = call i128 @llvm.clmul.i128(i128 poison, i128 poison)
-; RV64ZVBC-NEXT: Cost Model: Found costs of 3 for: %call_i64 = call i64 @llvm.clmul.i64(i64 poison, i64 poison)
-; RV64ZVBC-NEXT: Cost Model: Found costs of 3 for: %call_i32 = call i32 @llvm.clmul.i32(i32 poison, i32 poison)
-; RV64ZVBC-NEXT: Cost Model: Found costs of 3 for: %call_i16 = call i16 @llvm.clmul.i16(i16 poison, i16 poison)
-; RV64ZVBC-NEXT: Cost Model: Found costs of 3 for: %call_i8 = call i8 @llvm.clmul.i8(i8 poison, i8 poison)
+; RV64ZVBC-NEXT: Cost Model: Found costs of RThru:8 CodeSize:6 Lat:8 SizeLat:6 for: %call_i128 = call i128 @llvm.clmul.i128(i128 poison, i128 poison)
+; RV64ZVBC-NEXT: Cost Model: Found costs of RThru:4 CodeSize:3 Lat:4 SizeLat:3 for: %call_i64 = call i64 @llvm.clmul.i64(i64 poison, i64 poison)
+; RV64ZVBC-NEXT: Cost Model: Found costs of RThru:4 CodeSize:3 Lat:4 SizeLat:3 for: %call_i32 = call i32 @llvm.clmul.i32(i32 poison, i32 poison)
+; RV64ZVBC-NEXT: Cost Model: Found costs of RThru:4 CodeSize:3 Lat:4 SizeLat:3 for: %call_i16 = call i16 @llvm.clmul.i16(i16 poison, i16 poison)
+; RV64ZVBC-NEXT: Cost Model: Found costs of RThru:4 CodeSize:3 Lat:4 SizeLat:3 for: %call_i8 = call i8 @llvm.clmul.i8(i8 poison, i8 poison)
; RV64ZVBC-NEXT: Cost Model: Found costs of RThru:0 CodeSize:1 Lat:1 SizeLat:1 for: ret void
;
%call_i128 = call i128 @llvm.clmul.i128(i128 poison, i128 poison)
diff --git a/llvm/test/Analysis/CostModel/RISCV/cttz_elts.ll b/llvm/test/Analysis/CostModel/RISCV/cttz_elts.ll
index 094d73ddd0581..8ac08c311d02e 100644
--- a/llvm/test/Analysis/CostModel/RISCV/cttz_elts.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/cttz_elts.ll
@@ -9,28 +9,28 @@ define void @foo_no_vscale_range() {
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %res.i64.nxv16i1.zip = call i64 @llvm.experimental.cttz.elts.i64.nxv16i1(<vscale x 16 x i1> undef, i1 true)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %res.i64.nxv32i1.zip = call i64 @llvm.experimental.cttz.elts.i64.nxv32i1(<vscale x 32 x i1> undef, i1 true)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %res.i64.nxv64i1.zip = call i64 @llvm.experimental.cttz.elts.i64.nxv64i1(<vscale x 64 x i1> undef, i1 true)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 661 for instruction: %res.i64.nxv128i1.zip = call i64 @llvm.experimental.cttz.elts.i64.nxv128i1(<vscale x 128 x i1> undef, i1 true)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 662 for instruction: %res.i64.nxv128i1.zip = call i64 @llvm.experimental.cttz.elts.i64.nxv128i1(<vscale x 128 x i1> undef, i1 true)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %res.i32.nxv2i1.zip = call i32 @llvm.experimental.cttz.elts.i32.nxv2i1(<vscale x 2 x i1> undef, i1 true)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %res.i32.nxv4i1.zip = call i32 @llvm.experimental.cttz.elts.i32.nxv4i1(<vscale x 4 x i1> undef, i1 true)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %res.i32.nxv8i1.zip = call i32 @llvm.experimental.cttz.elts.i32.nxv8i1(<vscale x 8 x i1> undef, i1 true)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %res.i32.nxv16i1.zip = call i32 @llvm.experimental.cttz.elts.i32.nxv16i1(<vscale x 16 x i1> undef, i1 true)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %res.i32.nxv32i1.zip = call i32 @llvm.experimental.cttz.elts.i32.nxv32i1(<vscale x 32 x i1> undef, i1 true)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %res.i32.nxv64i1.zip = call i32 @llvm.experimental.cttz.elts.i32.nxv64i1(<vscale x 64 x i1> undef, i1 true)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 334 for instruction: %res.i32.nxv128i1.zip = call i32 @llvm.experimental.cttz.elts.i32.nxv128i1(<vscale x 128 x i1> undef, i1 true)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 335 for instruction: %res.i32.nxv128i1.zip = call i32 @llvm.experimental.cttz.elts.i32.nxv128i1(<vscale x 128 x i1> undef, i1 true)
; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %res.i64.nxv2i1.nzip = call i64 @llvm.experimental.cttz.elts.i64.nxv2i1(<vscale x 2 x i1> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %res.i64.nxv4i1.nzip = call i64 @llvm.experimental.cttz.elts.i64.nxv4i1(<vscale x 4 x i1> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %res.i64.nxv8i1.nzip = call i64 @llvm.experimental.cttz.elts.i64.nxv8i1(<vscale x 8 x i1> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %res.i64.nxv16i1.nzip = call i64 @llvm.experimental.cttz.elts.i64.nxv16i1(<vscale x 16 x i1> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %res.i64.nxv32i1.nzip = call i64 @llvm.experimental.cttz.elts.i64.nxv32i1(<vscale x 32 x i1> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %res.i64.nxv64i1.nzip = call i64 @llvm.experimental.cttz.elts.i64.nxv64i1(<vscale x 64 x i1> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 661 for instruction: %res.i64.nxv128i1.nzip = call i64 @llvm.experimental.cttz.elts.i64.nxv128i1(<vscale x 128 x i1> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 662 for instruction: %res.i64.nxv128i1.nzip = call i64 @llvm.experimental.cttz.elts.i64.nxv128i1(<vscale x 128 x i1> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %res.i32.nxv2i1.nzip = call i32 @llvm.experimental.cttz.elts.i32.nxv2i1(<vscale x 2 x i1> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %res.i32.nxv4i1.nzip = call i32 @llvm.experimental.cttz.elts.i32.nxv4i1(<vscale x 4 x i1> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %res.i32.nxv8i1.nzip = call i32 @llvm.experimental.cttz.elts.i32.nxv8i1(<vscale x 8 x i1> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %res.i32.nxv16i1.nzip = call i32 @llvm.experimental.cttz.elts.i32.nxv16i1(<vscale x 16 x i1> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %res.i32.nxv32i1.nzip = call i32 @llvm.experimental.cttz.elts.i32.nxv32i1(<vscale x 32 x i1> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %res.i32.nxv64i1.nzip = call i32 @llvm.experimental.cttz.elts.i32.nxv64i1(<vscale x 64 x i1> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 334 for instruction: %res.i32.nxv128i1.nzip = call i32 @llvm.experimental.cttz.elts.i32.nxv128i1(<vscale x 128 x i1> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 335 for instruction: %res.i32.nxv128i1.nzip = call i32 @llvm.experimental.cttz.elts.i32.nxv128i1(<vscale x 128 x i1> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%res.i64.nxv2i1.zip = call i64 @llvm.experimental.cttz.elts.i64.nxv2i1(<vscale x 2 x i1> undef, i1 true)
@@ -75,28 +75,28 @@ define void @foo_vscale_range_2_16() vscale_range(2,16) {
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %res.i64.nxv16i1.zip = call i64 @llvm.experimental.cttz.elts.i64.nxv16i1(<vscale x 16 x i1> undef, i1 true)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %res.i64.nxv32i1.zip = call i64 @llvm.experimental.cttz.elts.i64.nxv32i1(<vscale x 32 x i1> undef, i1 true)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %res.i64.nxv64i1.zip = call i64 @llvm.experimental.cttz.elts.i64.nxv64i1(<vscale x 64 x i1> undef, i1 true)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 171 for instruction: %res.i64.nxv128i1.zip = call i64 @llvm.experimental.cttz.elts.i64.nxv128i1(<vscale x 128 x i1> undef, i1 true)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 172 for instruction: %res.i64.nxv128i1.zip = call i64 @llvm.experimental.cttz.elts.i64.nxv128i1(<vscale x 128 x i1> undef, i1 true)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %res.i32.nxv2i1.zip = call i32 @llvm.experimental.cttz.elts.i32.nxv2i1(<vscale x 2 x i1> undef, i1 true)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %res.i32.nxv4i1.zip = call i32 @llvm.experimental.cttz.elts.i32.nxv4i1(<vscale x 4 x i1> undef, i1 true)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %res.i32.nxv8i1.zip = call i32 @llvm.experimental.cttz.elts.i32.nxv8i1(<vscale x 8 x i1> undef, i1 true)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %res.i32.nxv16i1.zip = call i32 @llvm.experimental.cttz.elts.i32.nxv16i1(<vscale x 16 x i1> undef, i1 true)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %res.i32.nxv32i1.zip = call i32 @llvm.experimental.cttz.elts.i32.nxv32i1(<vscale x 32 x i1> undef, i1 true)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %res.i32.nxv64i1.zip = call i32 @llvm.experimental.cttz.elts.i32.nxv64i1(<vscale x 64 x i1> undef, i1 true)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 171 for instruction: %res.i32.nxv128i1.zip = call i32 @llvm.experimental.cttz.elts.i32.nxv128i1(<vscale x 128 x i1> undef, i1 true)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 172 for instruction: %res.i32.nxv128i1.zip = call i32 @llvm.experimental.cttz.elts.i32.nxv128i1(<vscale x 128 x i1> undef, i1 true)
; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %res.i64.nxv2i1.nzip = call i64 @llvm.experimental.cttz.elts.i64.nxv2i1(<vscale x 2 x i1> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %res.i64.nxv4i1.nzip = call i64 @llvm.experimental.cttz.elts.i64.nxv4i1(<vscale x 4 x i1> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %res.i64.nxv8i1.nzip = call i64 @llvm.experimental.cttz.elts.i64.nxv8i1(<vscale x 8 x i1> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %res.i64.nxv16i1.nzip = call i64 @llvm.experimental.cttz.elts.i64.nxv16i1(<vscale x 16 x i1> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %res.i64.nxv32i1.nzip = call i64 @llvm.experimental.cttz.elts.i64.nxv32i1(<vscale x 32 x i1> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %res.i64.nxv64i1.nzip = call i64 @llvm.experimental.cttz.elts.i64.nxv64i1(<vscale x 64 x i1> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 171 for instruction: %res.i64.nxv128i1.nzip = call i64 @llvm.experimental.cttz.elts.i64.nxv128i1(<vscale x 128 x i1> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 172 for instruction: %res.i64.nxv128i1.nzip = call i64 @llvm.experimental.cttz.elts.i64.nxv128i1(<vscale x 128 x i1> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %res.i32.nxv2i1.nzip = call i32 @llvm.experimental.cttz.elts.i32.nxv2i1(<vscale x 2 x i1> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %res.i32.nxv4i1.nzip = call i32 @llvm.experimental.cttz.elts.i32.nxv4i1(<vscale x 4 x i1> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %res.i32.nxv8i1.nzip = call i32 @llvm.experimental.cttz.elts.i32.nxv8i1(<vscale x 8 x i1> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %res.i32.nxv16i1.nzip = call i32 @llvm.experimental.cttz.elts.i32.nxv16i1(<vscale x 16 x i1> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %res.i32.nxv32i1.nzip = call i32 @llvm.experimental.cttz.elts.i32.nxv32i1(<vscale x 32 x i1> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %res.i32.nxv64i1.nzip = call i32 @llvm.experimental.cttz.elts.i32.nxv64i1(<vscale x 64 x i1> undef, i1 false)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 171 for instruction: %res.i32.nxv128i1.nzip = call i32 @llvm.experimental.cttz.elts.i32.nxv128i1(<vscale x 128 x i1> undef, i1 false)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 172 for instruction: %res.i32.nxv128i1.nzip = call i32 @llvm.experimental.cttz.elts.i32.nxv128i1(<vscale x 128 x i1> undef, i1 false)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%res.i64.nxv2i1.zip = call i64 @llvm.experimental.cttz.elts.i64.nxv2i1(<vscale x 2 x i1> undef, i1 true)
diff --git a/llvm/test/Analysis/CostModel/RISCV/extract-last-active.ll b/llvm/test/Analysis/CostModel/RISCV/extract-last-active.ll
index 84f18e3df9015..065383e2f9d56 100644
--- a/llvm/test/Analysis/CostModel/RISCV/extract-last-active.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/extract-last-active.ll
@@ -3,54 +3,54 @@
define void @extractions() {
; CHECK-LABEL: 'extractions'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %v16i8 = call i8 @llvm.experimental.vector.extract.last.active.v16i8(<16 x i8> poison, <16 x i1> poison, i8 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v8i16 = call i16 @llvm.experimental.vector.extract.last.active.v8i16(<8 x i16> poison, <8 x i1> poison, i16 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %v4i32 = call i32 @llvm.experimental.vector.extract.last.active.v4i32(<4 x i32> poison, <4 x i1> poison, i32 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %v2i64 = call i64 @llvm.experimental.vector.extract.last.active.v2i64(<2 x i64> poison, <2 x i1> poison, i64 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v8f16 = call half @llvm.experimental.vector.extract.last.active.v8f16(<8 x half> poison, <8 x i1> poison, half poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v8bf16 = call bfloat @llvm.experimental.vector.extract.last.active.v8bf16(<8 x bfloat> poison, <8 x i1> poison, bfloat poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %v4f32 = call float @llvm.experimental.vector.extract.last.active.v4f32(<4 x float> poison, <4 x i1> poison, float poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %v2f64 = call double @llvm.experimental.vector.extract.last.active.v2f64(<2 x double> poison, <2 x i1> poison, double poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %v16i8 = call i8 @llvm.experimental.vector.extract.last.active.v16i8(<16 x i8> poison, <16 x i1> poison, i8 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %v8i16 = call i16 @llvm.experimental.vector.extract.last.active.v8i16(<8 x i16> poison, <8 x i1> poison, i16 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %v4i32 = call i32 @llvm.experimental.vector.extract.last.active.v4i32(<4 x i32> poison, <4 x i1> poison, i32 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v2i64 = call i64 @llvm.experimental.vector.extract.last.active.v2i64(<2 x i64> poison, <2 x i1> poison, i64 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %v8f16 = call half @llvm.experimental.vector.extract.last.active.v8f16(<8 x half> poison, <8 x i1> poison, half poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %v8bf16 = call bfloat @llvm.experimental.vector.extract.last.active.v8bf16(<8 x bfloat> poison, <8 x i1> poison, bfloat poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %v4f32 = call float @llvm.experimental.vector.extract.last.active.v4f32(<4 x float> poison, <4 x i1> poison, float poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v2f64 = call double @llvm.experimental.vector.extract.last.active.v2f64(<2 x double> poison, <2 x i1> poison, double poison)
; CHECK-NEXT: Cost Model: Invalid cost for instruction: %nxv16i8 = call i8 @llvm.experimental.vector.extract.last.active.nxv16i8(<vscale x 16 x i8> poison, <vscale x 16 x i1> poison, i8 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %nxv8i16 = call i16 @llvm.experimental.vector.extract.last.active.nxv8i16(<vscale x 8 x i16> poison, <vscale x 8 x i1> poison, i16 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %nxv4i32 = call i32 @llvm.experimental.vector.extract.last.active.nxv4i32(<vscale x 4 x i32> poison, <vscale x 4 x i1> poison, i32 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %nxv2i64 = call i64 @llvm.experimental.vector.extract.last.active.nxv2i64(<vscale x 2 x i64> poison, <vscale x 2 x i1> poison, i64 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %nxv8f16 = call half @llvm.experimental.vector.extract.last.active.nxv8f16(<vscale x 8 x half> poison, <vscale x 8 x i1> poison, half poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %nxv8bf16 = call bfloat @llvm.experimental.vector.extract.last.active.nxv8bf16(<vscale x 8 x bfloat> poison, <vscale x 8 x i1> poison, bfloat poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %nxv4f32 = call float @llvm.experimental.vector.extract.last.active.nxv4f32(<vscale x 4 x float> poison, <vscale x 4 x i1> poison, float poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %nxv2f64 = call double @llvm.experimental.vector.extract.last.active.nxv2f64(<vscale x 2 x double> poison, <vscale x 2 x i1> poison, double poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %v32i8 = call i8 @llvm.experimental.vector.extract.last.active.v32i8(<32 x i8> poison, <32 x i1> poison, i8 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %v16i16 = call i16 @llvm.experimental.vector.extract.last.active.v16i16(<16 x i16> poison, <16 x i1> poison, i16 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %v8i32 = call i32 @llvm.experimental.vector.extract.last.active.v8i32(<8 x i32> poison, <8 x i1> poison, i32 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v4i64 = call i64 @llvm.experimental.vector.extract.last.active.v4i64(<4 x i64> poison, <4 x i1> poison, i64 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %v16f16 = call half @llvm.experimental.vector.extract.last.active.v16f16(<16 x half> poison, <16 x i1> poison, half poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %v16bf16 = call bfloat @llvm.experimental.vector.extract.last.active.v16bf16(<16 x bfloat> poison, <16 x i1> poison, bfloat poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %v8f32 = call float @llvm.experimental.vector.extract.last.active.v8f32(<8 x float> poison, <8 x i1> poison, float poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v4f64 = call double @llvm.experimental.vector.extract.last.active.v4f64(<4 x double> poison, <4 x i1> poison, double poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %nxv8i16 = call i16 @llvm.experimental.vector.extract.last.active.nxv8i16(<vscale x 8 x i16> poison, <vscale x 8 x i1> poison, i16 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %nxv4i32 = call i32 @llvm.experimental.vector.extract.last.active.nxv4i32(<vscale x 4 x i32> poison, <vscale x 4 x i1> poison, i32 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %nxv2i64 = call i64 @llvm.experimental.vector.extract.last.active.nxv2i64(<vscale x 2 x i64> poison, <vscale x 2 x i1> poison, i64 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %nxv8f16 = call half @llvm.experimental.vector.extract.last.active.nxv8f16(<vscale x 8 x half> poison, <vscale x 8 x i1> poison, half poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %nxv8bf16 = call bfloat @llvm.experimental.vector.extract.last.active.nxv8bf16(<vscale x 8 x bfloat> poison, <vscale x 8 x i1> poison, bfloat poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %nxv4f32 = call float @llvm.experimental.vector.extract.last.active.nxv4f32(<vscale x 4 x float> poison, <vscale x 4 x i1> poison, float poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %nxv2f64 = call double @llvm.experimental.vector.extract.last.active.nxv2f64(<vscale x 2 x double> poison, <vscale x 2 x i1> poison, double poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %v32i8 = call i8 @llvm.experimental.vector.extract.last.active.v32i8(<32 x i8> poison, <32 x i1> poison, i8 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %v16i16 = call i16 @llvm.experimental.vector.extract.last.active.v16i16(<16 x i16> poison, <16 x i1> poison, i16 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %v8i32 = call i32 @llvm.experimental.vector.extract.last.active.v8i32(<8 x i32> poison, <8 x i1> poison, i32 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %v4i64 = call i64 @llvm.experimental.vector.extract.last.active.v4i64(<4 x i64> poison, <4 x i1> poison, i64 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %v16f16 = call half @llvm.experimental.vector.extract.last.active.v16f16(<16 x half> poison, <16 x i1> poison, half poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %v16bf16 = call bfloat @llvm.experimental.vector.extract.last.active.v16bf16(<16 x bfloat> poison, <16 x i1> poison, bfloat poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %v8f32 = call float @llvm.experimental.vector.extract.last.active.v8f32(<8 x float> poison, <8 x i1> poison, float poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %v4f64 = call double @llvm.experimental.vector.extract.last.active.v4f64(<4 x double> poison, <4 x i1> poison, double poison)
; CHECK-NEXT: Cost Model: Invalid cost for instruction: %nxv32i8 = call i8 @llvm.experimental.vector.extract.last.active.nxv32i8(<vscale x 32 x i8> poison, <vscale x 32 x i1> poison, i8 poison)
; CHECK-NEXT: Cost Model: Invalid cost for instruction: %nxv16i16 = call i16 @llvm.experimental.vector.extract.last.active.nxv16i16(<vscale x 16 x i16> poison, <vscale x 16 x i1> poison, i16 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %nxv8i32 = call i32 @llvm.experimental.vector.extract.last.active.nxv8i32(<vscale x 8 x i32> poison, <vscale x 8 x i1> poison, i32 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %nxv4i64 = call i64 @llvm.experimental.vector.extract.last.active.nxv4i64(<vscale x 4 x i64> poison, <vscale x 4 x i1> poison, i64 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %nxv8i32 = call i32 @llvm.experimental.vector.extract.last.active.nxv8i32(<vscale x 8 x i32> poison, <vscale x 8 x i1> poison, i32 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %nxv4i64 = call i64 @llvm.experimental.vector.extract.last.active.nxv4i64(<vscale x 4 x i64> poison, <vscale x 4 x i1> poison, i64 poison)
; CHECK-NEXT: Cost Model: Invalid cost for instruction: %nxv16f16 = call half @llvm.experimental.vector.extract.last.active.nxv16f16(<vscale x 16 x half> poison, <vscale x 16 x i1> poison, half poison)
; CHECK-NEXT: Cost Model: Invalid cost for instruction: %nxv16bf16 = call bfloat @llvm.experimental.vector.extract.last.active.nxv16bf16(<vscale x 16 x bfloat> poison, <vscale x 16 x i1> poison, bfloat poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %nxv8f32 = call float @llvm.experimental.vector.extract.last.active.nxv8f32(<vscale x 8 x float> poison, <vscale x 8 x i1> poison, float poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %nxv4f64 = call double @llvm.experimental.vector.extract.last.active.nxv4f64(<vscale x 4 x double> poison, <vscale x 4 x i1> poison, double poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v8i8 = call i8 @llvm.experimental.vector.extract.last.active.v8i8(<8 x i8> poison, <8 x i1> poison, i8 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %v4i16 = call i16 @llvm.experimental.vector.extract.last.active.v4i16(<4 x i16> poison, <4 x i1> poison, i16 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %v2i32 = call i32 @llvm.experimental.vector.extract.last.active.v2i32(<2 x i32> poison, <2 x i1> poison, i32 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v1i64 = call i64 @llvm.experimental.vector.extract.last.active.v1i64(<1 x i64> poison, <1 x i1> poison, i64 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %v4f16 = call half @llvm.experimental.vector.extract.last.active.v4f16(<4 x half> poison, <4 x i1> poison, half poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %v4bf16 = call bfloat @llvm.experimental.vector.extract.last.active.v4bf16(<4 x bfloat> poison, <4 x i1> poison, bfloat poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %v2f32 = call float @llvm.experimental.vector.extract.last.active.v2f32(<2 x float> poison, <2 x i1> poison, float poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v1f64 = call double @llvm.experimental.vector.extract.last.active.v1f64(<1 x double> poison, <1 x i1> poison, double poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %nxv8i8 = call i8 @llvm.experimental.vector.extract.last.active.nxv8i8(<vscale x 8 x i8> poison, <vscale x 8 x i1> poison, i8 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %nxv4i16 = call i16 @llvm.experimental.vector.extract.last.active.nxv4i16(<vscale x 4 x i16> poison, <vscale x 4 x i1> poison, i16 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %nxv2i32 = call i32 @llvm.experimental.vector.extract.last.active.nxv2i32(<vscale x 2 x i32> poison, <vscale x 2 x i1> poison, i32 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %nxv1i64 = call i64 @llvm.experimental.vector.extract.last.active.nxv1i64(<vscale x 1 x i64> poison, <vscale x 1 x i1> poison, i64 poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %nxv4f16 = call half @llvm.experimental.vector.extract.last.active.nxv4f16(<vscale x 4 x half> poison, <vscale x 4 x i1> poison, half poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %nxv4bf16 = call bfloat @llvm.experimental.vector.extract.last.active.nxv4bf16(<vscale x 4 x bfloat> poison, <vscale x 4 x i1> poison, bfloat poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %nxv2f32 = call float @llvm.experimental.vector.extract.last.active.nxv2f32(<vscale x 2 x float> poison, <vscale x 2 x i1> poison, float poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %nxv1f64 = call double @llvm.experimental.vector.extract.last.active.nxv1f64(<vscale x 1 x double> poison, <vscale x 1 x i1> poison, double poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %nxv8f32 = call float @llvm.experimental.vector.extract.last.active.nxv8f32(<vscale x 8 x float> poison, <vscale x 8 x i1> poison, float poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %nxv4f64 = call double @llvm.experimental.vector.extract.last.active.nxv4f64(<vscale x 4 x double> poison, <vscale x 4 x i1> poison, double poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %v8i8 = call i8 @llvm.experimental.vector.extract.last.active.v8i8(<8 x i8> poison, <8 x i1> poison, i8 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %v4i16 = call i16 @llvm.experimental.vector.extract.last.active.v4i16(<4 x i16> poison, <4 x i1> poison, i16 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v2i32 = call i32 @llvm.experimental.vector.extract.last.active.v2i32(<2 x i32> poison, <2 x i1> poison, i32 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %v1i64 = call i64 @llvm.experimental.vector.extract.last.active.v1i64(<1 x i64> poison, <1 x i1> poison, i64 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %v4f16 = call half @llvm.experimental.vector.extract.last.active.v4f16(<4 x half> poison, <4 x i1> poison, half poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %v4bf16 = call bfloat @llvm.experimental.vector.extract.last.active.v4bf16(<4 x bfloat> poison, <4 x i1> poison, bfloat poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v2f32 = call float @llvm.experimental.vector.extract.last.active.v2f32(<2 x float> poison, <2 x i1> poison, float poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %v1f64 = call double @llvm.experimental.vector.extract.last.active.v1f64(<1 x double> poison, <1 x i1> poison, double poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %nxv8i8 = call i8 @llvm.experimental.vector.extract.last.active.nxv8i8(<vscale x 8 x i8> poison, <vscale x 8 x i1> poison, i8 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %nxv4i16 = call i16 @llvm.experimental.vector.extract.last.active.nxv4i16(<vscale x 4 x i16> poison, <vscale x 4 x i1> poison, i16 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %nxv2i32 = call i32 @llvm.experimental.vector.extract.last.active.nxv2i32(<vscale x 2 x i32> poison, <vscale x 2 x i1> poison, i32 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %nxv1i64 = call i64 @llvm.experimental.vector.extract.last.active.nxv1i64(<vscale x 1 x i64> poison, <vscale x 1 x i1> poison, i64 poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %nxv4f16 = call half @llvm.experimental.vector.extract.last.active.nxv4f16(<vscale x 4 x half> poison, <vscale x 4 x i1> poison, half poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %nxv4bf16 = call bfloat @llvm.experimental.vector.extract.last.active.nxv4bf16(<vscale x 4 x bfloat> poison, <vscale x 4 x i1> poison, bfloat poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %nxv2f32 = call float @llvm.experimental.vector.extract.last.active.nxv2f32(<vscale x 2 x float> poison, <vscale x 2 x i1> poison, float poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %nxv1f64 = call double @llvm.experimental.vector.extract.last.active.nxv1f64(<vscale x 1 x double> poison, <vscale x 1 x i1> poison, double poison)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%v16i8 = call i8 @llvm.experimental.vector.extract.last.active.v16i8(<16 x i8> poison, <16 x i1> poison, i8 poison)
diff --git a/llvm/test/Analysis/CostModel/RISCV/fixed-vector-gather.ll b/llvm/test/Analysis/CostModel/RISCV/fixed-vector-gather.ll
index 50de9103be406..1c5709acdebb2 100644
--- a/llvm/test/Analysis/CostModel/RISCV/fixed-vector-gather.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/fixed-vector-gather.ll
@@ -49,42 +49,42 @@ define i32 @masked_gather() {
; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4I8 = call <4 x i8> @llvm.masked.gather.v4i8.v4p0(<4 x ptr> align 1 undef, <4 x i1> undef, <4 x i8> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2I8 = call <2 x i8> @llvm.masked.gather.v2i8.v2p0(<2 x ptr> align 1 undef, <2 x i1> undef, <2 x i8> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1I8 = call <1 x i8> @llvm.masked.gather.v1i8.v1p0(<1 x ptr> align 1 undef, <1 x i1> undef, <1 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 38 for instruction: %V8F64.u = call <8 x double> @llvm.masked.gather.v8f64.v8p0(<8 x ptr> align 2 undef, <8 x i1> undef, <8 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %V4F64.u = call <4 x double> @llvm.masked.gather.v4f64.v4p0(<4 x ptr> align 2 undef, <4 x i1> undef, <4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V2F64.u = call <2 x double> @llvm.masked.gather.v2f64.v2p0(<2 x ptr> align 2 undef, <2 x i1> undef, <2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1F64.u = call <1 x double> @llvm.masked.gather.v1f64.v1p0(<1 x ptr> align 2 undef, <1 x i1> undef, <1 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 78 for instruction: %V16F32.u = call <16 x float> @llvm.masked.gather.v16f32.v16p0(<16 x ptr> align 2 undef, <16 x i1> undef, <16 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 38 for instruction: %V8F32.u = call <8 x float> @llvm.masked.gather.v8f32.v8p0(<8 x ptr> align 2 undef, <8 x i1> undef, <8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V4F32.u = call <4 x float> @llvm.masked.gather.v4f32.v4p0(<4 x ptr> align 2 undef, <4 x i1> undef, <4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V2F32.u = call <2 x float> @llvm.masked.gather.v2f32.v2p0(<2 x ptr> align 2 undef, <2 x i1> undef, <2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1F32.u = call <1 x float> @llvm.masked.gather.v1f32.v1p0(<1 x ptr> align 2 undef, <1 x i1> undef, <1 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 157 for instruction: %V32BF16.u = call <32 x bfloat> @llvm.masked.gather.v32bf16.v32p0(<32 x ptr> align 1 undef, <32 x i1> undef, <32 x bfloat> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 78 for instruction: %V16BF16.u = call <16 x bfloat> @llvm.masked.gather.v16bf16.v16p0(<16 x ptr> align 1 undef, <16 x i1> undef, <16 x bfloat> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V8BF16.u = call <8 x bfloat> @llvm.masked.gather.v8bf16.v8p0(<8 x ptr> align 1 undef, <8 x i1> undef, <8 x bfloat> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V4BF16.u = call <4 x bfloat> @llvm.masked.gather.v4bf16.v4p0(<4 x ptr> align 1 undef, <4 x i1> undef, <4 x bfloat> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V2BF16.u = call <2 x bfloat> @llvm.masked.gather.v2bf16.v2p0(<2 x ptr> align 1 undef, <2 x i1> undef, <2 x bfloat> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1BF16.u = call <1 x bfloat> @llvm.masked.gather.v1bf16.v1p0(<1 x ptr> align 1 undef, <1 x i1> undef, <1 x bfloat> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 157 for instruction: %V32F16.u = call <32 x half> @llvm.masked.gather.v32f16.v32p0(<32 x ptr> align 1 undef, <32 x i1> undef, <32 x half> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 78 for instruction: %V16F16.u = call <16 x half> @llvm.masked.gather.v16f16.v16p0(<16 x ptr> align 1 undef, <16 x i1> undef, <16 x half> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V8F16.u = call <8 x half> @llvm.masked.gather.v8f16.v8p0(<8 x ptr> align 1 undef, <8 x i1> undef, <8 x half> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V4F16.u = call <4 x half> @llvm.masked.gather.v4f16.v4p0(<4 x ptr> align 1 undef, <4 x i1> undef, <4 x half> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V2F16.u = call <2 x half> @llvm.masked.gather.v2f16.v2p0(<2 x ptr> align 1 undef, <2 x i1> undef, <2 x half> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1F16.u = call <1 x half> @llvm.masked.gather.v1f16.v1p0(<1 x ptr> align 1 undef, <1 x i1> undef, <1 x half> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 38 for instruction: %V8I64.u = call <8 x i64> @llvm.masked.gather.v8i64.v8p0(<8 x ptr> align 4 undef, <8 x i1> undef, <8 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %V4I64.u = call <4 x i64> @llvm.masked.gather.v4i64.v4p0(<4 x ptr> align 4 undef, <4 x i1> undef, <4 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V2I64.u = call <2 x i64> @llvm.masked.gather.v2i64.v2p0(<2 x ptr> align 4 undef, <2 x i1> undef, <2 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1I64.u = call <1 x i64> @llvm.masked.gather.v1i64.v1p0(<1 x ptr> align 4 undef, <1 x i1> undef, <1 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 78 for instruction: %V16I32.u = call <16 x i32> @llvm.masked.gather.v16i32.v16p0(<16 x ptr> align 1 undef, <16 x i1> undef, <16 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 38 for instruction: %V8I32.u = call <8 x i32> @llvm.masked.gather.v8i32.v8p0(<8 x ptr> align 1 undef, <8 x i1> undef, <8 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V4I32.u = call <4 x i32> @llvm.masked.gather.v4i32.v4p0(<4 x ptr> align 1 undef, <4 x i1> undef, <4 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V2I32.u = call <2 x i32> @llvm.masked.gather.v2i32.v2p0(<2 x ptr> align 1 undef, <2 x i1> undef, <2 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1I32.u = call <1 x i32> @llvm.masked.gather.v1i32.v1p0(<1 x ptr> align 1 undef, <1 x i1> undef, <1 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 157 for instruction: %V32I16.u = call <32 x i16> @llvm.masked.gather.v32i16.v32p0(<32 x ptr> align 1 undef, <32 x i1> undef, <32 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 78 for instruction: %V16I16.u = call <16 x i16> @llvm.masked.gather.v16i16.v16p0(<16 x ptr> align 1 undef, <16 x i1> undef, <16 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V8I16.u = call <8 x i16> @llvm.masked.gather.v8i16.v8p0(<8 x ptr> align 1 undef, <8 x i1> undef, <8 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V4I16.u = call <4 x i16> @llvm.masked.gather.v4i16.v4p0(<4 x ptr> align 1 undef, <4 x i1> undef, <4 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V2I16.u = call <2 x i16> @llvm.masked.gather.v2i16.v2p0(<2 x ptr> align 1 undef, <2 x i1> undef, <2 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1I16.u = call <1 x i16> @llvm.masked.gather.v1i16.v1p0(<1 x ptr> align 1 undef, <1 x i1> undef, <1 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 46 for instruction: %V8F64.u = call <8 x double> @llvm.masked.gather.v8f64.v8p0(<8 x ptr> align 2 undef, <8 x i1> undef, <8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %V4F64.u = call <4 x double> @llvm.masked.gather.v4f64.v4p0(<4 x ptr> align 2 undef, <4 x i1> undef, <4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V2F64.u = call <2 x double> @llvm.masked.gather.v2f64.v2p0(<2 x ptr> align 2 undef, <2 x i1> undef, <2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V1F64.u = call <1 x double> @llvm.masked.gather.v1f64.v1p0(<1 x ptr> align 2 undef, <1 x i1> undef, <1 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 94 for instruction: %V16F32.u = call <16 x float> @llvm.masked.gather.v16f32.v16p0(<16 x ptr> align 2 undef, <16 x i1> undef, <16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 46 for instruction: %V8F32.u = call <8 x float> @llvm.masked.gather.v8f32.v8p0(<8 x ptr> align 2 undef, <8 x i1> undef, <8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V4F32.u = call <4 x float> @llvm.masked.gather.v4f32.v4p0(<4 x ptr> align 2 undef, <4 x i1> undef, <4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V2F32.u = call <2 x float> @llvm.masked.gather.v2f32.v2p0(<2 x ptr> align 2 undef, <2 x i1> undef, <2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V1F32.u = call <1 x float> @llvm.masked.gather.v1f32.v1p0(<1 x ptr> align 2 undef, <1 x i1> undef, <1 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 189 for instruction: %V32BF16.u = call <32 x bfloat> @llvm.masked.gather.v32bf16.v32p0(<32 x ptr> align 1 undef, <32 x i1> undef, <32 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 94 for instruction: %V16BF16.u = call <16 x bfloat> @llvm.masked.gather.v16bf16.v16p0(<16 x ptr> align 1 undef, <16 x i1> undef, <16 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %V8BF16.u = call <8 x bfloat> @llvm.masked.gather.v8bf16.v8p0(<8 x ptr> align 1 undef, <8 x i1> undef, <8 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V4BF16.u = call <4 x bfloat> @llvm.masked.gather.v4bf16.v4p0(<4 x ptr> align 1 undef, <4 x i1> undef, <4 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V2BF16.u = call <2 x bfloat> @llvm.masked.gather.v2bf16.v2p0(<2 x ptr> align 1 undef, <2 x i1> undef, <2 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V1BF16.u = call <1 x bfloat> @llvm.masked.gather.v1bf16.v1p0(<1 x ptr> align 1 undef, <1 x i1> undef, <1 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 189 for instruction: %V32F16.u = call <32 x half> @llvm.masked.gather.v32f16.v32p0(<32 x ptr> align 1 undef, <32 x i1> undef, <32 x half> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 94 for instruction: %V16F16.u = call <16 x half> @llvm.masked.gather.v16f16.v16p0(<16 x ptr> align 1 undef, <16 x i1> undef, <16 x half> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %V8F16.u = call <8 x half> @llvm.masked.gather.v8f16.v8p0(<8 x ptr> align 1 undef, <8 x i1> undef, <8 x half> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V4F16.u = call <4 x half> @llvm.masked.gather.v4f16.v4p0(<4 x ptr> align 1 undef, <4 x i1> undef, <4 x half> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V2F16.u = call <2 x half> @llvm.masked.gather.v2f16.v2p0(<2 x ptr> align 1 undef, <2 x i1> undef, <2 x half> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V1F16.u = call <1 x half> @llvm.masked.gather.v1f16.v1p0(<1 x ptr> align 1 undef, <1 x i1> undef, <1 x half> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 46 for instruction: %V8I64.u = call <8 x i64> @llvm.masked.gather.v8i64.v8p0(<8 x ptr> align 4 undef, <8 x i1> undef, <8 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %V4I64.u = call <4 x i64> @llvm.masked.gather.v4i64.v4p0(<4 x ptr> align 4 undef, <4 x i1> undef, <4 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V2I64.u = call <2 x i64> @llvm.masked.gather.v2i64.v2p0(<2 x ptr> align 4 undef, <2 x i1> undef, <2 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V1I64.u = call <1 x i64> @llvm.masked.gather.v1i64.v1p0(<1 x ptr> align 4 undef, <1 x i1> undef, <1 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 94 for instruction: %V16I32.u = call <16 x i32> @llvm.masked.gather.v16i32.v16p0(<16 x ptr> align 1 undef, <16 x i1> undef, <16 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 46 for instruction: %V8I32.u = call <8 x i32> @llvm.masked.gather.v8i32.v8p0(<8 x ptr> align 1 undef, <8 x i1> undef, <8 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V4I32.u = call <4 x i32> @llvm.masked.gather.v4i32.v4p0(<4 x ptr> align 1 undef, <4 x i1> undef, <4 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V2I32.u = call <2 x i32> @llvm.masked.gather.v2i32.v2p0(<2 x ptr> align 1 undef, <2 x i1> undef, <2 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V1I32.u = call <1 x i32> @llvm.masked.gather.v1i32.v1p0(<1 x ptr> align 1 undef, <1 x i1> undef, <1 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 189 for instruction: %V32I16.u = call <32 x i16> @llvm.masked.gather.v32i16.v32p0(<32 x ptr> align 1 undef, <32 x i1> undef, <32 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 94 for instruction: %V16I16.u = call <16 x i16> @llvm.masked.gather.v16i16.v16p0(<16 x ptr> align 1 undef, <16 x i1> undef, <16 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %V8I16.u = call <8 x i16> @llvm.masked.gather.v8i16.v8p0(<8 x ptr> align 1 undef, <8 x i1> undef, <8 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V4I16.u = call <4 x i16> @llvm.masked.gather.v4i16.v4p0(<4 x ptr> align 1 undef, <4 x i1> undef, <4 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V2I16.u = call <2 x i16> @llvm.masked.gather.v2i16.v2p0(<2 x ptr> align 1 undef, <2 x i1> undef, <2 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V1I16.u = call <1 x i16> @llvm.masked.gather.v1i16.v1p0(<1 x ptr> align 1 undef, <1 x i1> undef, <1 x i16> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 0
;
%V8F64 = call <8 x double> @llvm.masked.gather.v8f64.v8p0(<8 x ptr> undef, i32 8, <8 x i1> undef, <8 x double> undef)
diff --git a/llvm/test/Analysis/CostModel/RISCV/fixed-vector-scatter.ll b/llvm/test/Analysis/CostModel/RISCV/fixed-vector-scatter.ll
index d175b90ec900b..abdd64ad6125d 100644
--- a/llvm/test/Analysis/CostModel/RISCV/fixed-vector-scatter.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/fixed-vector-scatter.ll
@@ -49,42 +49,42 @@ define i32 @masked_scatter() {
; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: call void @llvm.masked.scatter.v4i8.v4p0(<4 x i8> undef, <4 x ptr> align 1 undef, <4 x i1> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: call void @llvm.masked.scatter.v2i8.v2p0(<2 x i8> undef, <2 x ptr> align 1 undef, <2 x i1> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: call void @llvm.masked.scatter.v1i8.v1p0(<1 x i8> undef, <1 x ptr> align 1 undef, <1 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 38 for instruction: call void @llvm.masked.scatter.v8f64.v8p0(<8 x double> undef, <8 x ptr> align 2 undef, <8 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: call void @llvm.masked.scatter.v4f64.v4p0(<4 x double> undef, <4 x ptr> align 2 undef, <4 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: call void @llvm.masked.scatter.v2f64.v2p0(<2 x double> undef, <2 x ptr> align 2 undef, <2 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: call void @llvm.masked.scatter.v1f64.v1p0(<1 x double> undef, <1 x ptr> align 2 undef, <1 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 78 for instruction: call void @llvm.masked.scatter.v16f32.v16p0(<16 x float> undef, <16 x ptr> align 2 undef, <16 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 38 for instruction: call void @llvm.masked.scatter.v8f32.v8p0(<8 x float> undef, <8 x ptr> align 2 undef, <8 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: call void @llvm.masked.scatter.v4f32.v4p0(<4 x float> undef, <4 x ptr> align 2 undef, <4 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: call void @llvm.masked.scatter.v2f32.v2p0(<2 x float> undef, <2 x ptr> align 2 undef, <2 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: call void @llvm.masked.scatter.v1f32.v1p0(<1 x float> undef, <1 x ptr> align 2 undef, <1 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 157 for instruction: call void @llvm.masked.scatter.v32bf16.v32p0(<32 x bfloat> undef, <32 x ptr> align 1 undef, <32 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 78 for instruction: call void @llvm.masked.scatter.v16bf16.v16p0(<16 x bfloat> undef, <16 x ptr> align 1 undef, <16 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 38 for instruction: call void @llvm.masked.scatter.v8bf16.v8p0(<8 x bfloat> undef, <8 x ptr> align 1 undef, <8 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: call void @llvm.masked.scatter.v4bf16.v4p0(<4 x bfloat> undef, <4 x ptr> align 1 undef, <4 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: call void @llvm.masked.scatter.v2bf16.v2p0(<2 x bfloat> undef, <2 x ptr> align 1 undef, <2 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: call void @llvm.masked.scatter.v1bf16.v1p0(<1 x bfloat> undef, <1 x ptr> align 1 undef, <1 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 157 for instruction: call void @llvm.masked.scatter.v32f16.v32p0(<32 x half> undef, <32 x ptr> align 1 undef, <32 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 78 for instruction: call void @llvm.masked.scatter.v16f16.v16p0(<16 x half> undef, <16 x ptr> align 1 undef, <16 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 38 for instruction: call void @llvm.masked.scatter.v8f16.v8p0(<8 x half> undef, <8 x ptr> align 1 undef, <8 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: call void @llvm.masked.scatter.v4f16.v4p0(<4 x half> undef, <4 x ptr> align 1 undef, <4 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: call void @llvm.masked.scatter.v2f16.v2p0(<2 x half> undef, <2 x ptr> align 1 undef, <2 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: call void @llvm.masked.scatter.v1f16.v1p0(<1 x half> undef, <1 x ptr> align 1 undef, <1 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 38 for instruction: call void @llvm.masked.scatter.v8i64.v8p0(<8 x i64> undef, <8 x ptr> align 1 undef, <8 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: call void @llvm.masked.scatter.v4i64.v4p0(<4 x i64> undef, <4 x ptr> align 1 undef, <4 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: call void @llvm.masked.scatter.v2i64.v2p0(<2 x i64> undef, <2 x ptr> align 1 undef, <2 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: call void @llvm.masked.scatter.v1i64.v1p0(<1 x i64> undef, <1 x ptr> align 1 undef, <1 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 78 for instruction: call void @llvm.masked.scatter.v16i32.v16p0(<16 x i32> undef, <16 x ptr> align 1 undef, <16 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 38 for instruction: call void @llvm.masked.scatter.v8i32.v8p0(<8 x i32> undef, <8 x ptr> align 1 undef, <8 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: call void @llvm.masked.scatter.v4i32.v4p0(<4 x i32> undef, <4 x ptr> align 1 undef, <4 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: call void @llvm.masked.scatter.v2i32.v2p0(<2 x i32> undef, <2 x ptr> align 1 undef, <2 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: call void @llvm.masked.scatter.v1i32.v1p0(<1 x i32> undef, <1 x ptr> align 1 undef, <1 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 157 for instruction: call void @llvm.masked.scatter.v32i16.v32p0(<32 x i16> undef, <32 x ptr> align 1 undef, <32 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 78 for instruction: call void @llvm.masked.scatter.v16i16.v16p0(<16 x i16> undef, <16 x ptr> align 1 undef, <16 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 38 for instruction: call void @llvm.masked.scatter.v8i16.v8p0(<8 x i16> undef, <8 x ptr> align 1 undef, <8 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: call void @llvm.masked.scatter.v4i16.v4p0(<4 x i16> undef, <4 x ptr> align 1 undef, <4 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: call void @llvm.masked.scatter.v2i16.v2p0(<2 x i16> undef, <2 x ptr> align 1 undef, <2 x i1> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: call void @llvm.masked.scatter.v1i16.v1p0(<1 x i16> undef, <1 x ptr> align 1 undef, <1 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 54 for instruction: call void @llvm.masked.scatter.v8f64.v8p0(<8 x double> undef, <8 x ptr> align 2 undef, <8 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 26 for instruction: call void @llvm.masked.scatter.v4f64.v4p0(<4 x double> undef, <4 x ptr> align 2 undef, <4 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: call void @llvm.masked.scatter.v2f64.v2p0(<2 x double> undef, <2 x ptr> align 2 undef, <2 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: call void @llvm.masked.scatter.v1f64.v1p0(<1 x double> undef, <1 x ptr> align 2 undef, <1 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 110 for instruction: call void @llvm.masked.scatter.v16f32.v16p0(<16 x float> undef, <16 x ptr> align 2 undef, <16 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 54 for instruction: call void @llvm.masked.scatter.v8f32.v8p0(<8 x float> undef, <8 x ptr> align 2 undef, <8 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 26 for instruction: call void @llvm.masked.scatter.v4f32.v4p0(<4 x float> undef, <4 x ptr> align 2 undef, <4 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: call void @llvm.masked.scatter.v2f32.v2p0(<2 x float> undef, <2 x ptr> align 2 undef, <2 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: call void @llvm.masked.scatter.v1f32.v1p0(<1 x float> undef, <1 x ptr> align 2 undef, <1 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 221 for instruction: call void @llvm.masked.scatter.v32bf16.v32p0(<32 x bfloat> undef, <32 x ptr> align 1 undef, <32 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 110 for instruction: call void @llvm.masked.scatter.v16bf16.v16p0(<16 x bfloat> undef, <16 x ptr> align 1 undef, <16 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 54 for instruction: call void @llvm.masked.scatter.v8bf16.v8p0(<8 x bfloat> undef, <8 x ptr> align 1 undef, <8 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 26 for instruction: call void @llvm.masked.scatter.v4bf16.v4p0(<4 x bfloat> undef, <4 x ptr> align 1 undef, <4 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: call void @llvm.masked.scatter.v2bf16.v2p0(<2 x bfloat> undef, <2 x ptr> align 1 undef, <2 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: call void @llvm.masked.scatter.v1bf16.v1p0(<1 x bfloat> undef, <1 x ptr> align 1 undef, <1 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 221 for instruction: call void @llvm.masked.scatter.v32f16.v32p0(<32 x half> undef, <32 x ptr> align 1 undef, <32 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 110 for instruction: call void @llvm.masked.scatter.v16f16.v16p0(<16 x half> undef, <16 x ptr> align 1 undef, <16 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 54 for instruction: call void @llvm.masked.scatter.v8f16.v8p0(<8 x half> undef, <8 x ptr> align 1 undef, <8 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 26 for instruction: call void @llvm.masked.scatter.v4f16.v4p0(<4 x half> undef, <4 x ptr> align 1 undef, <4 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: call void @llvm.masked.scatter.v2f16.v2p0(<2 x half> undef, <2 x ptr> align 1 undef, <2 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: call void @llvm.masked.scatter.v1f16.v1p0(<1 x half> undef, <1 x ptr> align 1 undef, <1 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 54 for instruction: call void @llvm.masked.scatter.v8i64.v8p0(<8 x i64> undef, <8 x ptr> align 1 undef, <8 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 26 for instruction: call void @llvm.masked.scatter.v4i64.v4p0(<4 x i64> undef, <4 x ptr> align 1 undef, <4 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: call void @llvm.masked.scatter.v2i64.v2p0(<2 x i64> undef, <2 x ptr> align 1 undef, <2 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: call void @llvm.masked.scatter.v1i64.v1p0(<1 x i64> undef, <1 x ptr> align 1 undef, <1 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 110 for instruction: call void @llvm.masked.scatter.v16i32.v16p0(<16 x i32> undef, <16 x ptr> align 1 undef, <16 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 54 for instruction: call void @llvm.masked.scatter.v8i32.v8p0(<8 x i32> undef, <8 x ptr> align 1 undef, <8 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 26 for instruction: call void @llvm.masked.scatter.v4i32.v4p0(<4 x i32> undef, <4 x ptr> align 1 undef, <4 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: call void @llvm.masked.scatter.v2i32.v2p0(<2 x i32> undef, <2 x ptr> align 1 undef, <2 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: call void @llvm.masked.scatter.v1i32.v1p0(<1 x i32> undef, <1 x ptr> align 1 undef, <1 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 221 for instruction: call void @llvm.masked.scatter.v32i16.v32p0(<32 x i16> undef, <32 x ptr> align 1 undef, <32 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 110 for instruction: call void @llvm.masked.scatter.v16i16.v16p0(<16 x i16> undef, <16 x ptr> align 1 undef, <16 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 54 for instruction: call void @llvm.masked.scatter.v8i16.v8p0(<8 x i16> undef, <8 x ptr> align 1 undef, <8 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 26 for instruction: call void @llvm.masked.scatter.v4i16.v4p0(<4 x i16> undef, <4 x ptr> align 1 undef, <4 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: call void @llvm.masked.scatter.v2i16.v2p0(<2 x i16> undef, <2 x ptr> align 1 undef, <2 x i1> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: call void @llvm.masked.scatter.v1i16.v1p0(<1 x i16> undef, <1 x ptr> align 1 undef, <1 x i1> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 0
;
call void @llvm.masked.scatter.v8f64.v8p0(<8 x double> undef, <8 x ptr> undef, i32 8, <8 x i1> undef)
diff --git a/llvm/test/Analysis/CostModel/RISCV/masked_ldst.ll b/llvm/test/Analysis/CostModel/RISCV/masked_ldst.ll
index e572be71004b2..a8dd9bfe05aa0 100644
--- a/llvm/test/Analysis/CostModel/RISCV/masked_ldst.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/masked_ldst.ll
@@ -13,14 +13,14 @@ define void @fixed() {
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v2i32 = call <2 x i32> @llvm.masked.load.v2i32.p0(ptr align 8 undef, <2 x i1> undef, <2 x i32> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v4i32 = call <4 x i32> @llvm.masked.load.v4i32.p0(ptr align 8 undef, <4 x i1> undef, <4 x i32> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v2i64 = call <2 x i64> @llvm.masked.load.v2i64.p0(ptr align 8 undef, <2 x i1> undef, <2 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v2f16 = call <2 x half> @llvm.masked.load.v2f16.p0(ptr align 8 undef, <2 x i1> undef, <2 x half> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %v4f16 = call <4 x half> @llvm.masked.load.v4f16.p0(ptr align 8 undef, <4 x i1> undef, <4 x half> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %v8f16 = call <8 x half> @llvm.masked.load.v8f16.p0(ptr align 8 undef, <8 x i1> undef, <8 x half> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %v2f16 = call <2 x half> @llvm.masked.load.v2f16.p0(ptr align 8 undef, <2 x i1> undef, <2 x half> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 23 for instruction: %v4f16 = call <4 x half> @llvm.masked.load.v4f16.p0(ptr align 8 undef, <4 x i1> undef, <4 x half> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 47 for instruction: %v8f16 = call <8 x half> @llvm.masked.load.v8f16.p0(ptr align 8 undef, <8 x i1> undef, <8 x half> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v2f32 = call <2 x float> @llvm.masked.load.v2f32.p0(ptr align 8 undef, <2 x i1> undef, <2 x float> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v4f32 = call <4 x float> @llvm.masked.load.v4f32.p0(ptr align 8 undef, <4 x i1> undef, <4 x float> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v2f64 = call <2 x double> @llvm.masked.load.v2f64.p0(ptr align 8 undef, <2 x i1> undef, <2 x double> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i64 = call <4 x i64> @llvm.masked.load.v4i64.p0(ptr align 8 undef, <4 x i1> undef, <4 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 223 for instruction: %v32f16 = call <32 x half> @llvm.masked.load.v32f16.p0(ptr align 8 undef, <32 x i1> undef, <32 x half> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 255 for instruction: %v32f16 = call <32 x half> @llvm.masked.load.v32f16.p0(ptr align 8 undef, <32 x i1> undef, <32 x half> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
entry:
diff --git a/llvm/test/Analysis/CostModel/RISCV/reduce-add.ll b/llvm/test/Analysis/CostModel/RISCV/reduce-add.ll
index 23ee689937de1..84fe48de3686d 100644
--- a/llvm/test/Analysis/CostModel/RISCV/reduce-add.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/reduce-add.ll
@@ -40,14 +40,14 @@ define i32 @reduce_i1(i32 %arg) {
define i32 @reduce_i8(i32 %arg) {
; CHECK-LABEL: 'reduce_i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i8 @llvm.vector.reduce.add.v1i8(<1 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.vector.reduce.add.v2i8(<2 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i8 @llvm.vector.reduce.add.v4i8(<4 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i8 @llvm.vector.reduce.add.v8i8(<8 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i8 @llvm.vector.reduce.add.v16i8(<16 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i8 @llvm.vector.reduce.add.v32i8(<32 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i8 @llvm.vector.reduce.add.v64i8(<64 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V128 = call i8 @llvm.vector.reduce.add.v128i8(<128 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1 = call i8 @llvm.vector.reduce.add.v1i8(<1 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i8 @llvm.vector.reduce.add.v2i8(<2 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.vector.reduce.add.v4i8(<4 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i8 @llvm.vector.reduce.add.v8i8(<8 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V16 = call i8 @llvm.vector.reduce.add.v16i8(<16 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i8 @llvm.vector.reduce.add.v32i8(<32 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V64 = call i8 @llvm.vector.reduce.add.v64i8(<64 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %V128 = call i8 @llvm.vector.reduce.add.v128i8(<128 x i8> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_i8'
@@ -74,14 +74,14 @@ define i32 @reduce_i8(i32 %arg) {
define i32 @reduce_i16(i32 %arg) {
; CHECK-LABEL: 'reduce_i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i16 @llvm.vector.reduce.add.v1i16(<1 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.vector.reduce.add.v2i16(<2 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.vector.reduce.add.v4i16(<4 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i16 @llvm.vector.reduce.add.v8i16(<8 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i16 @llvm.vector.reduce.add.v16i16(<16 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i16 @llvm.vector.reduce.add.v32i16(<32 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.vector.reduce.add.v64i16(<64 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i16 @llvm.vector.reduce.add.v128i16(<128 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1 = call i16 @llvm.vector.reduce.add.v1i16(<1 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i16 @llvm.vector.reduce.add.v2i16(<2 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.vector.reduce.add.v4i16(<4 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i16 @llvm.vector.reduce.add.v8i16(<8 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V16 = call i16 @llvm.vector.reduce.add.v16i16(<16 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i16 @llvm.vector.reduce.add.v32i16(<32 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V64 = call i16 @llvm.vector.reduce.add.v64i16(<64 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %V128 = call i16 @llvm.vector.reduce.add.v128i16(<128 x i16> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_i16'
@@ -108,14 +108,14 @@ define i32 @reduce_i16(i32 %arg) {
define i32 @reduce_i32(i32 %arg) {
; CHECK-LABEL: 'reduce_i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i32 @llvm.vector.reduce.add.v1i32(<1 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.vector.reduce.add.v2i32(<2 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i32 @llvm.vector.reduce.add.v4i32(<4 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i32 @llvm.vector.reduce.add.v8i32(<8 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i32 @llvm.vector.reduce.add.v16i32(<16 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i32 @llvm.vector.reduce.add.v32i32(<32 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V64 = call i32 @llvm.vector.reduce.add.v64i32(<64 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V128 = call i32 @llvm.vector.reduce.add.v128i32(<128 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1 = call i32 @llvm.vector.reduce.add.v1i32(<1 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i32 @llvm.vector.reduce.add.v2i32(<2 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.vector.reduce.add.v4i32(<4 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.vector.reduce.add.v8i32(<8 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V16 = call i32 @llvm.vector.reduce.add.v16i32(<16 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.vector.reduce.add.v32i32(<32 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V64 = call i32 @llvm.vector.reduce.add.v64i32(<64 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 32 for instruction: %V128 = call i32 @llvm.vector.reduce.add.v128i32(<128 x i32> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_i32'
@@ -142,14 +142,14 @@ define i32 @reduce_i32(i32 %arg) {
define i32 @reduce_i64(i32 %arg) {
; CHECK-LABEL: 'reduce_i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i64 @llvm.vector.reduce.add.v1i64(<1 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.vector.reduce.add.v2i64(<2 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.vector.reduce.add.v4i64(<4 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i64 @llvm.vector.reduce.add.v8i64(<8 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i64 @llvm.vector.reduce.add.v16i64(<16 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V32 = call i64 @llvm.vector.reduce.add.v32i64(<32 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %V64 = call i64 @llvm.vector.reduce.add.v64i64(<64 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 62 for instruction: %V128 = call i64 @llvm.vector.reduce.add.v128i64(<128 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1 = call i64 @llvm.vector.reduce.add.v1i64(<1 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i64 @llvm.vector.reduce.add.v2i64(<2 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.vector.reduce.add.v4i64(<4 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.vector.reduce.add.v8i64(<8 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V16 = call i64 @llvm.vector.reduce.add.v16i64(<16 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V32 = call i64 @llvm.vector.reduce.add.v32i64(<32 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V64 = call i64 @llvm.vector.reduce.add.v64i64(<64 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 63 for instruction: %V128 = call i64 @llvm.vector.reduce.add.v128i64(<128 x i64> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_i64'
diff --git a/llvm/test/Analysis/CostModel/RISCV/reduce-and.ll b/llvm/test/Analysis/CostModel/RISCV/reduce-and.ll
index eabbe1d43f117..7b9d8e146a001 100644
--- a/llvm/test/Analysis/CostModel/RISCV/reduce-and.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/reduce-and.ll
@@ -83,14 +83,14 @@ define i32 @reduce_i1(i32 %arg) {
define i32 @reduce_i8(i32 %arg) {
; CHECK-LABEL: 'reduce_i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i8 @llvm.vector.reduce.and.v1i8(<1 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.vector.reduce.and.v2i8(<2 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.vector.reduce.and.v4i8(<4 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.vector.reduce.and.v8i8(<8 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V16 = call i8 @llvm.vector.reduce.and.v16i8(<16 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i8 @llvm.vector.reduce.and.v32i8(<32 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V64 = call i8 @llvm.vector.reduce.and.v64i8(<64 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i8 @llvm.vector.reduce.and.v128i8(<128 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i8 @llvm.vector.reduce.and.v1i8(<1 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.vector.reduce.and.v2i8(<2 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i8 @llvm.vector.reduce.and.v4i8(<4 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i8 @llvm.vector.reduce.and.v8i8(<8 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i8 @llvm.vector.reduce.and.v16i8(<16 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i8 @llvm.vector.reduce.and.v32i8(<32 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i8 @llvm.vector.reduce.and.v64i8(<64 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V128 = call i8 @llvm.vector.reduce.and.v128i8(<128 x i8> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_i8'
@@ -117,14 +117,14 @@ define i32 @reduce_i8(i32 %arg) {
define i32 @reduce_i16(i32 %arg) {
; CHECK-LABEL: 'reduce_i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i16 @llvm.vector.reduce.and.v1i16(<1 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i16 @llvm.vector.reduce.and.v2i16(<2 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i16 @llvm.vector.reduce.and.v4i16(<4 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.vector.reduce.and.v8i16(<8 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V16 = call i16 @llvm.vector.reduce.and.v16i16(<16 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i16 @llvm.vector.reduce.and.v32i16(<32 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V64 = call i16 @llvm.vector.reduce.and.v64i16(<64 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V128 = call i16 @llvm.vector.reduce.and.v128i16(<128 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i16 @llvm.vector.reduce.and.v1i16(<1 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.vector.reduce.and.v2i16(<2 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.vector.reduce.and.v4i16(<4 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i16 @llvm.vector.reduce.and.v8i16(<8 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i16 @llvm.vector.reduce.and.v16i16(<16 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i16 @llvm.vector.reduce.and.v32i16(<32 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.vector.reduce.and.v64i16(<64 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i16 @llvm.vector.reduce.and.v128i16(<128 x i16> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_i16'
@@ -151,14 +151,14 @@ define i32 @reduce_i16(i32 %arg) {
define i32 @reduce_i32(i32 %arg) {
; CHECK-LABEL: 'reduce_i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i32 @llvm.vector.reduce.and.v1i32(<1 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i32 @llvm.vector.reduce.and.v2i32(<2 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i32 @llvm.vector.reduce.and.v4i32(<4 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i32 @llvm.vector.reduce.and.v8i32(<8 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V16 = call i32 @llvm.vector.reduce.and.v16i32(<16 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i32 @llvm.vector.reduce.and.v32i32(<32 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i32 @llvm.vector.reduce.and.v64i32(<64 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %V128 = call i32 @llvm.vector.reduce.and.v128i32(<128 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i32 @llvm.vector.reduce.and.v1i32(<1 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.vector.reduce.and.v2i32(<2 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i32 @llvm.vector.reduce.and.v4i32(<4 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i32 @llvm.vector.reduce.and.v8i32(<8 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i32 @llvm.vector.reduce.and.v16i32(<16 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i32 @llvm.vector.reduce.and.v32i32(<32 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V64 = call i32 @llvm.vector.reduce.and.v64i32(<64 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V128 = call i32 @llvm.vector.reduce.and.v128i32(<128 x i32> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_i32'
@@ -185,14 +185,14 @@ define i32 @reduce_i32(i32 %arg) {
define i32 @reduce_i64(i32 %arg) {
; CHECK-LABEL: 'reduce_i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i64 @llvm.vector.reduce.and.v1i64(<1 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i64 @llvm.vector.reduce.and.v2i64(<2 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i64 @llvm.vector.reduce.and.v4i64(<4 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i64 @llvm.vector.reduce.and.v8i64(<8 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V16 = call i64 @llvm.vector.reduce.and.v16i64(<16 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %V32 = call i64 @llvm.vector.reduce.and.v32i64(<32 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 29 for instruction: %V64 = call i64 @llvm.vector.reduce.and.v64i64(<64 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 61 for instruction: %V128 = call i64 @llvm.vector.reduce.and.v128i64(<128 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i64 @llvm.vector.reduce.and.v1i64(<1 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.vector.reduce.and.v2i64(<2 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.vector.reduce.and.v4i64(<4 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i64 @llvm.vector.reduce.and.v8i64(<8 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i64 @llvm.vector.reduce.and.v16i64(<16 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V32 = call i64 @llvm.vector.reduce.and.v32i64(<32 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %V64 = call i64 @llvm.vector.reduce.and.v64i64(<64 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 62 for instruction: %V128 = call i64 @llvm.vector.reduce.and.v128i64(<128 x i64> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_i64'
diff --git a/llvm/test/Analysis/CostModel/RISCV/reduce-fadd.ll b/llvm/test/Analysis/CostModel/RISCV/reduce-fadd.ll
index 9ee33d4b5a31e..95b0a0fa9ef5e 100644
--- a/llvm/test/Analysis/CostModel/RISCV/reduce-fadd.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/reduce-fadd.ll
@@ -6,14 +6,14 @@
define void @reduce_fadd_bfloat() {
; FP-REDUCE-ZVFH-LABEL: 'reduce_fadd_bfloat'
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call fast bfloat @llvm.vector.reduce.fadd.v1bf16(bfloat 0.000000e+00, <1 x bfloat> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V2 = call fast bfloat @llvm.vector.reduce.fadd.v2bf16(bfloat 0.000000e+00, <2 x bfloat> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V4 = call fast bfloat @llvm.vector.reduce.fadd.v4bf16(bfloat 0.000000e+00, <4 x bfloat> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %V8 = call fast bfloat @llvm.vector.reduce.fadd.v8bf16(bfloat 0.000000e+00, <8 x bfloat> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 73 for instruction: %V16 = call fast bfloat @llvm.vector.reduce.fadd.v16bf16(bfloat 0.000000e+00, <16 x bfloat> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 211 for instruction: %v32 = call fast bfloat @llvm.vector.reduce.fadd.v32bf16(bfloat 0.000000e+00, <32 x bfloat> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 541 for instruction: %V64 = call fast bfloat @llvm.vector.reduce.fadd.v64bf16(bfloat 0.000000e+00, <64 x bfloat> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 573 for instruction: %V128 = call fast bfloat @llvm.vector.reduce.fadd.v128bf16(bfloat 0.000000e+00, <128 x bfloat> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call fast bfloat @llvm.vector.reduce.fadd.v1bf16(bfloat 0.000000e+00, <1 x bfloat> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V2 = call fast bfloat @llvm.vector.reduce.fadd.v2bf16(bfloat 0.000000e+00, <2 x bfloat> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V4 = call fast bfloat @llvm.vector.reduce.fadd.v4bf16(bfloat 0.000000e+00, <4 x bfloat> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 29 for instruction: %V8 = call fast bfloat @llvm.vector.reduce.fadd.v8bf16(bfloat 0.000000e+00, <8 x bfloat> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 74 for instruction: %V16 = call fast bfloat @llvm.vector.reduce.fadd.v16bf16(bfloat 0.000000e+00, <16 x bfloat> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 212 for instruction: %v32 = call fast bfloat @llvm.vector.reduce.fadd.v32bf16(bfloat 0.000000e+00, <32 x bfloat> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 542 for instruction: %V64 = call fast bfloat @llvm.vector.reduce.fadd.v64bf16(bfloat 0.000000e+00, <64 x bfloat> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 574 for instruction: %V128 = call fast bfloat @llvm.vector.reduce.fadd.v128bf16(bfloat 0.000000e+00, <128 x bfloat> undef)
; FP-REDUCE-ZVFH-NEXT: Cost Model: Invalid cost for instruction: %NXV1 = call fast bfloat @llvm.vector.reduce.fadd.nxv1bf16(bfloat 0.000000e+00, <vscale x 1 x bfloat> undef)
; FP-REDUCE-ZVFH-NEXT: Cost Model: Invalid cost for instruction: %NXV2 = call fast bfloat @llvm.vector.reduce.fadd.nxv2bf16(bfloat 0.000000e+00, <vscale x 2 x bfloat> undef)
; FP-REDUCE-ZVFH-NEXT: Cost Model: Invalid cost for instruction: %NXV4 = call fast bfloat @llvm.vector.reduce.fadd.nxv4bf16(bfloat 0.000000e+00, <vscale x 4 x bfloat> undef)
@@ -23,14 +23,14 @@ define void @reduce_fadd_bfloat() {
; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; FP-REDUCE-ZVFHMIN-LABEL: 'reduce_fadd_bfloat'
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call fast bfloat @llvm.vector.reduce.fadd.v1bf16(bfloat 0.000000e+00, <1 x bfloat> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V2 = call fast bfloat @llvm.vector.reduce.fadd.v2bf16(bfloat 0.000000e+00, <2 x bfloat> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V4 = call fast bfloat @llvm.vector.reduce.fadd.v4bf16(bfloat 0.000000e+00, <4 x bfloat> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %V8 = call fast bfloat @llvm.vector.reduce.fadd.v8bf16(bfloat 0.000000e+00, <8 x bfloat> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 73 for instruction: %V16 = call fast bfloat @llvm.vector.reduce.fadd.v16bf16(bfloat 0.000000e+00, <16 x bfloat> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 211 for instruction: %v32 = call fast bfloat @llvm.vector.reduce.fadd.v32bf16(bfloat 0.000000e+00, <32 x bfloat> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 541 for instruction: %V64 = call fast bfloat @llvm.vector.reduce.fadd.v64bf16(bfloat 0.000000e+00, <64 x bfloat> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 573 for instruction: %V128 = call fast bfloat @llvm.vector.reduce.fadd.v128bf16(bfloat 0.000000e+00, <128 x bfloat> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call fast bfloat @llvm.vector.reduce.fadd.v1bf16(bfloat 0.000000e+00, <1 x bfloat> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V2 = call fast bfloat @llvm.vector.reduce.fadd.v2bf16(bfloat 0.000000e+00, <2 x bfloat> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V4 = call fast bfloat @llvm.vector.reduce.fadd.v4bf16(bfloat 0.000000e+00, <4 x bfloat> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 29 for instruction: %V8 = call fast bfloat @llvm.vector.reduce.fadd.v8bf16(bfloat 0.000000e+00, <8 x bfloat> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 74 for instruction: %V16 = call fast bfloat @llvm.vector.reduce.fadd.v16bf16(bfloat 0.000000e+00, <16 x bfloat> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 212 for instruction: %v32 = call fast bfloat @llvm.vector.reduce.fadd.v32bf16(bfloat 0.000000e+00, <32 x bfloat> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 542 for instruction: %V64 = call fast bfloat @llvm.vector.reduce.fadd.v64bf16(bfloat 0.000000e+00, <64 x bfloat> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 574 for instruction: %V128 = call fast bfloat @llvm.vector.reduce.fadd.v128bf16(bfloat 0.000000e+00, <128 x bfloat> undef)
; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV1 = call fast bfloat @llvm.vector.reduce.fadd.nxv1bf16(bfloat 0.000000e+00, <vscale x 1 x bfloat> undef)
; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV2 = call fast bfloat @llvm.vector.reduce.fadd.nxv2bf16(bfloat 0.000000e+00, <vscale x 2 x bfloat> undef)
; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV4 = call fast bfloat @llvm.vector.reduce.fadd.nxv4bf16(bfloat 0.000000e+00, <vscale x 4 x bfloat> undef)
@@ -92,31 +92,31 @@ define void @reduce_fadd_bfloat() {
define void @reduce_fadd_half() {
; FP-REDUCE-ZVFH-LABEL: 'reduce_fadd_half'
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call fast half @llvm.vector.reduce.fadd.v1f16(half 0.000000e+00, <1 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call fast half @llvm.vector.reduce.fadd.v2f16(half 0.000000e+00, <2 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call fast half @llvm.vector.reduce.fadd.v4f16(half 0.000000e+00, <4 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call fast half @llvm.vector.reduce.fadd.v8f16(half 0.000000e+00, <8 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call fast half @llvm.vector.reduce.fadd.v16f16(half 0.000000e+00, <16 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %v32 = call fast half @llvm.vector.reduce.fadd.v32f16(half 0.000000e+00, <32 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V64 = call fast half @llvm.vector.reduce.fadd.v64f16(half 0.000000e+00, <64 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V128 = call fast half @llvm.vector.reduce.fadd.v128f16(half 0.000000e+00, <128 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %NXV1 = call fast half @llvm.vector.reduce.fadd.nxv1f16(half 0.000000e+00, <vscale x 1 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %NXV2 = call fast half @llvm.vector.reduce.fadd.nxv2f16(half 0.000000e+00, <vscale x 2 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %NXV4 = call fast half @llvm.vector.reduce.fadd.nxv4f16(half 0.000000e+00, <vscale x 4 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %NXV8 = call fast half @llvm.vector.reduce.fadd.nxv8f16(half 0.000000e+00, <vscale x 8 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %NXV16 = call fast half @llvm.vector.reduce.fadd.nxv16f16(half 0.000000e+00, <vscale x 16 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %NXV32 = call fast half @llvm.vector.reduce.fadd.nxv32f16(half 0.000000e+00, <vscale x 32 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1 = call fast half @llvm.vector.reduce.fadd.v1f16(half 0.000000e+00, <1 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2 = call fast half @llvm.vector.reduce.fadd.v2f16(half 0.000000e+00, <2 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V4 = call fast half @llvm.vector.reduce.fadd.v4f16(half 0.000000e+00, <4 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V8 = call fast half @llvm.vector.reduce.fadd.v8f16(half 0.000000e+00, <8 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V16 = call fast half @llvm.vector.reduce.fadd.v16f16(half 0.000000e+00, <16 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %v32 = call fast half @llvm.vector.reduce.fadd.v32f16(half 0.000000e+00, <32 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V64 = call fast half @llvm.vector.reduce.fadd.v64f16(half 0.000000e+00, <64 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %V128 = call fast half @llvm.vector.reduce.fadd.v128f16(half 0.000000e+00, <128 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %NXV1 = call fast half @llvm.vector.reduce.fadd.nxv1f16(half 0.000000e+00, <vscale x 1 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %NXV2 = call fast half @llvm.vector.reduce.fadd.nxv2f16(half 0.000000e+00, <vscale x 2 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %NXV4 = call fast half @llvm.vector.reduce.fadd.nxv4f16(half 0.000000e+00, <vscale x 4 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %NXV8 = call fast half @llvm.vector.reduce.fadd.nxv8f16(half 0.000000e+00, <vscale x 8 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %NXV16 = call fast half @llvm.vector.reduce.fadd.nxv16f16(half 0.000000e+00, <vscale x 16 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %NXV32 = call fast half @llvm.vector.reduce.fadd.nxv32f16(half 0.000000e+00, <vscale x 32 x half> undef)
; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; FP-REDUCE-ZVFHMIN-LABEL: 'reduce_fadd_half'
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call fast half @llvm.vector.reduce.fadd.v1f16(half 0.000000e+00, <1 x half> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V2 = call fast half @llvm.vector.reduce.fadd.v2f16(half 0.000000e+00, <2 x half> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V4 = call fast half @llvm.vector.reduce.fadd.v4f16(half 0.000000e+00, <4 x half> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %V8 = call fast half @llvm.vector.reduce.fadd.v8f16(half 0.000000e+00, <8 x half> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 73 for instruction: %V16 = call fast half @llvm.vector.reduce.fadd.v16f16(half 0.000000e+00, <16 x half> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 211 for instruction: %v32 = call fast half @llvm.vector.reduce.fadd.v32f16(half 0.000000e+00, <32 x half> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 541 for instruction: %V64 = call fast half @llvm.vector.reduce.fadd.v64f16(half 0.000000e+00, <64 x half> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 573 for instruction: %V128 = call fast half @llvm.vector.reduce.fadd.v128f16(half 0.000000e+00, <128 x half> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call fast half @llvm.vector.reduce.fadd.v1f16(half 0.000000e+00, <1 x half> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V2 = call fast half @llvm.vector.reduce.fadd.v2f16(half 0.000000e+00, <2 x half> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V4 = call fast half @llvm.vector.reduce.fadd.v4f16(half 0.000000e+00, <4 x half> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 29 for instruction: %V8 = call fast half @llvm.vector.reduce.fadd.v8f16(half 0.000000e+00, <8 x half> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 74 for instruction: %V16 = call fast half @llvm.vector.reduce.fadd.v16f16(half 0.000000e+00, <16 x half> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 212 for instruction: %v32 = call fast half @llvm.vector.reduce.fadd.v32f16(half 0.000000e+00, <32 x half> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 542 for instruction: %V64 = call fast half @llvm.vector.reduce.fadd.v64f16(half 0.000000e+00, <64 x half> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 574 for instruction: %V128 = call fast half @llvm.vector.reduce.fadd.v128f16(half 0.000000e+00, <128 x half> undef)
; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV1 = call fast half @llvm.vector.reduce.fadd.nxv1f16(half 0.000000e+00, <vscale x 1 x half> undef)
; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV2 = call fast half @llvm.vector.reduce.fadd.nxv2f16(half 0.000000e+00, <vscale x 2 x half> undef)
; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV4 = call fast half @llvm.vector.reduce.fadd.nxv4f16(half 0.000000e+00, <vscale x 4 x half> undef)
@@ -178,19 +178,19 @@ define void @reduce_fadd_half() {
define void @reduce_fadd_float() {
; FP-REDUCE-LABEL: 'reduce_fadd_float'
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call fast float @llvm.vector.reduce.fadd.v1f32(float 0.000000e+00, <1 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call fast float @llvm.vector.reduce.fadd.v2f32(float 0.000000e+00, <2 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call fast float @llvm.vector.reduce.fadd.v4f32(float 0.000000e+00, <4 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call fast float @llvm.vector.reduce.fadd.v8f32(float 0.000000e+00, <8 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call fast float @llvm.vector.reduce.fadd.v16f32(float 0.000000e+00, <16 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %v32 = call fast float @llvm.vector.reduce.fadd.v32f32(float 0.000000e+00, <32 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V64 = call fast float @llvm.vector.reduce.fadd.v64f32(float 0.000000e+00, <64 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V128 = call fast float @llvm.vector.reduce.fadd.v128f32(float 0.000000e+00, <128 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %NXV1 = call fast float @llvm.vector.reduce.fadd.nxv1f32(float 0.000000e+00, <vscale x 1 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %NXV2 = call fast float @llvm.vector.reduce.fadd.nxv2f32(float 0.000000e+00, <vscale x 2 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %NXV4 = call fast float @llvm.vector.reduce.fadd.nxv4f32(float 0.000000e+00, <vscale x 4 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %NXV8 = call fast float @llvm.vector.reduce.fadd.nxv8f32(float 0.000000e+00, <vscale x 8 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %NXV16 = call fast float @llvm.vector.reduce.fadd.nxv16f32(float 0.000000e+00, <vscale x 16 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1 = call fast float @llvm.vector.reduce.fadd.v1f32(float 0.000000e+00, <1 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2 = call fast float @llvm.vector.reduce.fadd.v2f32(float 0.000000e+00, <2 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V4 = call fast float @llvm.vector.reduce.fadd.v4f32(float 0.000000e+00, <4 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V8 = call fast float @llvm.vector.reduce.fadd.v8f32(float 0.000000e+00, <8 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V16 = call fast float @llvm.vector.reduce.fadd.v16f32(float 0.000000e+00, <16 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %v32 = call fast float @llvm.vector.reduce.fadd.v32f32(float 0.000000e+00, <32 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V64 = call fast float @llvm.vector.reduce.fadd.v64f32(float 0.000000e+00, <64 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 32 for instruction: %V128 = call fast float @llvm.vector.reduce.fadd.v128f32(float 0.000000e+00, <128 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %NXV1 = call fast float @llvm.vector.reduce.fadd.nxv1f32(float 0.000000e+00, <vscale x 1 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %NXV2 = call fast float @llvm.vector.reduce.fadd.nxv2f32(float 0.000000e+00, <vscale x 2 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %NXV4 = call fast float @llvm.vector.reduce.fadd.nxv4f32(float 0.000000e+00, <vscale x 4 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %NXV8 = call fast float @llvm.vector.reduce.fadd.nxv8f32(float 0.000000e+00, <vscale x 8 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %NXV16 = call fast float @llvm.vector.reduce.fadd.nxv16f32(float 0.000000e+00, <vscale x 16 x float> undef)
; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; SIZE-LABEL: 'reduce_fadd_float'
@@ -227,18 +227,18 @@ define void @reduce_fadd_float() {
define void @reduce_fadd_double() {
; FP-REDUCE-LABEL: 'reduce_fadd_double'
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call fast double @llvm.vector.reduce.fadd.v1f64(double 0.000000e+00, <1 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call fast double @llvm.vector.reduce.fadd.v2f64(double 0.000000e+00, <2 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call fast double @llvm.vector.reduce.fadd.v4f64(double 0.000000e+00, <4 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call fast double @llvm.vector.reduce.fadd.v8f64(double 0.000000e+00, <8 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call fast double @llvm.vector.reduce.fadd.v16f64(double 0.000000e+00, <16 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %v32 = call fast double @llvm.vector.reduce.fadd.v32f64(double 0.000000e+00, <32 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %V64 = call fast double @llvm.vector.reduce.fadd.v64f64(double 0.000000e+00, <64 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 62 for instruction: %V128 = call fast double @llvm.vector.reduce.fadd.v128f64(double 0.000000e+00, <128 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %NXV1 = call fast double @llvm.vector.reduce.fadd.nxv1f64(double 0.000000e+00, <vscale x 1 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %NXV2 = call fast double @llvm.vector.reduce.fadd.nxv2f64(double 0.000000e+00, <vscale x 2 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %NXV4 = call fast double @llvm.vector.reduce.fadd.nxv4f64(double 0.000000e+00, <vscale x 4 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %NXV8 = call fast double @llvm.vector.reduce.fadd.nxv8f64(double 0.000000e+00, <vscale x 8 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1 = call fast double @llvm.vector.reduce.fadd.v1f64(double 0.000000e+00, <1 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2 = call fast double @llvm.vector.reduce.fadd.v2f64(double 0.000000e+00, <2 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V4 = call fast double @llvm.vector.reduce.fadd.v4f64(double 0.000000e+00, <4 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V8 = call fast double @llvm.vector.reduce.fadd.v8f64(double 0.000000e+00, <8 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V16 = call fast double @llvm.vector.reduce.fadd.v16f64(double 0.000000e+00, <16 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %v32 = call fast double @llvm.vector.reduce.fadd.v32f64(double 0.000000e+00, <32 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V64 = call fast double @llvm.vector.reduce.fadd.v64f64(double 0.000000e+00, <64 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 63 for instruction: %V128 = call fast double @llvm.vector.reduce.fadd.v128f64(double 0.000000e+00, <128 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %NXV1 = call fast double @llvm.vector.reduce.fadd.nxv1f64(double 0.000000e+00, <vscale x 1 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %NXV2 = call fast double @llvm.vector.reduce.fadd.nxv2f64(double 0.000000e+00, <vscale x 2 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %NXV4 = call fast double @llvm.vector.reduce.fadd.nxv4f64(double 0.000000e+00, <vscale x 4 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %NXV8 = call fast double @llvm.vector.reduce.fadd.nxv8f64(double 0.000000e+00, <vscale x 8 x double> undef)
; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; SIZE-LABEL: 'reduce_fadd_double'
@@ -273,14 +273,14 @@ define void @reduce_fadd_double() {
define void @reduce_ordered_fadd_bfloat() {
; FP-REDUCE-ZVFH-LABEL: 'reduce_ordered_fadd_bfloat'
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1 = call bfloat @llvm.vector.reduce.fadd.v1bf16(bfloat 0.000000e+00, <1 x bfloat> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V2 = call bfloat @llvm.vector.reduce.fadd.v2bf16(bfloat 0.000000e+00, <2 x bfloat> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V4 = call bfloat @llvm.vector.reduce.fadd.v4bf16(bfloat 0.000000e+00, <4 x bfloat> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V8 = call bfloat @llvm.vector.reduce.fadd.v8bf16(bfloat 0.000000e+00, <8 x bfloat> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 63 for instruction: %V16 = call bfloat @llvm.vector.reduce.fadd.v16bf16(bfloat 0.000000e+00, <16 x bfloat> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 127 for instruction: %v32 = call bfloat @llvm.vector.reduce.fadd.v32bf16(bfloat 0.000000e+00, <32 x bfloat> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 255 for instruction: %V64 = call bfloat @llvm.vector.reduce.fadd.v64bf16(bfloat 0.000000e+00, <64 x bfloat> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 510 for instruction: %V128 = call bfloat @llvm.vector.reduce.fadd.v128bf16(bfloat 0.000000e+00, <128 x bfloat> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V1 = call bfloat @llvm.vector.reduce.fadd.v1bf16(bfloat 0.000000e+00, <1 x bfloat> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V2 = call bfloat @llvm.vector.reduce.fadd.v2bf16(bfloat 0.000000e+00, <2 x bfloat> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V4 = call bfloat @llvm.vector.reduce.fadd.v4bf16(bfloat 0.000000e+00, <4 x bfloat> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %V8 = call bfloat @llvm.vector.reduce.fadd.v8bf16(bfloat 0.000000e+00, <8 x bfloat> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 79 for instruction: %V16 = call bfloat @llvm.vector.reduce.fadd.v16bf16(bfloat 0.000000e+00, <16 x bfloat> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 159 for instruction: %v32 = call bfloat @llvm.vector.reduce.fadd.v32bf16(bfloat 0.000000e+00, <32 x bfloat> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 319 for instruction: %V64 = call bfloat @llvm.vector.reduce.fadd.v64bf16(bfloat 0.000000e+00, <64 x bfloat> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 638 for instruction: %V128 = call bfloat @llvm.vector.reduce.fadd.v128bf16(bfloat 0.000000e+00, <128 x bfloat> undef)
; FP-REDUCE-ZVFH-NEXT: Cost Model: Invalid cost for instruction: %NXV1 = call bfloat @llvm.vector.reduce.fadd.nxv1bf16(bfloat 0.000000e+00, <vscale x 1 x bfloat> undef)
; FP-REDUCE-ZVFH-NEXT: Cost Model: Invalid cost for instruction: %NXV2 = call bfloat @llvm.vector.reduce.fadd.nxv2bf16(bfloat 0.000000e+00, <vscale x 2 x bfloat> undef)
; FP-REDUCE-ZVFH-NEXT: Cost Model: Invalid cost for instruction: %NXV4 = call bfloat @llvm.vector.reduce.fadd.nxv4bf16(bfloat 0.000000e+00, <vscale x 4 x bfloat> undef)
@@ -290,14 +290,14 @@ define void @reduce_ordered_fadd_bfloat() {
; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; FP-REDUCE-ZVFHMIN-LABEL: 'reduce_ordered_fadd_bfloat'
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1 = call bfloat @llvm.vector.reduce.fadd.v1bf16(bfloat 0.000000e+00, <1 x bfloat> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V2 = call bfloat @llvm.vector.reduce.fadd.v2bf16(bfloat 0.000000e+00, <2 x bfloat> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V4 = call bfloat @llvm.vector.reduce.fadd.v4bf16(bfloat 0.000000e+00, <4 x bfloat> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V8 = call bfloat @llvm.vector.reduce.fadd.v8bf16(bfloat 0.000000e+00, <8 x bfloat> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 63 for instruction: %V16 = call bfloat @llvm.vector.reduce.fadd.v16bf16(bfloat 0.000000e+00, <16 x bfloat> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 127 for instruction: %v32 = call bfloat @llvm.vector.reduce.fadd.v32bf16(bfloat 0.000000e+00, <32 x bfloat> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 255 for instruction: %V64 = call bfloat @llvm.vector.reduce.fadd.v64bf16(bfloat 0.000000e+00, <64 x bfloat> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 510 for instruction: %V128 = call bfloat @llvm.vector.reduce.fadd.v128bf16(bfloat 0.000000e+00, <128 x bfloat> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V1 = call bfloat @llvm.vector.reduce.fadd.v1bf16(bfloat 0.000000e+00, <1 x bfloat> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V2 = call bfloat @llvm.vector.reduce.fadd.v2bf16(bfloat 0.000000e+00, <2 x bfloat> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V4 = call bfloat @llvm.vector.reduce.fadd.v4bf16(bfloat 0.000000e+00, <4 x bfloat> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %V8 = call bfloat @llvm.vector.reduce.fadd.v8bf16(bfloat 0.000000e+00, <8 x bfloat> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 79 for instruction: %V16 = call bfloat @llvm.vector.reduce.fadd.v16bf16(bfloat 0.000000e+00, <16 x bfloat> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 159 for instruction: %v32 = call bfloat @llvm.vector.reduce.fadd.v32bf16(bfloat 0.000000e+00, <32 x bfloat> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 319 for instruction: %V64 = call bfloat @llvm.vector.reduce.fadd.v64bf16(bfloat 0.000000e+00, <64 x bfloat> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 638 for instruction: %V128 = call bfloat @llvm.vector.reduce.fadd.v128bf16(bfloat 0.000000e+00, <128 x bfloat> undef)
; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV1 = call bfloat @llvm.vector.reduce.fadd.nxv1bf16(bfloat 0.000000e+00, <vscale x 1 x bfloat> undef)
; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV2 = call bfloat @llvm.vector.reduce.fadd.nxv2bf16(bfloat 0.000000e+00, <vscale x 2 x bfloat> undef)
; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV4 = call bfloat @llvm.vector.reduce.fadd.nxv4bf16(bfloat 0.000000e+00, <vscale x 4 x bfloat> undef)
@@ -359,31 +359,31 @@ define void @reduce_ordered_fadd_bfloat() {
define void @reduce_ordered_fadd_half() {
; FP-REDUCE-ZVFH-LABEL: 'reduce_ordered_fadd_half'
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1 = call half @llvm.vector.reduce.fadd.v1f16(half 0.000000e+00, <1 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2 = call half @llvm.vector.reduce.fadd.v2f16(half 0.000000e+00, <2 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V4 = call half @llvm.vector.reduce.fadd.v4f16(half 0.000000e+00, <4 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %V8 = call half @llvm.vector.reduce.fadd.v8f16(half 0.000000e+00, <8 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %V16 = call half @llvm.vector.reduce.fadd.v16f16(half 0.000000e+00, <16 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 34 for instruction: %v32 = call half @llvm.vector.reduce.fadd.v32f16(half 0.000000e+00, <32 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 66 for instruction: %V64 = call half @llvm.vector.reduce.fadd.v64f16(half 0.000000e+00, <64 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 130 for instruction: %V128 = call half @llvm.vector.reduce.fadd.v128f16(half 0.000000e+00, <128 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %NXV1 = call half @llvm.vector.reduce.fadd.nxv1f16(half 0.000000e+00, <vscale x 1 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %NXV2 = call half @llvm.vector.reduce.fadd.nxv2f16(half 0.000000e+00, <vscale x 2 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %NXV4 = call half @llvm.vector.reduce.fadd.nxv4f16(half 0.000000e+00, <vscale x 4 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %NXV8 = call half @llvm.vector.reduce.fadd.nxv8f16(half 0.000000e+00, <vscale x 8 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 34 for instruction: %NXV16 = call half @llvm.vector.reduce.fadd.nxv16f16(half 0.000000e+00, <vscale x 16 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 66 for instruction: %NXV32 = call half @llvm.vector.reduce.fadd.nxv32f16(half 0.000000e+00, <vscale x 32 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V1 = call half @llvm.vector.reduce.fadd.v1f16(half 0.000000e+00, <1 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V2 = call half @llvm.vector.reduce.fadd.v2f16(half 0.000000e+00, <2 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V4 = call half @llvm.vector.reduce.fadd.v4f16(half 0.000000e+00, <4 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %V8 = call half @llvm.vector.reduce.fadd.v8f16(half 0.000000e+00, <8 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V16 = call half @llvm.vector.reduce.fadd.v16f16(half 0.000000e+00, <16 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 35 for instruction: %v32 = call half @llvm.vector.reduce.fadd.v32f16(half 0.000000e+00, <32 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 67 for instruction: %V64 = call half @llvm.vector.reduce.fadd.v64f16(half 0.000000e+00, <64 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 131 for instruction: %V128 = call half @llvm.vector.reduce.fadd.v128f16(half 0.000000e+00, <128 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %NXV1 = call half @llvm.vector.reduce.fadd.nxv1f16(half 0.000000e+00, <vscale x 1 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %NXV2 = call half @llvm.vector.reduce.fadd.nxv2f16(half 0.000000e+00, <vscale x 2 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %NXV4 = call half @llvm.vector.reduce.fadd.nxv4f16(half 0.000000e+00, <vscale x 4 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %NXV8 = call half @llvm.vector.reduce.fadd.nxv8f16(half 0.000000e+00, <vscale x 8 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 35 for instruction: %NXV16 = call half @llvm.vector.reduce.fadd.nxv16f16(half 0.000000e+00, <vscale x 16 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 67 for instruction: %NXV32 = call half @llvm.vector.reduce.fadd.nxv32f16(half 0.000000e+00, <vscale x 32 x half> undef)
; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; FP-REDUCE-ZVFHMIN-LABEL: 'reduce_ordered_fadd_half'
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1 = call half @llvm.vector.reduce.fadd.v1f16(half 0.000000e+00, <1 x half> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V2 = call half @llvm.vector.reduce.fadd.v2f16(half 0.000000e+00, <2 x half> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V4 = call half @llvm.vector.reduce.fadd.v4f16(half 0.000000e+00, <4 x half> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V8 = call half @llvm.vector.reduce.fadd.v8f16(half 0.000000e+00, <8 x half> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 63 for instruction: %V16 = call half @llvm.vector.reduce.fadd.v16f16(half 0.000000e+00, <16 x half> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 127 for instruction: %v32 = call half @llvm.vector.reduce.fadd.v32f16(half 0.000000e+00, <32 x half> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 255 for instruction: %V64 = call half @llvm.vector.reduce.fadd.v64f16(half 0.000000e+00, <64 x half> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 510 for instruction: %V128 = call half @llvm.vector.reduce.fadd.v128f16(half 0.000000e+00, <128 x half> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V1 = call half @llvm.vector.reduce.fadd.v1f16(half 0.000000e+00, <1 x half> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V2 = call half @llvm.vector.reduce.fadd.v2f16(half 0.000000e+00, <2 x half> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V4 = call half @llvm.vector.reduce.fadd.v4f16(half 0.000000e+00, <4 x half> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %V8 = call half @llvm.vector.reduce.fadd.v8f16(half 0.000000e+00, <8 x half> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 79 for instruction: %V16 = call half @llvm.vector.reduce.fadd.v16f16(half 0.000000e+00, <16 x half> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 159 for instruction: %v32 = call half @llvm.vector.reduce.fadd.v32f16(half 0.000000e+00, <32 x half> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 319 for instruction: %V64 = call half @llvm.vector.reduce.fadd.v64f16(half 0.000000e+00, <64 x half> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 638 for instruction: %V128 = call half @llvm.vector.reduce.fadd.v128f16(half 0.000000e+00, <128 x half> undef)
; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV1 = call half @llvm.vector.reduce.fadd.nxv1f16(half 0.000000e+00, <vscale x 1 x half> undef)
; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV2 = call half @llvm.vector.reduce.fadd.nxv2f16(half 0.000000e+00, <vscale x 2 x half> undef)
; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV4 = call half @llvm.vector.reduce.fadd.nxv4f16(half 0.000000e+00, <vscale x 4 x half> undef)
@@ -445,19 +445,19 @@ define void @reduce_ordered_fadd_half() {
define void @reduce_ordered_fadd_float() {
; FP-REDUCE-LABEL: 'reduce_ordered_fadd_float'
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1 = call float @llvm.vector.reduce.fadd.v1f32(float 0.000000e+00, <1 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2 = call float @llvm.vector.reduce.fadd.v2f32(float 0.000000e+00, <2 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V4 = call float @llvm.vector.reduce.fadd.v4f32(float 0.000000e+00, <4 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %V8 = call float @llvm.vector.reduce.fadd.v8f32(float 0.000000e+00, <8 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %V16 = call float @llvm.vector.reduce.fadd.v16f32(float 0.000000e+00, <16 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 34 for instruction: %v32 = call float @llvm.vector.reduce.fadd.v32f32(float 0.000000e+00, <32 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 66 for instruction: %V64 = call float @llvm.vector.reduce.fadd.v64f32(float 0.000000e+00, <64 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 130 for instruction: %V128 = call float @llvm.vector.reduce.fadd.v128f32(float 0.000000e+00, <128 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %NXV1 = call float @llvm.vector.reduce.fadd.nxv1f32(float 0.000000e+00, <vscale x 1 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %NXV2 = call float @llvm.vector.reduce.fadd.nxv2f32(float 0.000000e+00, <vscale x 2 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %NXV4 = call float @llvm.vector.reduce.fadd.nxv4f32(float 0.000000e+00, <vscale x 4 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %NXV8 = call float @llvm.vector.reduce.fadd.nxv8f32(float 0.000000e+00, <vscale x 8 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 34 for instruction: %NXV16 = call float @llvm.vector.reduce.fadd.nxv16f32(float 0.000000e+00, <vscale x 16 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V1 = call float @llvm.vector.reduce.fadd.v1f32(float 0.000000e+00, <1 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V2 = call float @llvm.vector.reduce.fadd.v2f32(float 0.000000e+00, <2 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V4 = call float @llvm.vector.reduce.fadd.v4f32(float 0.000000e+00, <4 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %V8 = call float @llvm.vector.reduce.fadd.v8f32(float 0.000000e+00, <8 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V16 = call float @llvm.vector.reduce.fadd.v16f32(float 0.000000e+00, <16 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 35 for instruction: %v32 = call float @llvm.vector.reduce.fadd.v32f32(float 0.000000e+00, <32 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 67 for instruction: %V64 = call float @llvm.vector.reduce.fadd.v64f32(float 0.000000e+00, <64 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 131 for instruction: %V128 = call float @llvm.vector.reduce.fadd.v128f32(float 0.000000e+00, <128 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %NXV1 = call float @llvm.vector.reduce.fadd.nxv1f32(float 0.000000e+00, <vscale x 1 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %NXV2 = call float @llvm.vector.reduce.fadd.nxv2f32(float 0.000000e+00, <vscale x 2 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %NXV4 = call float @llvm.vector.reduce.fadd.nxv4f32(float 0.000000e+00, <vscale x 4 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %NXV8 = call float @llvm.vector.reduce.fadd.nxv8f32(float 0.000000e+00, <vscale x 8 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 35 for instruction: %NXV16 = call float @llvm.vector.reduce.fadd.nxv16f32(float 0.000000e+00, <vscale x 16 x float> undef)
; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; SIZE-LABEL: 'reduce_ordered_fadd_float'
@@ -494,18 +494,18 @@ define void @reduce_ordered_fadd_float() {
define void @reduce_ordered_fadd_double() {
; FP-REDUCE-LABEL: 'reduce_ordered_fadd_double'
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1 = call double @llvm.vector.reduce.fadd.v1f64(double 0.000000e+00, <1 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2 = call double @llvm.vector.reduce.fadd.v2f64(double 0.000000e+00, <2 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V4 = call double @llvm.vector.reduce.fadd.v4f64(double 0.000000e+00, <4 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %V8 = call double @llvm.vector.reduce.fadd.v8f64(double 0.000000e+00, <8 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %V16 = call double @llvm.vector.reduce.fadd.v16f64(double 0.000000e+00, <16 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 34 for instruction: %v32 = call double @llvm.vector.reduce.fadd.v32f64(double 0.000000e+00, <32 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 66 for instruction: %V64 = call double @llvm.vector.reduce.fadd.v64f64(double 0.000000e+00, <64 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 130 for instruction: %V128 = call double @llvm.vector.reduce.fadd.v128f64(double 0.000000e+00, <128 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %NXV1 = call double @llvm.vector.reduce.fadd.nxv1f64(double 0.000000e+00, <vscale x 1 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %NXV2 = call double @llvm.vector.reduce.fadd.nxv2f64(double 0.000000e+00, <vscale x 2 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %NXV4 = call double @llvm.vector.reduce.fadd.nxv4f64(double 0.000000e+00, <vscale x 4 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %NXV8 = call double @llvm.vector.reduce.fadd.nxv8f64(double 0.000000e+00, <vscale x 8 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V1 = call double @llvm.vector.reduce.fadd.v1f64(double 0.000000e+00, <1 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V2 = call double @llvm.vector.reduce.fadd.v2f64(double 0.000000e+00, <2 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V4 = call double @llvm.vector.reduce.fadd.v4f64(double 0.000000e+00, <4 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %V8 = call double @llvm.vector.reduce.fadd.v8f64(double 0.000000e+00, <8 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V16 = call double @llvm.vector.reduce.fadd.v16f64(double 0.000000e+00, <16 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 35 for instruction: %v32 = call double @llvm.vector.reduce.fadd.v32f64(double 0.000000e+00, <32 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 67 for instruction: %V64 = call double @llvm.vector.reduce.fadd.v64f64(double 0.000000e+00, <64 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 131 for instruction: %V128 = call double @llvm.vector.reduce.fadd.v128f64(double 0.000000e+00, <128 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %NXV1 = call double @llvm.vector.reduce.fadd.nxv1f64(double 0.000000e+00, <vscale x 1 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %NXV2 = call double @llvm.vector.reduce.fadd.nxv2f64(double 0.000000e+00, <vscale x 2 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %NXV4 = call double @llvm.vector.reduce.fadd.nxv4f64(double 0.000000e+00, <vscale x 4 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %NXV8 = call double @llvm.vector.reduce.fadd.nxv8f64(double 0.000000e+00, <vscale x 8 x double> undef)
; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; SIZE-LABEL: 'reduce_ordered_fadd_double'
diff --git a/llvm/test/Analysis/CostModel/RISCV/reduce-fmaximum.ll b/llvm/test/Analysis/CostModel/RISCV/reduce-fmaximum.ll
index d937d066d6ebe..ba9e42a2598a9 100644
--- a/llvm/test/Analysis/CostModel/RISCV/reduce-fmaximum.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/reduce-fmaximum.ll
@@ -6,20 +6,20 @@
define float @reduce_fmaximum_f32(float %arg) {
; CHECK-LABEL: 'reduce_fmaximum_f32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V2 = call float @llvm.vector.reduce.fmaximum.v2f32(<2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V4 = call float @llvm.vector.reduce.fmaximum.v4f32(<4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V8 = call float @llvm.vector.reduce.fmaximum.v8f32(<8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %V16 = call float @llvm.vector.reduce.fmaximum.v16f32(<16 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %V32 = call float @llvm.vector.reduce.fmaximum.v32f32(<32 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %V64 = call float @llvm.vector.reduce.fmaximum.v64f32(<64 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %V128 = call float @llvm.vector.reduce.fmaximum.v128f32(<128 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %1 = call fast float @llvm.vector.reduce.fmaximum.v2f32(<2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %2 = call fast float @llvm.vector.reduce.fmaximum.v4f32(<4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %3 = call fast float @llvm.vector.reduce.fmaximum.v8f32(<8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %4 = call fast float @llvm.vector.reduce.fmaximum.v16f32(<16 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %5 = call fast float @llvm.vector.reduce.fmaximum.v32f32(<32 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %6 = call fast float @llvm.vector.reduce.fmaximum.v64f32(<64 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %7 = call fast float @llvm.vector.reduce.fmaximum.v128f32(<128 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V2 = call float @llvm.vector.reduce.fmaximum.v2f32(<2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V4 = call float @llvm.vector.reduce.fmaximum.v4f32(<4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %V8 = call float @llvm.vector.reduce.fmaximum.v8f32(<8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %V16 = call float @llvm.vector.reduce.fmaximum.v16f32(<16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %V32 = call float @llvm.vector.reduce.fmaximum.v32f32(<32 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %V64 = call float @llvm.vector.reduce.fmaximum.v64f32(<64 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %V128 = call float @llvm.vector.reduce.fmaximum.v128f32(<128 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %1 = call fast float @llvm.vector.reduce.fmaximum.v2f32(<2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %2 = call fast float @llvm.vector.reduce.fmaximum.v4f32(<4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %3 = call fast float @llvm.vector.reduce.fmaximum.v8f32(<8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %4 = call fast float @llvm.vector.reduce.fmaximum.v16f32(<16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %5 = call fast float @llvm.vector.reduce.fmaximum.v32f32(<32 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %6 = call fast float @llvm.vector.reduce.fmaximum.v64f32(<64 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %7 = call fast float @llvm.vector.reduce.fmaximum.v128f32(<128 x float> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float undef
;
; SIZE-LABEL: 'reduce_fmaximum_f32'
@@ -58,18 +58,18 @@ ret float undef
define double @reduce_fmaximum_f64(double %arg) {
; CHECK-LABEL: 'reduce_fmaximum_f64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V2 = call double @llvm.vector.reduce.fmaximum.v2f64(<2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V4 = call double @llvm.vector.reduce.fmaximum.v4f64(<4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %V8 = call double @llvm.vector.reduce.fmaximum.v8f64(<8 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V16 = call double @llvm.vector.reduce.fmaximum.v16f64(<16 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V32 = call double @llvm.vector.reduce.fmaximum.v32f64(<32 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V64 = call double @llvm.vector.reduce.fmaximum.v64f64(<64 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %1 = call fast double @llvm.vector.reduce.fmaximum.v2f64(<2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %2 = call fast double @llvm.vector.reduce.fmaximum.v4f64(<4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %3 = call fast double @llvm.vector.reduce.fmaximum.v8f64(<8 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %4 = call fast double @llvm.vector.reduce.fmaximum.v16f64(<16 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %5 = call fast double @llvm.vector.reduce.fmaximum.v32f64(<32 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %6 = call fast double @llvm.vector.reduce.fmaximum.v64f64(<64 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V2 = call double @llvm.vector.reduce.fmaximum.v2f64(<2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V4 = call double @llvm.vector.reduce.fmaximum.v4f64(<4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %V8 = call double @llvm.vector.reduce.fmaximum.v8f64(<8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %V16 = call double @llvm.vector.reduce.fmaximum.v16f64(<16 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %V32 = call double @llvm.vector.reduce.fmaximum.v32f64(<32 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %V64 = call double @llvm.vector.reduce.fmaximum.v64f64(<64 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %1 = call fast double @llvm.vector.reduce.fmaximum.v2f64(<2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %2 = call fast double @llvm.vector.reduce.fmaximum.v4f64(<4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %3 = call fast double @llvm.vector.reduce.fmaximum.v8f64(<8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %4 = call fast double @llvm.vector.reduce.fmaximum.v16f64(<16 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %5 = call fast double @llvm.vector.reduce.fmaximum.v32f64(<32 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %6 = call fast double @llvm.vector.reduce.fmaximum.v64f64(<64 x double> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double undef
;
; SIZE-LABEL: 'reduce_fmaximum_f64'
diff --git a/llvm/test/Analysis/CostModel/RISCV/reduce-fminimum.ll b/llvm/test/Analysis/CostModel/RISCV/reduce-fminimum.ll
index 397f25c2599b1..9f41a3be7a5a7 100644
--- a/llvm/test/Analysis/CostModel/RISCV/reduce-fminimum.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/reduce-fminimum.ll
@@ -6,13 +6,13 @@
define float @reduce_fmaximum_f32(float %arg) {
; CHECK-LABEL: 'reduce_fmaximum_f32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V2 = call float @llvm.vector.reduce.fminimum.v2f32(<2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V4 = call float @llvm.vector.reduce.fminimum.v4f32(<4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V8 = call float @llvm.vector.reduce.fminimum.v8f32(<8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %V16 = call float @llvm.vector.reduce.fminimum.v16f32(<16 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %V32 = call float @llvm.vector.reduce.fminimum.v32f32(<32 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %V64 = call float @llvm.vector.reduce.fminimum.v64f32(<64 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %V128 = call float @llvm.vector.reduce.fminimum.v128f32(<128 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V2 = call float @llvm.vector.reduce.fminimum.v2f32(<2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V4 = call float @llvm.vector.reduce.fminimum.v4f32(<4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %V8 = call float @llvm.vector.reduce.fminimum.v8f32(<8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %V16 = call float @llvm.vector.reduce.fminimum.v16f32(<16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %V32 = call float @llvm.vector.reduce.fminimum.v32f32(<32 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %V64 = call float @llvm.vector.reduce.fminimum.v64f32(<64 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %V128 = call float @llvm.vector.reduce.fminimum.v128f32(<128 x float> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float undef
;
; SIZE-LABEL: 'reduce_fmaximum_f32'
@@ -37,12 +37,12 @@ ret float undef
define double @reduce_fmaximum_f64(double %arg) {
; CHECK-LABEL: 'reduce_fmaximum_f64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V2 = call double @llvm.vector.reduce.fminimum.v2f64(<2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V4 = call double @llvm.vector.reduce.fminimum.v4f64(<4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %V8 = call double @llvm.vector.reduce.fminimum.v8f64(<8 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V16 = call double @llvm.vector.reduce.fminimum.v16f64(<16 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V32 = call double @llvm.vector.reduce.fminimum.v32f64(<32 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V64 = call double @llvm.vector.reduce.fminimum.v64f64(<64 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V2 = call double @llvm.vector.reduce.fminimum.v2f64(<2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V4 = call double @llvm.vector.reduce.fminimum.v4f64(<4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %V8 = call double @llvm.vector.reduce.fminimum.v8f64(<8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %V16 = call double @llvm.vector.reduce.fminimum.v16f64(<16 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %V32 = call double @llvm.vector.reduce.fminimum.v32f64(<32 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %V64 = call double @llvm.vector.reduce.fminimum.v64f64(<64 x double> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double undef
;
; SIZE-LABEL: 'reduce_fmaximum_f64'
diff --git a/llvm/test/Analysis/CostModel/RISCV/reduce-fmul.ll b/llvm/test/Analysis/CostModel/RISCV/reduce-fmul.ll
index 2fb74b479e94a..78471d7d57b6a 100644
--- a/llvm/test/Analysis/CostModel/RISCV/reduce-fmul.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/reduce-fmul.ll
@@ -5,14 +5,14 @@
define void @reduce_fmul_bfloat() {
; FP-REDUCE-LABEL: 'reduce_fmul_bfloat'
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call fast bfloat @llvm.vector.reduce.fmul.v1bf16(bfloat 0.000000e+00, <1 x bfloat> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V2 = call fast bfloat @llvm.vector.reduce.fmul.v2bf16(bfloat 0.000000e+00, <2 x bfloat> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V4 = call fast bfloat @llvm.vector.reduce.fmul.v4bf16(bfloat 0.000000e+00, <4 x bfloat> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %V8 = call fast bfloat @llvm.vector.reduce.fmul.v8bf16(bfloat 0.000000e+00, <8 x bfloat> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 73 for instruction: %V16 = call fast bfloat @llvm.vector.reduce.fmul.v16bf16(bfloat 0.000000e+00, <16 x bfloat> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 211 for instruction: %v32 = call fast bfloat @llvm.vector.reduce.fmul.v32bf16(bfloat 0.000000e+00, <32 x bfloat> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 541 for instruction: %V64 = call fast bfloat @llvm.vector.reduce.fmul.v64bf16(bfloat 0.000000e+00, <64 x bfloat> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 573 for instruction: %V128 = call fast bfloat @llvm.vector.reduce.fmul.v128bf16(bfloat 0.000000e+00, <128 x bfloat> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call fast bfloat @llvm.vector.reduce.fmul.v1bf16(bfloat 0.000000e+00, <1 x bfloat> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V2 = call fast bfloat @llvm.vector.reduce.fmul.v2bf16(bfloat 0.000000e+00, <2 x bfloat> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V4 = call fast bfloat @llvm.vector.reduce.fmul.v4bf16(bfloat 0.000000e+00, <4 x bfloat> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 29 for instruction: %V8 = call fast bfloat @llvm.vector.reduce.fmul.v8bf16(bfloat 0.000000e+00, <8 x bfloat> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 74 for instruction: %V16 = call fast bfloat @llvm.vector.reduce.fmul.v16bf16(bfloat 0.000000e+00, <16 x bfloat> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 212 for instruction: %v32 = call fast bfloat @llvm.vector.reduce.fmul.v32bf16(bfloat 0.000000e+00, <32 x bfloat> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 542 for instruction: %V64 = call fast bfloat @llvm.vector.reduce.fmul.v64bf16(bfloat 0.000000e+00, <64 x bfloat> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 574 for instruction: %V128 = call fast bfloat @llvm.vector.reduce.fmul.v128bf16(bfloat 0.000000e+00, <128 x bfloat> undef)
; FP-REDUCE-NEXT: Cost Model: Invalid cost for instruction: %NXV1 = call fast bfloat @llvm.vector.reduce.fmul.nxv1bf16(bfloat 0.000000e+00, <vscale x 1 x bfloat> undef)
; FP-REDUCE-NEXT: Cost Model: Invalid cost for instruction: %NXV2 = call fast bfloat @llvm.vector.reduce.fmul.nxv2bf16(bfloat 0.000000e+00, <vscale x 2 x bfloat> undef)
; FP-REDUCE-NEXT: Cost Model: Invalid cost for instruction: %NXV4 = call fast bfloat @llvm.vector.reduce.fmul.nxv4bf16(bfloat 0.000000e+00, <vscale x 4 x bfloat> undef)
@@ -57,14 +57,14 @@ define void @reduce_fmul_bfloat() {
define void @reduce_fmul_half() {
; FP-REDUCE-ZVFH-LABEL: 'reduce_fmul_half'
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call fast half @llvm.vector.reduce.fmul.v1f16(half 0.000000e+00, <1 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V2 = call fast half @llvm.vector.reduce.fmul.v2f16(half 0.000000e+00, <2 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %V4 = call fast half @llvm.vector.reduce.fmul.v4f16(half 0.000000e+00, <4 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V8 = call fast half @llvm.vector.reduce.fmul.v8f16(half 0.000000e+00, <8 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 49 for instruction: %V16 = call fast half @llvm.vector.reduce.fmul.v16f16(half 0.000000e+00, <16 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 151 for instruction: %v32 = call fast half @llvm.vector.reduce.fmul.v32f16(half 0.000000e+00, <32 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 541 for instruction: %V64 = call fast half @llvm.vector.reduce.fmul.v64f16(half 0.000000e+00, <64 x half> undef)
-; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 573 for instruction: %V128 = call fast half @llvm.vector.reduce.fmul.v128f16(half 0.000000e+00, <128 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call fast half @llvm.vector.reduce.fmul.v1f16(half 0.000000e+00, <1 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V2 = call fast half @llvm.vector.reduce.fmul.v2f16(half 0.000000e+00, <2 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V4 = call fast half @llvm.vector.reduce.fmul.v4f16(half 0.000000e+00, <4 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %V8 = call fast half @llvm.vector.reduce.fmul.v8f16(half 0.000000e+00, <8 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 50 for instruction: %V16 = call fast half @llvm.vector.reduce.fmul.v16f16(half 0.000000e+00, <16 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 152 for instruction: %v32 = call fast half @llvm.vector.reduce.fmul.v32f16(half 0.000000e+00, <32 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 542 for instruction: %V64 = call fast half @llvm.vector.reduce.fmul.v64f16(half 0.000000e+00, <64 x half> undef)
+; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 574 for instruction: %V128 = call fast half @llvm.vector.reduce.fmul.v128f16(half 0.000000e+00, <128 x half> undef)
; FP-REDUCE-ZVFH-NEXT: Cost Model: Invalid cost for instruction: %NXV1 = call fast half @llvm.vector.reduce.fmul.nxv1f16(half 0.000000e+00, <vscale x 1 x half> undef)
; FP-REDUCE-ZVFH-NEXT: Cost Model: Invalid cost for instruction: %NXV2 = call fast half @llvm.vector.reduce.fmul.nxv2f16(half 0.000000e+00, <vscale x 2 x half> undef)
; FP-REDUCE-ZVFH-NEXT: Cost Model: Invalid cost for instruction: %NXV4 = call fast half @llvm.vector.reduce.fmul.nxv4f16(half 0.000000e+00, <vscale x 4 x half> undef)
@@ -74,14 +74,14 @@ define void @reduce_fmul_half() {
; FP-REDUCE-ZVFH-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; FP-REDUCE-ZVFHMIN-LABEL: 'reduce_fmul_half'
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call fast half @llvm.vector.reduce.fmul.v1f16(half 0.000000e+00, <1 x half> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V2 = call fast half @llvm.vector.reduce.fmul.v2f16(half 0.000000e+00, <2 x half> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V4 = call fast half @llvm.vector.reduce.fmul.v4f16(half 0.000000e+00, <4 x half> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 28 for instruction: %V8 = call fast half @llvm.vector.reduce.fmul.v8f16(half 0.000000e+00, <8 x half> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 73 for instruction: %V16 = call fast half @llvm.vector.reduce.fmul.v16f16(half 0.000000e+00, <16 x half> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 211 for instruction: %v32 = call fast half @llvm.vector.reduce.fmul.v32f16(half 0.000000e+00, <32 x half> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 541 for instruction: %V64 = call fast half @llvm.vector.reduce.fmul.v64f16(half 0.000000e+00, <64 x half> undef)
-; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 573 for instruction: %V128 = call fast half @llvm.vector.reduce.fmul.v128f16(half 0.000000e+00, <128 x half> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call fast half @llvm.vector.reduce.fmul.v1f16(half 0.000000e+00, <1 x half> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V2 = call fast half @llvm.vector.reduce.fmul.v2f16(half 0.000000e+00, <2 x half> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V4 = call fast half @llvm.vector.reduce.fmul.v4f16(half 0.000000e+00, <4 x half> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 29 for instruction: %V8 = call fast half @llvm.vector.reduce.fmul.v8f16(half 0.000000e+00, <8 x half> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 74 for instruction: %V16 = call fast half @llvm.vector.reduce.fmul.v16f16(half 0.000000e+00, <16 x half> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 212 for instruction: %v32 = call fast half @llvm.vector.reduce.fmul.v32f16(half 0.000000e+00, <32 x half> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 542 for instruction: %V64 = call fast half @llvm.vector.reduce.fmul.v64f16(half 0.000000e+00, <64 x half> undef)
+; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 574 for instruction: %V128 = call fast half @llvm.vector.reduce.fmul.v128f16(half 0.000000e+00, <128 x half> undef)
; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV1 = call fast half @llvm.vector.reduce.fmul.nxv1f16(half 0.000000e+00, <vscale x 1 x half> undef)
; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV2 = call fast half @llvm.vector.reduce.fmul.nxv2f16(half 0.000000e+00, <vscale x 2 x half> undef)
; FP-REDUCE-ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV4 = call fast half @llvm.vector.reduce.fmul.nxv4f16(half 0.000000e+00, <vscale x 4 x half> undef)
@@ -126,14 +126,14 @@ define void @reduce_fmul_half() {
define void @reduce_fmul_float() {
; FP-REDUCE-LABEL: 'reduce_fmul_float'
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call fast float @llvm.vector.reduce.fmul.v1f32(float 0.000000e+00, <1 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V2 = call fast float @llvm.vector.reduce.fmul.v2f32(float 0.000000e+00, <2 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %V4 = call fast float @llvm.vector.reduce.fmul.v4f32(float 0.000000e+00, <4 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 37 for instruction: %V8 = call fast float @llvm.vector.reduce.fmul.v8f32(float 0.000000e+00, <8 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 121 for instruction: %V16 = call fast float @llvm.vector.reduce.fmul.v16f32(float 0.000000e+00, <16 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 451 for instruction: %v32 = call fast float @llvm.vector.reduce.fmul.v32f32(float 0.000000e+00, <32 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 483 for instruction: %V64 = call fast float @llvm.vector.reduce.fmul.v64f32(float 0.000000e+00, <64 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 547 for instruction: %V128 = call fast float @llvm.vector.reduce.fmul.v128f32(float 0.000000e+00, <128 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call fast float @llvm.vector.reduce.fmul.v1f32(float 0.000000e+00, <1 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V2 = call fast float @llvm.vector.reduce.fmul.v2f32(float 0.000000e+00, <2 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V4 = call fast float @llvm.vector.reduce.fmul.v4f32(float 0.000000e+00, <4 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 38 for instruction: %V8 = call fast float @llvm.vector.reduce.fmul.v8f32(float 0.000000e+00, <8 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 122 for instruction: %V16 = call fast float @llvm.vector.reduce.fmul.v16f32(float 0.000000e+00, <16 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 452 for instruction: %v32 = call fast float @llvm.vector.reduce.fmul.v32f32(float 0.000000e+00, <32 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 484 for instruction: %V64 = call fast float @llvm.vector.reduce.fmul.v64f32(float 0.000000e+00, <64 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 548 for instruction: %V128 = call fast float @llvm.vector.reduce.fmul.v128f32(float 0.000000e+00, <128 x float> undef)
; FP-REDUCE-NEXT: Cost Model: Invalid cost for instruction: %NXV1 = call fast float @llvm.vector.reduce.fmul.nxv1f32(float 0.000000e+00, <vscale x 1 x float> undef)
; FP-REDUCE-NEXT: Cost Model: Invalid cost for instruction: %NXV2 = call fast float @llvm.vector.reduce.fmul.nxv2f32(float 0.000000e+00, <vscale x 2 x float> undef)
; FP-REDUCE-NEXT: Cost Model: Invalid cost for instruction: %NXV4 = call fast float @llvm.vector.reduce.fmul.nxv4f32(float 0.000000e+00, <vscale x 4 x float> undef)
@@ -175,14 +175,14 @@ define void @reduce_fmul_float() {
define void @reduce_fmul_double() {
; FP-REDUCE-LABEL: 'reduce_fmul_double'
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call fast double @llvm.vector.reduce.fmul.v1f64(double 0.000000e+00, <1 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V2 = call fast double @llvm.vector.reduce.fmul.v2f64(double 0.000000e+00, <2 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 25 for instruction: %V4 = call fast double @llvm.vector.reduce.fmul.v4f64(double 0.000000e+00, <4 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 91 for instruction: %V8 = call fast double @llvm.vector.reduce.fmul.v8f64(double 0.000000e+00, <8 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 361 for instruction: %V16 = call fast double @llvm.vector.reduce.fmul.v16f64(double 0.000000e+00, <16 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 393 for instruction: %v32 = call fast double @llvm.vector.reduce.fmul.v32f64(double 0.000000e+00, <32 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 457 for instruction: %V64 = call fast double @llvm.vector.reduce.fmul.v64f64(double 0.000000e+00, <64 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 585 for instruction: %V128 = call fast double @llvm.vector.reduce.fmul.v128f64(double 0.000000e+00, <128 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call fast double @llvm.vector.reduce.fmul.v1f64(double 0.000000e+00, <1 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V2 = call fast double @llvm.vector.reduce.fmul.v2f64(double 0.000000e+00, <2 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %V4 = call fast double @llvm.vector.reduce.fmul.v4f64(double 0.000000e+00, <4 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 92 for instruction: %V8 = call fast double @llvm.vector.reduce.fmul.v8f64(double 0.000000e+00, <8 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 362 for instruction: %V16 = call fast double @llvm.vector.reduce.fmul.v16f64(double 0.000000e+00, <16 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 394 for instruction: %v32 = call fast double @llvm.vector.reduce.fmul.v32f64(double 0.000000e+00, <32 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 458 for instruction: %V64 = call fast double @llvm.vector.reduce.fmul.v64f64(double 0.000000e+00, <64 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 586 for instruction: %V128 = call fast double @llvm.vector.reduce.fmul.v128f64(double 0.000000e+00, <128 x double> undef)
; FP-REDUCE-NEXT: Cost Model: Invalid cost for instruction: %NXV1 = call fast double @llvm.vector.reduce.fmul.nxv1f64(double 0.000000e+00, <vscale x 1 x double> undef)
; FP-REDUCE-NEXT: Cost Model: Invalid cost for instruction: %NXV2 = call fast double @llvm.vector.reduce.fmul.nxv2f64(double 0.000000e+00, <vscale x 2 x double> undef)
; FP-REDUCE-NEXT: Cost Model: Invalid cost for instruction: %NXV4 = call fast double @llvm.vector.reduce.fmul.nxv4f64(double 0.000000e+00, <vscale x 4 x double> undef)
@@ -221,14 +221,14 @@ define void @reduce_fmul_double() {
define void @reduce_ordered_fmul_bfloat() {
; FP-REDUCE-LABEL: 'reduce_ordered_fmul_bfloat'
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1 = call bfloat @llvm.vector.reduce.fmul.v1bf16(bfloat 0.000000e+00, <1 x bfloat> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V2 = call bfloat @llvm.vector.reduce.fmul.v2bf16(bfloat 0.000000e+00, <2 x bfloat> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V4 = call bfloat @llvm.vector.reduce.fmul.v4bf16(bfloat 0.000000e+00, <4 x bfloat> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V8 = call bfloat @llvm.vector.reduce.fmul.v8bf16(bfloat 0.000000e+00, <8 x bfloat> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 63 for instruction: %V16 = call bfloat @llvm.vector.reduce.fmul.v16bf16(bfloat 0.000000e+00, <16 x bfloat> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 127 for instruction: %v32 = call bfloat @llvm.vector.reduce.fmul.v32bf16(bfloat 0.000000e+00, <32 x bfloat> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 255 for instruction: %V64 = call bfloat @llvm.vector.reduce.fmul.v64bf16(bfloat 0.000000e+00, <64 x bfloat> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 510 for instruction: %V128 = call bfloat @llvm.vector.reduce.fmul.v128bf16(bfloat 0.000000e+00, <128 x bfloat> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V1 = call bfloat @llvm.vector.reduce.fmul.v1bf16(bfloat 0.000000e+00, <1 x bfloat> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V2 = call bfloat @llvm.vector.reduce.fmul.v2bf16(bfloat 0.000000e+00, <2 x bfloat> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V4 = call bfloat @llvm.vector.reduce.fmul.v4bf16(bfloat 0.000000e+00, <4 x bfloat> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %V8 = call bfloat @llvm.vector.reduce.fmul.v8bf16(bfloat 0.000000e+00, <8 x bfloat> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 79 for instruction: %V16 = call bfloat @llvm.vector.reduce.fmul.v16bf16(bfloat 0.000000e+00, <16 x bfloat> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 159 for instruction: %v32 = call bfloat @llvm.vector.reduce.fmul.v32bf16(bfloat 0.000000e+00, <32 x bfloat> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 319 for instruction: %V64 = call bfloat @llvm.vector.reduce.fmul.v64bf16(bfloat 0.000000e+00, <64 x bfloat> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 638 for instruction: %V128 = call bfloat @llvm.vector.reduce.fmul.v128bf16(bfloat 0.000000e+00, <128 x bfloat> undef)
; FP-REDUCE-NEXT: Cost Model: Invalid cost for instruction: %NXV1 = call bfloat @llvm.vector.reduce.fmul.nxv1bf16(bfloat 0.000000e+00, <vscale x 1 x bfloat> undef)
; FP-REDUCE-NEXT: Cost Model: Invalid cost for instruction: %NXV2 = call bfloat @llvm.vector.reduce.fmul.nxv2bf16(bfloat 0.000000e+00, <vscale x 2 x bfloat> undef)
; FP-REDUCE-NEXT: Cost Model: Invalid cost for instruction: %NXV4 = call bfloat @llvm.vector.reduce.fmul.nxv4bf16(bfloat 0.000000e+00, <vscale x 4 x bfloat> undef)
@@ -273,14 +273,14 @@ define void @reduce_ordered_fmul_bfloat() {
define void @reduce_ordered_fmul_half() {
; FP-REDUCE-LABEL: 'reduce_ordered_fmul_half'
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1 = call half @llvm.vector.reduce.fmul.v1f16(half 0.000000e+00, <1 x half> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V2 = call half @llvm.vector.reduce.fmul.v2f16(half 0.000000e+00, <2 x half> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V4 = call half @llvm.vector.reduce.fmul.v4f16(half 0.000000e+00, <4 x half> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V8 = call half @llvm.vector.reduce.fmul.v8f16(half 0.000000e+00, <8 x half> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 63 for instruction: %V16 = call half @llvm.vector.reduce.fmul.v16f16(half 0.000000e+00, <16 x half> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 127 for instruction: %v32 = call half @llvm.vector.reduce.fmul.v32f16(half 0.000000e+00, <32 x half> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 255 for instruction: %V64 = call half @llvm.vector.reduce.fmul.v64f16(half 0.000000e+00, <64 x half> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 510 for instruction: %V128 = call half @llvm.vector.reduce.fmul.v128f16(half 0.000000e+00, <128 x half> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V1 = call half @llvm.vector.reduce.fmul.v1f16(half 0.000000e+00, <1 x half> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V2 = call half @llvm.vector.reduce.fmul.v2f16(half 0.000000e+00, <2 x half> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V4 = call half @llvm.vector.reduce.fmul.v4f16(half 0.000000e+00, <4 x half> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %V8 = call half @llvm.vector.reduce.fmul.v8f16(half 0.000000e+00, <8 x half> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 79 for instruction: %V16 = call half @llvm.vector.reduce.fmul.v16f16(half 0.000000e+00, <16 x half> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 159 for instruction: %v32 = call half @llvm.vector.reduce.fmul.v32f16(half 0.000000e+00, <32 x half> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 319 for instruction: %V64 = call half @llvm.vector.reduce.fmul.v64f16(half 0.000000e+00, <64 x half> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 638 for instruction: %V128 = call half @llvm.vector.reduce.fmul.v128f16(half 0.000000e+00, <128 x half> undef)
; FP-REDUCE-NEXT: Cost Model: Invalid cost for instruction: %NXV1 = call half @llvm.vector.reduce.fmul.nxv1f16(half 0.000000e+00, <vscale x 1 x half> undef)
; FP-REDUCE-NEXT: Cost Model: Invalid cost for instruction: %NXV2 = call half @llvm.vector.reduce.fmul.nxv2f16(half 0.000000e+00, <vscale x 2 x half> undef)
; FP-REDUCE-NEXT: Cost Model: Invalid cost for instruction: %NXV4 = call half @llvm.vector.reduce.fmul.nxv4f16(half 0.000000e+00, <vscale x 4 x half> undef)
@@ -325,14 +325,14 @@ define void @reduce_ordered_fmul_half() {
define void @reduce_ordered_fmul_float() {
; FP-REDUCE-LABEL: 'reduce_ordered_fmul_float'
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1 = call float @llvm.vector.reduce.fmul.v1f32(float 0.000000e+00, <1 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V2 = call float @llvm.vector.reduce.fmul.v2f32(float 0.000000e+00, <2 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V4 = call float @llvm.vector.reduce.fmul.v4f32(float 0.000000e+00, <4 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V8 = call float @llvm.vector.reduce.fmul.v8f32(float 0.000000e+00, <8 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 63 for instruction: %V16 = call float @llvm.vector.reduce.fmul.v16f32(float 0.000000e+00, <16 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 127 for instruction: %v32 = call float @llvm.vector.reduce.fmul.v32f32(float 0.000000e+00, <32 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 254 for instruction: %V64 = call float @llvm.vector.reduce.fmul.v64f32(float 0.000000e+00, <64 x float> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 508 for instruction: %V128 = call float @llvm.vector.reduce.fmul.v128f32(float 0.000000e+00, <128 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V1 = call float @llvm.vector.reduce.fmul.v1f32(float 0.000000e+00, <1 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V2 = call float @llvm.vector.reduce.fmul.v2f32(float 0.000000e+00, <2 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V4 = call float @llvm.vector.reduce.fmul.v4f32(float 0.000000e+00, <4 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %V8 = call float @llvm.vector.reduce.fmul.v8f32(float 0.000000e+00, <8 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 79 for instruction: %V16 = call float @llvm.vector.reduce.fmul.v16f32(float 0.000000e+00, <16 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 159 for instruction: %v32 = call float @llvm.vector.reduce.fmul.v32f32(float 0.000000e+00, <32 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 318 for instruction: %V64 = call float @llvm.vector.reduce.fmul.v64f32(float 0.000000e+00, <64 x float> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 636 for instruction: %V128 = call float @llvm.vector.reduce.fmul.v128f32(float 0.000000e+00, <128 x float> undef)
; FP-REDUCE-NEXT: Cost Model: Invalid cost for instruction: %NXV1 = call float @llvm.vector.reduce.fmul.nxv1f32(float 0.000000e+00, <vscale x 1 x float> undef)
; FP-REDUCE-NEXT: Cost Model: Invalid cost for instruction: %NXV2 = call float @llvm.vector.reduce.fmul.nxv2f32(float 0.000000e+00, <vscale x 2 x float> undef)
; FP-REDUCE-NEXT: Cost Model: Invalid cost for instruction: %NXV4 = call float @llvm.vector.reduce.fmul.nxv4f32(float 0.000000e+00, <vscale x 4 x float> undef)
@@ -374,14 +374,14 @@ define void @reduce_ordered_fmul_float() {
define void @reduce_ordered_fmul_double() {
; FP-REDUCE-LABEL: 'reduce_ordered_fmul_double'
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1 = call double @llvm.vector.reduce.fmul.v1f64(double 0.000000e+00, <1 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V2 = call double @llvm.vector.reduce.fmul.v2f64(double 0.000000e+00, <2 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V4 = call double @llvm.vector.reduce.fmul.v4f64(double 0.000000e+00, <4 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V8 = call double @llvm.vector.reduce.fmul.v8f64(double 0.000000e+00, <8 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 63 for instruction: %V16 = call double @llvm.vector.reduce.fmul.v16f64(double 0.000000e+00, <16 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 126 for instruction: %v32 = call double @llvm.vector.reduce.fmul.v32f64(double 0.000000e+00, <32 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 252 for instruction: %V64 = call double @llvm.vector.reduce.fmul.v64f64(double 0.000000e+00, <64 x double> undef)
-; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 504 for instruction: %V128 = call double @llvm.vector.reduce.fmul.v128f64(double 0.000000e+00, <128 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V1 = call double @llvm.vector.reduce.fmul.v1f64(double 0.000000e+00, <1 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V2 = call double @llvm.vector.reduce.fmul.v2f64(double 0.000000e+00, <2 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V4 = call double @llvm.vector.reduce.fmul.v4f64(double 0.000000e+00, <4 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %V8 = call double @llvm.vector.reduce.fmul.v8f64(double 0.000000e+00, <8 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 79 for instruction: %V16 = call double @llvm.vector.reduce.fmul.v16f64(double 0.000000e+00, <16 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 158 for instruction: %v32 = call double @llvm.vector.reduce.fmul.v32f64(double 0.000000e+00, <32 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 316 for instruction: %V64 = call double @llvm.vector.reduce.fmul.v64f64(double 0.000000e+00, <64 x double> undef)
+; FP-REDUCE-NEXT: Cost Model: Found an estimated cost of 632 for instruction: %V128 = call double @llvm.vector.reduce.fmul.v128f64(double 0.000000e+00, <128 x double> undef)
; FP-REDUCE-NEXT: Cost Model: Invalid cost for instruction: %NXV1 = call double @llvm.vector.reduce.fmul.nxv1f64(double 0.000000e+00, <vscale x 1 x double> undef)
; FP-REDUCE-NEXT: Cost Model: Invalid cost for instruction: %NXV2 = call double @llvm.vector.reduce.fmul.nxv2f64(double 0.000000e+00, <vscale x 2 x double> undef)
; FP-REDUCE-NEXT: Cost Model: Invalid cost for instruction: %NXV4 = call double @llvm.vector.reduce.fmul.nxv4f64(double 0.000000e+00, <vscale x 4 x double> undef)
diff --git a/llvm/test/Analysis/CostModel/RISCV/reduce-max.ll b/llvm/test/Analysis/CostModel/RISCV/reduce-max.ll
index 1370beb880d03..8045e4c9921fc 100644
--- a/llvm/test/Analysis/CostModel/RISCV/reduce-max.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/reduce-max.ll
@@ -40,14 +40,14 @@ define i32 @reduce_umin_i1(i32 %arg) {
define i32 @reduce_umax_i8(i32 %arg) {
; CHECK-LABEL: 'reduce_umax_i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i8 @llvm.vector.reduce.umax.v1i8(<1 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.vector.reduce.umax.v2i8(<2 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.vector.reduce.umax.v4i8(<4 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.vector.reduce.umax.v8i8(<8 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V16 = call i8 @llvm.vector.reduce.umax.v16i8(<16 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i8 @llvm.vector.reduce.umax.v32i8(<32 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V64 = call i8 @llvm.vector.reduce.umax.v64i8(<64 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i8 @llvm.vector.reduce.umax.v128i8(<128 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i8 @llvm.vector.reduce.umax.v1i8(<1 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.vector.reduce.umax.v2i8(<2 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i8 @llvm.vector.reduce.umax.v4i8(<4 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i8 @llvm.vector.reduce.umax.v8i8(<8 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i8 @llvm.vector.reduce.umax.v16i8(<16 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i8 @llvm.vector.reduce.umax.v32i8(<32 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i8 @llvm.vector.reduce.umax.v64i8(<64 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V128 = call i8 @llvm.vector.reduce.umax.v128i8(<128 x i8> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_umax_i8'
@@ -74,14 +74,14 @@ define i32 @reduce_umax_i8(i32 %arg) {
define i32 @reduce_umax_i16(i32 %arg) {
; CHECK-LABEL: 'reduce_umax_i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i16 @llvm.vector.reduce.umax.v1i16(<1 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i16 @llvm.vector.reduce.umax.v2i16(<2 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i16 @llvm.vector.reduce.umax.v4i16(<4 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.vector.reduce.umax.v8i16(<8 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V16 = call i16 @llvm.vector.reduce.umax.v16i16(<16 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i16 @llvm.vector.reduce.umax.v32i16(<32 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V64 = call i16 @llvm.vector.reduce.umax.v64i16(<64 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V128 = call i16 @llvm.vector.reduce.umax.v128i16(<128 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i16 @llvm.vector.reduce.umax.v1i16(<1 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.vector.reduce.umax.v2i16(<2 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.vector.reduce.umax.v4i16(<4 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i16 @llvm.vector.reduce.umax.v8i16(<8 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i16 @llvm.vector.reduce.umax.v16i16(<16 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i16 @llvm.vector.reduce.umax.v32i16(<32 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.vector.reduce.umax.v64i16(<64 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i16 @llvm.vector.reduce.umax.v128i16(<128 x i16> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_umax_i16'
@@ -108,14 +108,14 @@ define i32 @reduce_umax_i16(i32 %arg) {
define i32 @reduce_umax_i32(i32 %arg) {
; CHECK-LABEL: 'reduce_umax_i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i32 @llvm.vector.reduce.umax.v1i32(<1 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i32 @llvm.vector.reduce.umax.v2i32(<2 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i32 @llvm.vector.reduce.umax.v4i32(<4 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i32 @llvm.vector.reduce.umax.v8i32(<8 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V16 = call i32 @llvm.vector.reduce.umax.v16i32(<16 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i32 @llvm.vector.reduce.umax.v32i32(<32 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i32 @llvm.vector.reduce.umax.v64i32(<64 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %V128 = call i32 @llvm.vector.reduce.umax.v128i32(<128 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i32 @llvm.vector.reduce.umax.v1i32(<1 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.vector.reduce.umax.v2i32(<2 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i32 @llvm.vector.reduce.umax.v4i32(<4 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i32 @llvm.vector.reduce.umax.v8i32(<8 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i32 @llvm.vector.reduce.umax.v16i32(<16 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i32 @llvm.vector.reduce.umax.v32i32(<32 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V64 = call i32 @llvm.vector.reduce.umax.v64i32(<64 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V128 = call i32 @llvm.vector.reduce.umax.v128i32(<128 x i32> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_umax_i32'
@@ -142,14 +142,14 @@ define i32 @reduce_umax_i32(i32 %arg) {
define i32 @reduce_umax_i64(i32 %arg) {
; CHECK-LABEL: 'reduce_umax_i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i64 @llvm.vector.reduce.umax.v1i64(<1 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i64 @llvm.vector.reduce.umax.v2i64(<2 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i64 @llvm.vector.reduce.umax.v4i64(<4 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i64 @llvm.vector.reduce.umax.v8i64(<8 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V16 = call i64 @llvm.vector.reduce.umax.v16i64(<16 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %V32 = call i64 @llvm.vector.reduce.umax.v32i64(<32 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 29 for instruction: %V64 = call i64 @llvm.vector.reduce.umax.v64i64(<64 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 61 for instruction: %V128 = call i64 @llvm.vector.reduce.umax.v128i64(<128 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i64 @llvm.vector.reduce.umax.v1i64(<1 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.vector.reduce.umax.v2i64(<2 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.vector.reduce.umax.v4i64(<4 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i64 @llvm.vector.reduce.umax.v8i64(<8 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i64 @llvm.vector.reduce.umax.v16i64(<16 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V32 = call i64 @llvm.vector.reduce.umax.v32i64(<32 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %V64 = call i64 @llvm.vector.reduce.umax.v64i64(<64 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 62 for instruction: %V128 = call i64 @llvm.vector.reduce.umax.v128i64(<128 x i64> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_umax_i64'
@@ -210,14 +210,14 @@ define i32 @reduce_smin_i1(i32 %arg) {
define i32 @reduce_smax_i8(i32 %arg) {
; CHECK-LABEL: 'reduce_smax_i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i8 @llvm.vector.reduce.smax.v1i8(<1 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.vector.reduce.smax.v2i8(<2 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.vector.reduce.smax.v4i8(<4 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.vector.reduce.smax.v8i8(<8 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V16 = call i8 @llvm.vector.reduce.smax.v16i8(<16 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i8 @llvm.vector.reduce.smax.v32i8(<32 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V64 = call i8 @llvm.vector.reduce.smax.v64i8(<64 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i8 @llvm.vector.reduce.smax.v128i8(<128 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i8 @llvm.vector.reduce.smax.v1i8(<1 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.vector.reduce.smax.v2i8(<2 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i8 @llvm.vector.reduce.smax.v4i8(<4 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i8 @llvm.vector.reduce.smax.v8i8(<8 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i8 @llvm.vector.reduce.smax.v16i8(<16 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i8 @llvm.vector.reduce.smax.v32i8(<32 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i8 @llvm.vector.reduce.smax.v64i8(<64 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V128 = call i8 @llvm.vector.reduce.smax.v128i8(<128 x i8> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_smax_i8'
@@ -244,14 +244,14 @@ define i32 @reduce_smax_i8(i32 %arg) {
define i32 @reduce_smax_i16(i32 %arg) {
; CHECK-LABEL: 'reduce_smax_i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i16 @llvm.vector.reduce.smax.v1i16(<1 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i16 @llvm.vector.reduce.smax.v2i16(<2 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i16 @llvm.vector.reduce.smax.v4i16(<4 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.vector.reduce.smax.v8i16(<8 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V16 = call i16 @llvm.vector.reduce.smax.v16i16(<16 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i16 @llvm.vector.reduce.smax.v32i16(<32 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V64 = call i16 @llvm.vector.reduce.smax.v64i16(<64 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V128 = call i16 @llvm.vector.reduce.smax.v128i16(<128 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i16 @llvm.vector.reduce.smax.v1i16(<1 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.vector.reduce.smax.v2i16(<2 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.vector.reduce.smax.v4i16(<4 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i16 @llvm.vector.reduce.smax.v8i16(<8 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i16 @llvm.vector.reduce.smax.v16i16(<16 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i16 @llvm.vector.reduce.smax.v32i16(<32 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.vector.reduce.smax.v64i16(<64 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i16 @llvm.vector.reduce.smax.v128i16(<128 x i16> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_smax_i16'
@@ -278,14 +278,14 @@ define i32 @reduce_smax_i16(i32 %arg) {
define i32 @reduce_smax_i32(i32 %arg) {
; CHECK-LABEL: 'reduce_smax_i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i32 @llvm.vector.reduce.smax.v1i32(<1 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i32 @llvm.vector.reduce.smax.v2i32(<2 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i32 @llvm.vector.reduce.smax.v4i32(<4 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i32 @llvm.vector.reduce.smax.v8i32(<8 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V16 = call i32 @llvm.vector.reduce.smax.v16i32(<16 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i32 @llvm.vector.reduce.smax.v32i32(<32 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i32 @llvm.vector.reduce.smax.v64i32(<64 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %V128 = call i32 @llvm.vector.reduce.smax.v128i32(<128 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i32 @llvm.vector.reduce.smax.v1i32(<1 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.vector.reduce.smax.v2i32(<2 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i32 @llvm.vector.reduce.smax.v4i32(<4 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i32 @llvm.vector.reduce.smax.v8i32(<8 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i32 @llvm.vector.reduce.smax.v16i32(<16 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i32 @llvm.vector.reduce.smax.v32i32(<32 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V64 = call i32 @llvm.vector.reduce.smax.v64i32(<64 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V128 = call i32 @llvm.vector.reduce.smax.v128i32(<128 x i32> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_smax_i32'
@@ -312,14 +312,14 @@ define i32 @reduce_smax_i32(i32 %arg) {
define i32 @reduce_smax_i64(i32 %arg) {
; CHECK-LABEL: 'reduce_smax_i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i64 @llvm.vector.reduce.smax.v1i64(<1 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i64 @llvm.vector.reduce.smax.v2i64(<2 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i64 @llvm.vector.reduce.smax.v4i64(<4 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i64 @llvm.vector.reduce.smax.v8i64(<8 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V16 = call i64 @llvm.vector.reduce.smax.v16i64(<16 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %V32 = call i64 @llvm.vector.reduce.smax.v32i64(<32 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 29 for instruction: %V64 = call i64 @llvm.vector.reduce.smax.v64i64(<64 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 61 for instruction: %V128 = call i64 @llvm.vector.reduce.smax.v128i64(<128 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i64 @llvm.vector.reduce.smax.v1i64(<1 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.vector.reduce.smax.v2i64(<2 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.vector.reduce.smax.v4i64(<4 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i64 @llvm.vector.reduce.smax.v8i64(<8 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i64 @llvm.vector.reduce.smax.v16i64(<16 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V32 = call i64 @llvm.vector.reduce.smax.v32i64(<32 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %V64 = call i64 @llvm.vector.reduce.smax.v64i64(<64 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 62 for instruction: %V128 = call i64 @llvm.vector.reduce.smax.v128i64(<128 x i64> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_smax_i64'
diff --git a/llvm/test/Analysis/CostModel/RISCV/reduce-min.ll b/llvm/test/Analysis/CostModel/RISCV/reduce-min.ll
index 7acf76b53bb2d..d117c6e91419f 100644
--- a/llvm/test/Analysis/CostModel/RISCV/reduce-min.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/reduce-min.ll
@@ -40,14 +40,14 @@ define i32 @reduce_umin_i1(i32 %arg) {
define i32 @reduce_umin_i8(i32 %arg) {
; CHECK-LABEL: 'reduce_umin_i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i8 @llvm.vector.reduce.umin.v1i8(<1 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.vector.reduce.umin.v2i8(<2 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.vector.reduce.umin.v4i8(<4 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.vector.reduce.umin.v8i8(<8 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V16 = call i8 @llvm.vector.reduce.umin.v16i8(<16 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i8 @llvm.vector.reduce.umin.v32i8(<32 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V64 = call i8 @llvm.vector.reduce.umin.v64i8(<64 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i8 @llvm.vector.reduce.umin.v128i8(<128 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i8 @llvm.vector.reduce.umin.v1i8(<1 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.vector.reduce.umin.v2i8(<2 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i8 @llvm.vector.reduce.umin.v4i8(<4 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i8 @llvm.vector.reduce.umin.v8i8(<8 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i8 @llvm.vector.reduce.umin.v16i8(<16 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i8 @llvm.vector.reduce.umin.v32i8(<32 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i8 @llvm.vector.reduce.umin.v64i8(<64 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V128 = call i8 @llvm.vector.reduce.umin.v128i8(<128 x i8> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_umin_i8'
@@ -74,14 +74,14 @@ define i32 @reduce_umin_i8(i32 %arg) {
define i32 @reduce_umin_i16(i32 %arg) {
; CHECK-LABEL: 'reduce_umin_i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i16 @llvm.vector.reduce.umin.v1i16(<1 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i16 @llvm.vector.reduce.umin.v2i16(<2 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i16 @llvm.vector.reduce.umin.v4i16(<4 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.vector.reduce.umin.v8i16(<8 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V16 = call i16 @llvm.vector.reduce.umin.v16i16(<16 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i16 @llvm.vector.reduce.umin.v32i16(<32 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V64 = call i16 @llvm.vector.reduce.umin.v64i16(<64 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V128 = call i16 @llvm.vector.reduce.umin.v128i16(<128 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i16 @llvm.vector.reduce.umin.v1i16(<1 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.vector.reduce.umin.v2i16(<2 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.vector.reduce.umin.v4i16(<4 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i16 @llvm.vector.reduce.umin.v8i16(<8 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i16 @llvm.vector.reduce.umin.v16i16(<16 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i16 @llvm.vector.reduce.umin.v32i16(<32 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.vector.reduce.umin.v64i16(<64 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i16 @llvm.vector.reduce.umin.v128i16(<128 x i16> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_umin_i16'
@@ -108,14 +108,14 @@ define i32 @reduce_umin_i16(i32 %arg) {
define i32 @reduce_umin_i32(i32 %arg) {
; CHECK-LABEL: 'reduce_umin_i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i32 @llvm.vector.reduce.umin.v1i32(<1 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i32 @llvm.vector.reduce.umin.v2i32(<2 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i32 @llvm.vector.reduce.umin.v4i32(<4 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i32 @llvm.vector.reduce.umin.v8i32(<8 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V16 = call i32 @llvm.vector.reduce.umin.v16i32(<16 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i32 @llvm.vector.reduce.umin.v32i32(<32 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i32 @llvm.vector.reduce.umin.v64i32(<64 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %V128 = call i32 @llvm.vector.reduce.umin.v128i32(<128 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i32 @llvm.vector.reduce.umin.v1i32(<1 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.vector.reduce.umin.v2i32(<2 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i32 @llvm.vector.reduce.umin.v4i32(<4 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i32 @llvm.vector.reduce.umin.v8i32(<8 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i32 @llvm.vector.reduce.umin.v16i32(<16 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i32 @llvm.vector.reduce.umin.v32i32(<32 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V64 = call i32 @llvm.vector.reduce.umin.v64i32(<64 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V128 = call i32 @llvm.vector.reduce.umin.v128i32(<128 x i32> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_umin_i32'
@@ -142,14 +142,14 @@ define i32 @reduce_umin_i32(i32 %arg) {
define i32 @reduce_umin_i64(i32 %arg) {
; CHECK-LABEL: 'reduce_umin_i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i64 @llvm.vector.reduce.umin.v1i64(<1 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i64 @llvm.vector.reduce.umin.v2i64(<2 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i64 @llvm.vector.reduce.umin.v4i64(<4 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i64 @llvm.vector.reduce.umin.v8i64(<8 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V16 = call i64 @llvm.vector.reduce.umin.v16i64(<16 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %V32 = call i64 @llvm.vector.reduce.umin.v32i64(<32 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 29 for instruction: %V64 = call i64 @llvm.vector.reduce.umin.v64i64(<64 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 61 for instruction: %V128 = call i64 @llvm.vector.reduce.umin.v128i64(<128 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i64 @llvm.vector.reduce.umin.v1i64(<1 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.vector.reduce.umin.v2i64(<2 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.vector.reduce.umin.v4i64(<4 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i64 @llvm.vector.reduce.umin.v8i64(<8 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i64 @llvm.vector.reduce.umin.v16i64(<16 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V32 = call i64 @llvm.vector.reduce.umin.v32i64(<32 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %V64 = call i64 @llvm.vector.reduce.umin.v64i64(<64 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 62 for instruction: %V128 = call i64 @llvm.vector.reduce.umin.v128i64(<128 x i64> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_umin_i64'
@@ -210,14 +210,14 @@ define i32 @reduce_smin_i1(i32 %arg) {
define i32 @reduce_smin_i8(i32 %arg) {
; CHECK-LABEL: 'reduce_smin_i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i8 @llvm.vector.reduce.smin.v1i8(<1 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.vector.reduce.smin.v2i8(<2 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.vector.reduce.smin.v4i8(<4 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.vector.reduce.smin.v8i8(<8 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V16 = call i8 @llvm.vector.reduce.smin.v16i8(<16 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i8 @llvm.vector.reduce.smin.v32i8(<32 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V64 = call i8 @llvm.vector.reduce.smin.v64i8(<64 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i8 @llvm.vector.reduce.smin.v128i8(<128 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i8 @llvm.vector.reduce.smin.v1i8(<1 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.vector.reduce.smin.v2i8(<2 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i8 @llvm.vector.reduce.smin.v4i8(<4 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i8 @llvm.vector.reduce.smin.v8i8(<8 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i8 @llvm.vector.reduce.smin.v16i8(<16 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i8 @llvm.vector.reduce.smin.v32i8(<32 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i8 @llvm.vector.reduce.smin.v64i8(<64 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V128 = call i8 @llvm.vector.reduce.smin.v128i8(<128 x i8> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_smin_i8'
@@ -244,14 +244,14 @@ define i32 @reduce_smin_i8(i32 %arg) {
define i32 @reduce_smin_i16(i32 %arg) {
; CHECK-LABEL: 'reduce_smin_i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i16 @llvm.vector.reduce.smin.v1i16(<1 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i16 @llvm.vector.reduce.smin.v2i16(<2 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i16 @llvm.vector.reduce.smin.v4i16(<4 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.vector.reduce.smin.v8i16(<8 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V16 = call i16 @llvm.vector.reduce.smin.v16i16(<16 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i16 @llvm.vector.reduce.smin.v32i16(<32 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V64 = call i16 @llvm.vector.reduce.smin.v64i16(<64 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V128 = call i16 @llvm.vector.reduce.smin.v128i16(<128 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i16 @llvm.vector.reduce.smin.v1i16(<1 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.vector.reduce.smin.v2i16(<2 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.vector.reduce.smin.v4i16(<4 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i16 @llvm.vector.reduce.smin.v8i16(<8 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i16 @llvm.vector.reduce.smin.v16i16(<16 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i16 @llvm.vector.reduce.smin.v32i16(<32 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.vector.reduce.smin.v64i16(<64 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i16 @llvm.vector.reduce.smin.v128i16(<128 x i16> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_smin_i16'
@@ -278,14 +278,14 @@ define i32 @reduce_smin_i16(i32 %arg) {
define i32 @reduce_smin_i32(i32 %arg) {
; CHECK-LABEL: 'reduce_smin_i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i32 @llvm.vector.reduce.smin.v1i32(<1 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i32 @llvm.vector.reduce.smin.v2i32(<2 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i32 @llvm.vector.reduce.smin.v4i32(<4 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i32 @llvm.vector.reduce.smin.v8i32(<8 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V16 = call i32 @llvm.vector.reduce.smin.v16i32(<16 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i32 @llvm.vector.reduce.smin.v32i32(<32 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i32 @llvm.vector.reduce.smin.v64i32(<64 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %V128 = call i32 @llvm.vector.reduce.smin.v128i32(<128 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i32 @llvm.vector.reduce.smin.v1i32(<1 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.vector.reduce.smin.v2i32(<2 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i32 @llvm.vector.reduce.smin.v4i32(<4 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i32 @llvm.vector.reduce.smin.v8i32(<8 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i32 @llvm.vector.reduce.smin.v16i32(<16 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i32 @llvm.vector.reduce.smin.v32i32(<32 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V64 = call i32 @llvm.vector.reduce.smin.v64i32(<64 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V128 = call i32 @llvm.vector.reduce.smin.v128i32(<128 x i32> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_smin_i32'
@@ -312,14 +312,14 @@ define i32 @reduce_smin_i32(i32 %arg) {
define i32 @reduce_smin_i64(i32 %arg) {
; CHECK-LABEL: 'reduce_smin_i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i64 @llvm.vector.reduce.smin.v1i64(<1 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i64 @llvm.vector.reduce.smin.v2i64(<2 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i64 @llvm.vector.reduce.smin.v4i64(<4 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i64 @llvm.vector.reduce.smin.v8i64(<8 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V16 = call i64 @llvm.vector.reduce.smin.v16i64(<16 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %V32 = call i64 @llvm.vector.reduce.smin.v32i64(<32 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 29 for instruction: %V64 = call i64 @llvm.vector.reduce.smin.v64i64(<64 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 61 for instruction: %V128 = call i64 @llvm.vector.reduce.smin.v128i64(<128 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i64 @llvm.vector.reduce.smin.v1i64(<1 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.vector.reduce.smin.v2i64(<2 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.vector.reduce.smin.v4i64(<4 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i64 @llvm.vector.reduce.smin.v8i64(<8 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i64 @llvm.vector.reduce.smin.v16i64(<16 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V32 = call i64 @llvm.vector.reduce.smin.v32i64(<32 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %V64 = call i64 @llvm.vector.reduce.smin.v64i64(<64 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 62 for instruction: %V128 = call i64 @llvm.vector.reduce.smin.v128i64(<128 x i64> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_smin_i64'
diff --git a/llvm/test/Analysis/CostModel/RISCV/reduce-or.ll b/llvm/test/Analysis/CostModel/RISCV/reduce-or.ll
index db7d4ddf4c5ae..029b0c2828097 100644
--- a/llvm/test/Analysis/CostModel/RISCV/reduce-or.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/reduce-or.ll
@@ -49,14 +49,14 @@ define i32 @reduce_i1(i32 %arg) {
define i32 @reduce_i8(i32 %arg) {
; CHECK-LABEL: 'reduce_i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i8 @llvm.vector.reduce.or.v1i8(<1 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.vector.reduce.or.v2i8(<2 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i8 @llvm.vector.reduce.or.v4i8(<4 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i8 @llvm.vector.reduce.or.v8i8(<8 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V16 = call i8 @llvm.vector.reduce.or.v16i8(<16 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i8 @llvm.vector.reduce.or.v32i8(<32 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V64 = call i8 @llvm.vector.reduce.or.v64i8(<64 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V128 = call i8 @llvm.vector.reduce.or.v128i8(<128 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i8 @llvm.vector.reduce.or.v1i8(<1 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.vector.reduce.or.v2i8(<2 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i8 @llvm.vector.reduce.or.v4i8(<4 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i8 @llvm.vector.reduce.or.v8i8(<8 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i8 @llvm.vector.reduce.or.v16i8(<16 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i8 @llvm.vector.reduce.or.v32i8(<32 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i8 @llvm.vector.reduce.or.v64i8(<64 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V128 = call i8 @llvm.vector.reduce.or.v128i8(<128 x i8> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_i8'
@@ -83,14 +83,14 @@ define i32 @reduce_i8(i32 %arg) {
define i32 @reduce_i16(i32 %arg) {
; CHECK-LABEL: 'reduce_i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i16 @llvm.vector.reduce.or.v1i16(<1 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i16 @llvm.vector.reduce.or.v2i16(<2 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i16 @llvm.vector.reduce.or.v4i16(<4 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.vector.reduce.or.v8i16(<8 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V16 = call i16 @llvm.vector.reduce.or.v16i16(<16 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i16 @llvm.vector.reduce.or.v32i16(<32 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V64 = call i16 @llvm.vector.reduce.or.v64i16(<64 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V128 = call i16 @llvm.vector.reduce.or.v128i16(<128 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i16 @llvm.vector.reduce.or.v1i16(<1 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.vector.reduce.or.v2i16(<2 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.vector.reduce.or.v4i16(<4 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i16 @llvm.vector.reduce.or.v8i16(<8 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i16 @llvm.vector.reduce.or.v16i16(<16 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i16 @llvm.vector.reduce.or.v32i16(<32 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.vector.reduce.or.v64i16(<64 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i16 @llvm.vector.reduce.or.v128i16(<128 x i16> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_i16'
@@ -117,14 +117,14 @@ define i32 @reduce_i16(i32 %arg) {
define i32 @reduce_i32(i32 %arg) {
; CHECK-LABEL: 'reduce_i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i32 @llvm.vector.reduce.or.v1i32(<1 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i32 @llvm.vector.reduce.or.v2i32(<2 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i32 @llvm.vector.reduce.or.v4i32(<4 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i32 @llvm.vector.reduce.or.v8i32(<8 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V16 = call i32 @llvm.vector.reduce.or.v16i32(<16 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i32 @llvm.vector.reduce.or.v32i32(<32 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V64 = call i32 @llvm.vector.reduce.or.v64i32(<64 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %V128 = call i32 @llvm.vector.reduce.or.v128i32(<128 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i32 @llvm.vector.reduce.or.v1i32(<1 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.vector.reduce.or.v2i32(<2 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i32 @llvm.vector.reduce.or.v4i32(<4 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i32 @llvm.vector.reduce.or.v8i32(<8 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i32 @llvm.vector.reduce.or.v16i32(<16 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i32 @llvm.vector.reduce.or.v32i32(<32 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V64 = call i32 @llvm.vector.reduce.or.v64i32(<64 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V128 = call i32 @llvm.vector.reduce.or.v128i32(<128 x i32> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_i32'
@@ -151,14 +151,14 @@ define i32 @reduce_i32(i32 %arg) {
define i32 @reduce_i64(i32 %arg) {
; CHECK-LABEL: 'reduce_i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V1 = call i64 @llvm.vector.reduce.or.v1i64(<1 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i64 @llvm.vector.reduce.or.v2i64(<2 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i64 @llvm.vector.reduce.or.v4i64(<4 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i64 @llvm.vector.reduce.or.v8i64(<8 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V16 = call i64 @llvm.vector.reduce.or.v16i64(<16 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %V32 = call i64 @llvm.vector.reduce.or.v32i64(<32 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 29 for instruction: %V64 = call i64 @llvm.vector.reduce.or.v64i64(<64 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 61 for instruction: %V128 = call i64 @llvm.vector.reduce.or.v128i64(<128 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i64 @llvm.vector.reduce.or.v1i64(<1 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.vector.reduce.or.v2i64(<2 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.vector.reduce.or.v4i64(<4 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i64 @llvm.vector.reduce.or.v8i64(<8 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i64 @llvm.vector.reduce.or.v16i64(<16 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V32 = call i64 @llvm.vector.reduce.or.v32i64(<32 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %V64 = call i64 @llvm.vector.reduce.or.v64i64(<64 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 62 for instruction: %V128 = call i64 @llvm.vector.reduce.or.v128i64(<128 x i64> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_i64'
diff --git a/llvm/test/Analysis/CostModel/RISCV/reduce-scalable-fp.ll b/llvm/test/Analysis/CostModel/RISCV/reduce-scalable-fp.ll
index 6f631b2a0aaa9..e14dabdc44852 100644
--- a/llvm/test/Analysis/CostModel/RISCV/reduce-scalable-fp.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/reduce-scalable-fp.ll
@@ -6,7 +6,7 @@
define half @vreduce_fadd_nxv1f16(<vscale x 1 x half> %v, half %s) {
; CHECK-LABEL: 'vreduce_fadd_nxv1f16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call reassoc half @llvm.vector.reduce.fadd.nxv1f16(half %s, <vscale x 1 x half> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call reassoc half @llvm.vector.reduce.fadd.nxv1f16(half %s, <vscale x 1 x half> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret half %red
;
; SIZE-LABEL: 'vreduce_fadd_nxv1f16'
@@ -19,7 +19,7 @@ define half @vreduce_fadd_nxv1f16(<vscale x 1 x half> %v, half %s) {
define half @vreduce_ord_fadd_nxv1f16(<vscale x 1 x half> %v, half %s) {
; CHECK-LABEL: 'vreduce_ord_fadd_nxv1f16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call half @llvm.vector.reduce.fadd.nxv1f16(half %s, <vscale x 1 x half> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call half @llvm.vector.reduce.fadd.nxv1f16(half %s, <vscale x 1 x half> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret half %red
;
; SIZE-LABEL: 'vreduce_ord_fadd_nxv1f16'
@@ -32,7 +32,7 @@ define half @vreduce_ord_fadd_nxv1f16(<vscale x 1 x half> %v, half %s) {
define half @vreduce_fadd_nxv2f16(<vscale x 2 x half> %v, half %s) {
; CHECK-LABEL: 'vreduce_fadd_nxv2f16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call reassoc half @llvm.vector.reduce.fadd.nxv2f16(half %s, <vscale x 2 x half> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call reassoc half @llvm.vector.reduce.fadd.nxv2f16(half %s, <vscale x 2 x half> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret half %red
;
; SIZE-LABEL: 'vreduce_fadd_nxv2f16'
@@ -45,7 +45,7 @@ define half @vreduce_fadd_nxv2f16(<vscale x 2 x half> %v, half %s) {
define half @vreduce_ord_fadd_nxv2f16(<vscale x 2 x half> %v, half %s) {
; CHECK-LABEL: 'vreduce_ord_fadd_nxv2f16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %red = call half @llvm.vector.reduce.fadd.nxv2f16(half %s, <vscale x 2 x half> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %red = call half @llvm.vector.reduce.fadd.nxv2f16(half %s, <vscale x 2 x half> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret half %red
;
; SIZE-LABEL: 'vreduce_ord_fadd_nxv2f16'
@@ -58,7 +58,7 @@ define half @vreduce_ord_fadd_nxv2f16(<vscale x 2 x half> %v, half %s) {
define half @vreduce_fadd_nxv4f16(<vscale x 4 x half> %v, half %s) {
; CHECK-LABEL: 'vreduce_fadd_nxv4f16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call reassoc half @llvm.vector.reduce.fadd.nxv4f16(half %s, <vscale x 4 x half> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %red = call reassoc half @llvm.vector.reduce.fadd.nxv4f16(half %s, <vscale x 4 x half> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret half %red
;
; SIZE-LABEL: 'vreduce_fadd_nxv4f16'
@@ -71,7 +71,7 @@ define half @vreduce_fadd_nxv4f16(<vscale x 4 x half> %v, half %s) {
define half @vreduce_ord_fadd_nxv4f16(<vscale x 4 x half> %v, half %s) {
; CHECK-LABEL: 'vreduce_ord_fadd_nxv4f16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %red = call half @llvm.vector.reduce.fadd.nxv4f16(half %s, <vscale x 4 x half> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %red = call half @llvm.vector.reduce.fadd.nxv4f16(half %s, <vscale x 4 x half> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret half %red
;
; SIZE-LABEL: 'vreduce_ord_fadd_nxv4f16'
@@ -84,7 +84,7 @@ define half @vreduce_ord_fadd_nxv4f16(<vscale x 4 x half> %v, half %s) {
define float @vreduce_fadd_nxv1f32(<vscale x 1 x float> %v, float %s) {
; CHECK-LABEL: 'vreduce_fadd_nxv1f32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call reassoc float @llvm.vector.reduce.fadd.nxv1f32(float %s, <vscale x 1 x float> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call reassoc float @llvm.vector.reduce.fadd.nxv1f32(float %s, <vscale x 1 x float> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %red
;
; SIZE-LABEL: 'vreduce_fadd_nxv1f32'
@@ -97,7 +97,7 @@ define float @vreduce_fadd_nxv1f32(<vscale x 1 x float> %v, float %s) {
define float @vreduce_ord_fadd_nxv1f32(<vscale x 1 x float> %v, float %s) {
; CHECK-LABEL: 'vreduce_ord_fadd_nxv1f32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call float @llvm.vector.reduce.fadd.nxv1f32(float %s, <vscale x 1 x float> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call float @llvm.vector.reduce.fadd.nxv1f32(float %s, <vscale x 1 x float> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %red
;
; SIZE-LABEL: 'vreduce_ord_fadd_nxv1f32'
@@ -111,7 +111,7 @@ define float @vreduce_ord_fadd_nxv1f32(<vscale x 1 x float> %v, float %s) {
define float @vreduce_fwadd_nxv1f32(<vscale x 1 x half> %v, float %s) {
; CHECK-LABEL: 'vreduce_fwadd_nxv1f32'
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %e = fpext <vscale x 1 x half> %v to <vscale x 1 x float>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call reassoc float @llvm.vector.reduce.fadd.nxv1f32(float %s, <vscale x 1 x float> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call reassoc float @llvm.vector.reduce.fadd.nxv1f32(float %s, <vscale x 1 x float> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %red
;
; SIZE-LABEL: 'vreduce_fwadd_nxv1f32'
@@ -127,7 +127,7 @@ define float @vreduce_fwadd_nxv1f32(<vscale x 1 x half> %v, float %s) {
define float @vreduce_ord_fwadd_nxv1f32(<vscale x 1 x half> %v, float %s) {
; CHECK-LABEL: 'vreduce_ord_fwadd_nxv1f32'
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %e = fpext <vscale x 1 x half> %v to <vscale x 1 x float>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call float @llvm.vector.reduce.fadd.nxv1f32(float %s, <vscale x 1 x float> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call float @llvm.vector.reduce.fadd.nxv1f32(float %s, <vscale x 1 x float> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %red
;
; SIZE-LABEL: 'vreduce_ord_fwadd_nxv1f32'
@@ -142,7 +142,7 @@ define float @vreduce_ord_fwadd_nxv1f32(<vscale x 1 x half> %v, float %s) {
define float @vreduce_fadd_nxv2f32(<vscale x 2 x float> %v, float %s) {
; CHECK-LABEL: 'vreduce_fadd_nxv2f32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call reassoc float @llvm.vector.reduce.fadd.nxv2f32(float %s, <vscale x 2 x float> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call reassoc float @llvm.vector.reduce.fadd.nxv2f32(float %s, <vscale x 2 x float> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %red
;
; SIZE-LABEL: 'vreduce_fadd_nxv2f32'
@@ -155,7 +155,7 @@ define float @vreduce_fadd_nxv2f32(<vscale x 2 x float> %v, float %s) {
define float @vreduce_ord_fadd_nxv2f32(<vscale x 2 x float> %v, float %s) {
; CHECK-LABEL: 'vreduce_ord_fadd_nxv2f32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %red = call float @llvm.vector.reduce.fadd.nxv2f32(float %s, <vscale x 2 x float> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %red = call float @llvm.vector.reduce.fadd.nxv2f32(float %s, <vscale x 2 x float> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %red
;
; SIZE-LABEL: 'vreduce_ord_fadd_nxv2f32'
@@ -169,7 +169,7 @@ define float @vreduce_ord_fadd_nxv2f32(<vscale x 2 x float> %v, float %s) {
define float @vreduce_fwadd_nxv2f32(<vscale x 2 x half> %v, float %s) {
; CHECK-LABEL: 'vreduce_fwadd_nxv2f32'
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %e = fpext <vscale x 2 x half> %v to <vscale x 2 x float>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call reassoc float @llvm.vector.reduce.fadd.nxv2f32(float %s, <vscale x 2 x float> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call reassoc float @llvm.vector.reduce.fadd.nxv2f32(float %s, <vscale x 2 x float> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %red
;
; SIZE-LABEL: 'vreduce_fwadd_nxv2f32'
@@ -185,7 +185,7 @@ define float @vreduce_fwadd_nxv2f32(<vscale x 2 x half> %v, float %s) {
define float @vreduce_ord_fwadd_nxv2f32(<vscale x 2 x half> %v, float %s) {
; CHECK-LABEL: 'vreduce_ord_fwadd_nxv2f32'
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %e = fpext <vscale x 2 x half> %v to <vscale x 2 x float>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %red = call float @llvm.vector.reduce.fadd.nxv2f32(float %s, <vscale x 2 x float> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %red = call float @llvm.vector.reduce.fadd.nxv2f32(float %s, <vscale x 2 x float> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %red
;
; SIZE-LABEL: 'vreduce_ord_fwadd_nxv2f32'
@@ -200,7 +200,7 @@ define float @vreduce_ord_fwadd_nxv2f32(<vscale x 2 x half> %v, float %s) {
define float @vreduce_fadd_nxv4f32(<vscale x 4 x float> %v, float %s) {
; CHECK-LABEL: 'vreduce_fadd_nxv4f32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call reassoc float @llvm.vector.reduce.fadd.nxv4f32(float %s, <vscale x 4 x float> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %red = call reassoc float @llvm.vector.reduce.fadd.nxv4f32(float %s, <vscale x 4 x float> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %red
;
; SIZE-LABEL: 'vreduce_fadd_nxv4f32'
@@ -213,7 +213,7 @@ define float @vreduce_fadd_nxv4f32(<vscale x 4 x float> %v, float %s) {
define float @vreduce_ord_fadd_nxv4f32(<vscale x 4 x float> %v, float %s) {
; CHECK-LABEL: 'vreduce_ord_fadd_nxv4f32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %red = call float @llvm.vector.reduce.fadd.nxv4f32(float %s, <vscale x 4 x float> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %red = call float @llvm.vector.reduce.fadd.nxv4f32(float %s, <vscale x 4 x float> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %red
;
; SIZE-LABEL: 'vreduce_ord_fadd_nxv4f32'
@@ -227,7 +227,7 @@ define float @vreduce_ord_fadd_nxv4f32(<vscale x 4 x float> %v, float %s) {
define float @vreduce_fwadd_nxv4f32(<vscale x 4 x half> %v, float %s) {
; CHECK-LABEL: 'vreduce_fwadd_nxv4f32'
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %e = fpext <vscale x 4 x half> %v to <vscale x 4 x float>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call reassoc float @llvm.vector.reduce.fadd.nxv4f32(float %s, <vscale x 4 x float> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %red = call reassoc float @llvm.vector.reduce.fadd.nxv4f32(float %s, <vscale x 4 x float> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %red
;
; SIZE-LABEL: 'vreduce_fwadd_nxv4f32'
@@ -243,7 +243,7 @@ define float @vreduce_fwadd_nxv4f32(<vscale x 4 x half> %v, float %s) {
define float @vreduce_ord_fwadd_nxv4f32(<vscale x 4 x half> %v, float %s) {
; CHECK-LABEL: 'vreduce_ord_fwadd_nxv4f32'
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %e = fpext <vscale x 4 x half> %v to <vscale x 4 x float>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %red = call float @llvm.vector.reduce.fadd.nxv4f32(float %s, <vscale x 4 x float> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %red = call float @llvm.vector.reduce.fadd.nxv4f32(float %s, <vscale x 4 x float> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %red
;
; SIZE-LABEL: 'vreduce_ord_fwadd_nxv4f32'
@@ -258,7 +258,7 @@ define float @vreduce_ord_fwadd_nxv4f32(<vscale x 4 x half> %v, float %s) {
define double @vreduce_fadd_nxv1f64(<vscale x 1 x double> %v, double %s) {
; CHECK-LABEL: 'vreduce_fadd_nxv1f64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call reassoc double @llvm.vector.reduce.fadd.nxv1f64(double %s, <vscale x 1 x double> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call reassoc double @llvm.vector.reduce.fadd.nxv1f64(double %s, <vscale x 1 x double> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %red
;
; SIZE-LABEL: 'vreduce_fadd_nxv1f64'
@@ -271,7 +271,7 @@ define double @vreduce_fadd_nxv1f64(<vscale x 1 x double> %v, double %s) {
define double @vreduce_ord_fadd_nxv1f64(<vscale x 1 x double> %v, double %s) {
; CHECK-LABEL: 'vreduce_ord_fadd_nxv1f64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call double @llvm.vector.reduce.fadd.nxv1f64(double %s, <vscale x 1 x double> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call double @llvm.vector.reduce.fadd.nxv1f64(double %s, <vscale x 1 x double> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %red
;
; SIZE-LABEL: 'vreduce_ord_fadd_nxv1f64'
@@ -285,7 +285,7 @@ define double @vreduce_ord_fadd_nxv1f64(<vscale x 1 x double> %v, double %s) {
define double @vreduce_fwadd_nxv1f64(<vscale x 1 x float> %v, double %s) {
; CHECK-LABEL: 'vreduce_fwadd_nxv1f64'
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %e = fpext <vscale x 1 x float> %v to <vscale x 1 x double>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call reassoc double @llvm.vector.reduce.fadd.nxv1f64(double %s, <vscale x 1 x double> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call reassoc double @llvm.vector.reduce.fadd.nxv1f64(double %s, <vscale x 1 x double> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %red
;
; SIZE-LABEL: 'vreduce_fwadd_nxv1f64'
@@ -301,7 +301,7 @@ define double @vreduce_fwadd_nxv1f64(<vscale x 1 x float> %v, double %s) {
define double @vreduce_ord_fwadd_nxv1f64(<vscale x 1 x float> %v, double %s) {
; CHECK-LABEL: 'vreduce_ord_fwadd_nxv1f64'
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %e = fpext <vscale x 1 x float> %v to <vscale x 1 x double>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call double @llvm.vector.reduce.fadd.nxv1f64(double %s, <vscale x 1 x double> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call double @llvm.vector.reduce.fadd.nxv1f64(double %s, <vscale x 1 x double> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %red
;
; SIZE-LABEL: 'vreduce_ord_fwadd_nxv1f64'
@@ -316,7 +316,7 @@ define double @vreduce_ord_fwadd_nxv1f64(<vscale x 1 x float> %v, double %s) {
define double @vreduce_fadd_nxv2f64(<vscale x 2 x double> %v, double %s) {
; CHECK-LABEL: 'vreduce_fadd_nxv2f64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call reassoc double @llvm.vector.reduce.fadd.nxv2f64(double %s, <vscale x 2 x double> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call reassoc double @llvm.vector.reduce.fadd.nxv2f64(double %s, <vscale x 2 x double> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %red
;
; SIZE-LABEL: 'vreduce_fadd_nxv2f64'
@@ -329,7 +329,7 @@ define double @vreduce_fadd_nxv2f64(<vscale x 2 x double> %v, double %s) {
define double @vreduce_ord_fadd_nxv2f64(<vscale x 2 x double> %v, double %s) {
; CHECK-LABEL: 'vreduce_ord_fadd_nxv2f64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %red = call double @llvm.vector.reduce.fadd.nxv2f64(double %s, <vscale x 2 x double> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %red = call double @llvm.vector.reduce.fadd.nxv2f64(double %s, <vscale x 2 x double> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %red
;
; SIZE-LABEL: 'vreduce_ord_fadd_nxv2f64'
@@ -343,7 +343,7 @@ define double @vreduce_ord_fadd_nxv2f64(<vscale x 2 x double> %v, double %s) {
define double @vreduce_fwadd_nxv2f64(<vscale x 2 x float> %v, double %s) {
; CHECK-LABEL: 'vreduce_fwadd_nxv2f64'
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %e = fpext <vscale x 2 x float> %v to <vscale x 2 x double>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call reassoc double @llvm.vector.reduce.fadd.nxv2f64(double %s, <vscale x 2 x double> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call reassoc double @llvm.vector.reduce.fadd.nxv2f64(double %s, <vscale x 2 x double> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %red
;
; SIZE-LABEL: 'vreduce_fwadd_nxv2f64'
@@ -359,7 +359,7 @@ define double @vreduce_fwadd_nxv2f64(<vscale x 2 x float> %v, double %s) {
define double @vreduce_ord_fwadd_nxv2f64(<vscale x 2 x float> %v, double %s) {
; CHECK-LABEL: 'vreduce_ord_fwadd_nxv2f64'
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %e = fpext <vscale x 2 x float> %v to <vscale x 2 x double>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %red = call double @llvm.vector.reduce.fadd.nxv2f64(double %s, <vscale x 2 x double> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %red = call double @llvm.vector.reduce.fadd.nxv2f64(double %s, <vscale x 2 x double> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %red
;
; SIZE-LABEL: 'vreduce_ord_fwadd_nxv2f64'
@@ -374,7 +374,7 @@ define double @vreduce_ord_fwadd_nxv2f64(<vscale x 2 x float> %v, double %s) {
define double @vreduce_fadd_nxv4f64(<vscale x 4 x double> %v, double %s) {
; CHECK-LABEL: 'vreduce_fadd_nxv4f64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call reassoc double @llvm.vector.reduce.fadd.nxv4f64(double %s, <vscale x 4 x double> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %red = call reassoc double @llvm.vector.reduce.fadd.nxv4f64(double %s, <vscale x 4 x double> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %red
;
; SIZE-LABEL: 'vreduce_fadd_nxv4f64'
@@ -387,7 +387,7 @@ define double @vreduce_fadd_nxv4f64(<vscale x 4 x double> %v, double %s) {
define double @vreduce_ord_fadd_nxv4f64(<vscale x 4 x double> %v, double %s) {
; CHECK-LABEL: 'vreduce_ord_fadd_nxv4f64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %red = call double @llvm.vector.reduce.fadd.nxv4f64(double %s, <vscale x 4 x double> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %red = call double @llvm.vector.reduce.fadd.nxv4f64(double %s, <vscale x 4 x double> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %red
;
; SIZE-LABEL: 'vreduce_ord_fadd_nxv4f64'
@@ -401,7 +401,7 @@ define double @vreduce_ord_fadd_nxv4f64(<vscale x 4 x double> %v, double %s) {
define double @vreduce_fwadd_nxv4f64(<vscale x 4 x float> %v, double %s) {
; CHECK-LABEL: 'vreduce_fwadd_nxv4f64'
; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %e = fpext <vscale x 4 x float> %v to <vscale x 4 x double>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call reassoc double @llvm.vector.reduce.fadd.nxv4f64(double %s, <vscale x 4 x double> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %red = call reassoc double @llvm.vector.reduce.fadd.nxv4f64(double %s, <vscale x 4 x double> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %red
;
; SIZE-LABEL: 'vreduce_fwadd_nxv4f64'
@@ -417,7 +417,7 @@ define double @vreduce_fwadd_nxv4f64(<vscale x 4 x float> %v, double %s) {
define double @vreduce_ord_fwadd_nxv4f64(<vscale x 4 x float> %v, double %s) {
; CHECK-LABEL: 'vreduce_ord_fwadd_nxv4f64'
; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %e = fpext <vscale x 4 x float> %v to <vscale x 4 x double>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %red = call double @llvm.vector.reduce.fadd.nxv4f64(double %s, <vscale x 4 x double> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %red = call double @llvm.vector.reduce.fadd.nxv4f64(double %s, <vscale x 4 x double> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %red
;
; SIZE-LABEL: 'vreduce_ord_fwadd_nxv4f64'
@@ -432,7 +432,7 @@ define double @vreduce_ord_fwadd_nxv4f64(<vscale x 4 x float> %v, double %s) {
define half @vreduce_fmin_nxv1f16(<vscale x 1 x half> %v) {
; CHECK-LABEL: 'vreduce_fmin_nxv1f16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call half @llvm.vector.reduce.fmin.nxv1f16(<vscale x 1 x half> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call half @llvm.vector.reduce.fmin.nxv1f16(<vscale x 1 x half> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret half %red
;
; SIZE-LABEL: 'vreduce_fmin_nxv1f16'
@@ -445,7 +445,7 @@ define half @vreduce_fmin_nxv1f16(<vscale x 1 x half> %v) {
define half @vreduce_fmin_nxv1f16_nonans(<vscale x 1 x half> %v) {
; CHECK-LABEL: 'vreduce_fmin_nxv1f16_nonans'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call nnan half @llvm.vector.reduce.fmin.nxv1f16(<vscale x 1 x half> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call nnan half @llvm.vector.reduce.fmin.nxv1f16(<vscale x 1 x half> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret half %red
;
; SIZE-LABEL: 'vreduce_fmin_nxv1f16_nonans'
@@ -458,7 +458,7 @@ define half @vreduce_fmin_nxv1f16_nonans(<vscale x 1 x half> %v) {
define half @vreduce_fmin_nxv1f16_nonans_noinfs(<vscale x 1 x half> %v) #1 {
; CHECK-LABEL: 'vreduce_fmin_nxv1f16_nonans_noinfs'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call nnan ninf half @llvm.vector.reduce.fmin.nxv1f16(<vscale x 1 x half> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call nnan ninf half @llvm.vector.reduce.fmin.nxv1f16(<vscale x 1 x half> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret half %red
;
; SIZE-LABEL: 'vreduce_fmin_nxv1f16_nonans_noinfs'
@@ -471,7 +471,7 @@ define half @vreduce_fmin_nxv1f16_nonans_noinfs(<vscale x 1 x half> %v) #1 {
define half @vreduce_fmin_nxv2f16(<vscale x 2 x half> %v) {
; CHECK-LABEL: 'vreduce_fmin_nxv2f16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call half @llvm.vector.reduce.fmin.nxv2f16(<vscale x 2 x half> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call half @llvm.vector.reduce.fmin.nxv2f16(<vscale x 2 x half> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret half %red
;
; SIZE-LABEL: 'vreduce_fmin_nxv2f16'
@@ -484,7 +484,7 @@ define half @vreduce_fmin_nxv2f16(<vscale x 2 x half> %v) {
define half @vreduce_fmin_nxv4f16(<vscale x 4 x half> %v) {
; CHECK-LABEL: 'vreduce_fmin_nxv4f16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call half @llvm.vector.reduce.fmin.nxv4f16(<vscale x 4 x half> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call half @llvm.vector.reduce.fmin.nxv4f16(<vscale x 4 x half> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret half %red
;
; SIZE-LABEL: 'vreduce_fmin_nxv4f16'
@@ -497,7 +497,7 @@ define half @vreduce_fmin_nxv4f16(<vscale x 4 x half> %v) {
define half @vreduce_fmin_nxv64f16(<vscale x 64 x half> %v) {
; CHECK-LABEL: 'vreduce_fmin_nxv64f16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %red = call half @llvm.vector.reduce.fmin.nxv64f16(<vscale x 64 x half> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %red = call half @llvm.vector.reduce.fmin.nxv64f16(<vscale x 64 x half> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret half %red
;
; SIZE-LABEL: 'vreduce_fmin_nxv64f16'
@@ -510,7 +510,7 @@ define half @vreduce_fmin_nxv64f16(<vscale x 64 x half> %v) {
define float @vreduce_fmin_nxv1f32(<vscale x 1 x float> %v) {
; CHECK-LABEL: 'vreduce_fmin_nxv1f32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call float @llvm.vector.reduce.fmin.nxv1f32(<vscale x 1 x float> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call float @llvm.vector.reduce.fmin.nxv1f32(<vscale x 1 x float> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %red
;
; SIZE-LABEL: 'vreduce_fmin_nxv1f32'
@@ -523,7 +523,7 @@ define float @vreduce_fmin_nxv1f32(<vscale x 1 x float> %v) {
define float @vreduce_fmin_nxv1f32_nonans(<vscale x 1 x float> %v) {
; CHECK-LABEL: 'vreduce_fmin_nxv1f32_nonans'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call nnan float @llvm.vector.reduce.fmin.nxv1f32(<vscale x 1 x float> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call nnan float @llvm.vector.reduce.fmin.nxv1f32(<vscale x 1 x float> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %red
;
; SIZE-LABEL: 'vreduce_fmin_nxv1f32_nonans'
@@ -536,7 +536,7 @@ define float @vreduce_fmin_nxv1f32_nonans(<vscale x 1 x float> %v) {
define float @vreduce_fmin_nxv1f32_nonans_noinfs(<vscale x 1 x float> %v) {
; CHECK-LABEL: 'vreduce_fmin_nxv1f32_nonans_noinfs'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call nnan ninf float @llvm.vector.reduce.fmin.nxv1f32(<vscale x 1 x float> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call nnan ninf float @llvm.vector.reduce.fmin.nxv1f32(<vscale x 1 x float> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %red
;
; SIZE-LABEL: 'vreduce_fmin_nxv1f32_nonans_noinfs'
@@ -549,7 +549,7 @@ define float @vreduce_fmin_nxv1f32_nonans_noinfs(<vscale x 1 x float> %v) {
define float @vreduce_fmin_nxv2f32(<vscale x 2 x float> %v) {
; CHECK-LABEL: 'vreduce_fmin_nxv2f32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call float @llvm.vector.reduce.fmin.nxv2f32(<vscale x 2 x float> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call float @llvm.vector.reduce.fmin.nxv2f32(<vscale x 2 x float> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %red
;
; SIZE-LABEL: 'vreduce_fmin_nxv2f32'
@@ -562,7 +562,7 @@ define float @vreduce_fmin_nxv2f32(<vscale x 2 x float> %v) {
define float @vreduce_fmin_nxv4f32(<vscale x 4 x float> %v) {
; CHECK-LABEL: 'vreduce_fmin_nxv4f32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call float @llvm.vector.reduce.fmin.nxv4f32(<vscale x 4 x float> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call float @llvm.vector.reduce.fmin.nxv4f32(<vscale x 4 x float> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %red
;
; SIZE-LABEL: 'vreduce_fmin_nxv4f32'
@@ -575,7 +575,7 @@ define float @vreduce_fmin_nxv4f32(<vscale x 4 x float> %v) {
define float @vreduce_fmin_nxv32f32(<vscale x 32 x float> %v) {
; CHECK-LABEL: 'vreduce_fmin_nxv32f32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %red = call float @llvm.vector.reduce.fmin.nxv32f32(<vscale x 32 x float> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %red = call float @llvm.vector.reduce.fmin.nxv32f32(<vscale x 32 x float> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %red
;
; SIZE-LABEL: 'vreduce_fmin_nxv32f32'
@@ -588,7 +588,7 @@ define float @vreduce_fmin_nxv32f32(<vscale x 32 x float> %v) {
define double @vreduce_fmin_nxv1f64(<vscale x 1 x double> %v) {
; CHECK-LABEL: 'vreduce_fmin_nxv1f64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call double @llvm.vector.reduce.fmin.nxv1f64(<vscale x 1 x double> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call double @llvm.vector.reduce.fmin.nxv1f64(<vscale x 1 x double> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %red
;
; SIZE-LABEL: 'vreduce_fmin_nxv1f64'
@@ -601,7 +601,7 @@ define double @vreduce_fmin_nxv1f64(<vscale x 1 x double> %v) {
define double @vreduce_fmin_nxv1f64_nonans(<vscale x 1 x double> %v) {
; CHECK-LABEL: 'vreduce_fmin_nxv1f64_nonans'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call nnan double @llvm.vector.reduce.fmin.nxv1f64(<vscale x 1 x double> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call nnan double @llvm.vector.reduce.fmin.nxv1f64(<vscale x 1 x double> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %red
;
; SIZE-LABEL: 'vreduce_fmin_nxv1f64_nonans'
@@ -614,7 +614,7 @@ define double @vreduce_fmin_nxv1f64_nonans(<vscale x 1 x double> %v) {
define double @vreduce_fmin_nxv1f64_nonans_noinfs(<vscale x 1 x double> %v) {
; CHECK-LABEL: 'vreduce_fmin_nxv1f64_nonans_noinfs'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call nnan ninf double @llvm.vector.reduce.fmin.nxv1f64(<vscale x 1 x double> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call nnan ninf double @llvm.vector.reduce.fmin.nxv1f64(<vscale x 1 x double> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %red
;
; SIZE-LABEL: 'vreduce_fmin_nxv1f64_nonans_noinfs'
@@ -627,7 +627,7 @@ define double @vreduce_fmin_nxv1f64_nonans_noinfs(<vscale x 1 x double> %v) {
define double @vreduce_fmin_nxv2f64(<vscale x 2 x double> %v) {
; CHECK-LABEL: 'vreduce_fmin_nxv2f64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call double @llvm.vector.reduce.fmin.nxv2f64(<vscale x 2 x double> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call double @llvm.vector.reduce.fmin.nxv2f64(<vscale x 2 x double> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %red
;
; SIZE-LABEL: 'vreduce_fmin_nxv2f64'
@@ -640,7 +640,7 @@ define double @vreduce_fmin_nxv2f64(<vscale x 2 x double> %v) {
define double @vreduce_fmin_nxv4f64(<vscale x 4 x double> %v) {
; CHECK-LABEL: 'vreduce_fmin_nxv4f64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call double @llvm.vector.reduce.fmin.nxv4f64(<vscale x 4 x double> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call double @llvm.vector.reduce.fmin.nxv4f64(<vscale x 4 x double> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %red
;
; SIZE-LABEL: 'vreduce_fmin_nxv4f64'
@@ -653,7 +653,7 @@ define double @vreduce_fmin_nxv4f64(<vscale x 4 x double> %v) {
define double @vreduce_fmin_nxv16f64(<vscale x 16 x double> %v) {
; CHECK-LABEL: 'vreduce_fmin_nxv16f64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %red = call double @llvm.vector.reduce.fmin.nxv16f64(<vscale x 16 x double> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %red = call double @llvm.vector.reduce.fmin.nxv16f64(<vscale x 16 x double> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %red
;
; SIZE-LABEL: 'vreduce_fmin_nxv16f64'
@@ -666,7 +666,7 @@ define double @vreduce_fmin_nxv16f64(<vscale x 16 x double> %v) {
define half @vreduce_fmax_nxv1f16(<vscale x 1 x half> %v) {
; CHECK-LABEL: 'vreduce_fmax_nxv1f16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call half @llvm.vector.reduce.fmax.nxv1f16(<vscale x 1 x half> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call half @llvm.vector.reduce.fmax.nxv1f16(<vscale x 1 x half> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret half %red
;
; SIZE-LABEL: 'vreduce_fmax_nxv1f16'
@@ -679,7 +679,7 @@ define half @vreduce_fmax_nxv1f16(<vscale x 1 x half> %v) {
define half @vreduce_fmax_nxv1f16_nonans(<vscale x 1 x half> %v) {
; CHECK-LABEL: 'vreduce_fmax_nxv1f16_nonans'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call nnan half @llvm.vector.reduce.fmax.nxv1f16(<vscale x 1 x half> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call nnan half @llvm.vector.reduce.fmax.nxv1f16(<vscale x 1 x half> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret half %red
;
; SIZE-LABEL: 'vreduce_fmax_nxv1f16_nonans'
@@ -692,7 +692,7 @@ define half @vreduce_fmax_nxv1f16_nonans(<vscale x 1 x half> %v) {
define half @vreduce_fmax_nxv1f16_nonans_noinfs(<vscale x 1 x half> %v) #1 {
; CHECK-LABEL: 'vreduce_fmax_nxv1f16_nonans_noinfs'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call nnan ninf half @llvm.vector.reduce.fmax.nxv1f16(<vscale x 1 x half> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call nnan ninf half @llvm.vector.reduce.fmax.nxv1f16(<vscale x 1 x half> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret half %red
;
; SIZE-LABEL: 'vreduce_fmax_nxv1f16_nonans_noinfs'
@@ -705,7 +705,7 @@ define half @vreduce_fmax_nxv1f16_nonans_noinfs(<vscale x 1 x half> %v) #1 {
define half @vreduce_fmax_nxv2f16(<vscale x 2 x half> %v) {
; CHECK-LABEL: 'vreduce_fmax_nxv2f16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call half @llvm.vector.reduce.fmax.nxv2f16(<vscale x 2 x half> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call half @llvm.vector.reduce.fmax.nxv2f16(<vscale x 2 x half> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret half %red
;
; SIZE-LABEL: 'vreduce_fmax_nxv2f16'
@@ -718,7 +718,7 @@ define half @vreduce_fmax_nxv2f16(<vscale x 2 x half> %v) {
define half @vreduce_fmax_nxv4f16(<vscale x 4 x half> %v) {
; CHECK-LABEL: 'vreduce_fmax_nxv4f16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call half @llvm.vector.reduce.fmax.nxv4f16(<vscale x 4 x half> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call half @llvm.vector.reduce.fmax.nxv4f16(<vscale x 4 x half> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret half %red
;
; SIZE-LABEL: 'vreduce_fmax_nxv4f16'
@@ -731,7 +731,7 @@ define half @vreduce_fmax_nxv4f16(<vscale x 4 x half> %v) {
define half @vreduce_fmax_nxv64f16(<vscale x 64 x half> %v) {
; CHECK-LABEL: 'vreduce_fmax_nxv64f16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %red = call half @llvm.vector.reduce.fmax.nxv64f16(<vscale x 64 x half> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %red = call half @llvm.vector.reduce.fmax.nxv64f16(<vscale x 64 x half> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret half %red
;
; SIZE-LABEL: 'vreduce_fmax_nxv64f16'
@@ -744,7 +744,7 @@ define half @vreduce_fmax_nxv64f16(<vscale x 64 x half> %v) {
define float @vreduce_fmax_nxv1f32(<vscale x 1 x float> %v) {
; CHECK-LABEL: 'vreduce_fmax_nxv1f32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call float @llvm.vector.reduce.fmax.nxv1f32(<vscale x 1 x float> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call float @llvm.vector.reduce.fmax.nxv1f32(<vscale x 1 x float> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %red
;
; SIZE-LABEL: 'vreduce_fmax_nxv1f32'
@@ -757,7 +757,7 @@ define float @vreduce_fmax_nxv1f32(<vscale x 1 x float> %v) {
define float @vreduce_fmax_nxv1f32_nonans(<vscale x 1 x float> %v) {
; CHECK-LABEL: 'vreduce_fmax_nxv1f32_nonans'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call nnan float @llvm.vector.reduce.fmax.nxv1f32(<vscale x 1 x float> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call nnan float @llvm.vector.reduce.fmax.nxv1f32(<vscale x 1 x float> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %red
;
; SIZE-LABEL: 'vreduce_fmax_nxv1f32_nonans'
@@ -770,7 +770,7 @@ define float @vreduce_fmax_nxv1f32_nonans(<vscale x 1 x float> %v) {
define float @vreduce_fmax_nxv1f32_nonans_noinfs(<vscale x 1 x float> %v) {
; CHECK-LABEL: 'vreduce_fmax_nxv1f32_nonans_noinfs'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call nnan ninf float @llvm.vector.reduce.fmax.nxv1f32(<vscale x 1 x float> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call nnan ninf float @llvm.vector.reduce.fmax.nxv1f32(<vscale x 1 x float> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %red
;
; SIZE-LABEL: 'vreduce_fmax_nxv1f32_nonans_noinfs'
@@ -783,7 +783,7 @@ define float @vreduce_fmax_nxv1f32_nonans_noinfs(<vscale x 1 x float> %v) {
define float @vreduce_fmax_nxv2f32(<vscale x 2 x float> %v) {
; CHECK-LABEL: 'vreduce_fmax_nxv2f32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call float @llvm.vector.reduce.fmax.nxv2f32(<vscale x 2 x float> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call float @llvm.vector.reduce.fmax.nxv2f32(<vscale x 2 x float> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %red
;
; SIZE-LABEL: 'vreduce_fmax_nxv2f32'
@@ -796,7 +796,7 @@ define float @vreduce_fmax_nxv2f32(<vscale x 2 x float> %v) {
define float @vreduce_fmax_nxv4f32(<vscale x 4 x float> %v) {
; CHECK-LABEL: 'vreduce_fmax_nxv4f32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call float @llvm.vector.reduce.fmax.nxv4f32(<vscale x 4 x float> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call float @llvm.vector.reduce.fmax.nxv4f32(<vscale x 4 x float> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %red
;
; SIZE-LABEL: 'vreduce_fmax_nxv4f32'
@@ -809,7 +809,7 @@ define float @vreduce_fmax_nxv4f32(<vscale x 4 x float> %v) {
define float @vreduce_fmax_nxv32f32(<vscale x 32 x float> %v) {
; CHECK-LABEL: 'vreduce_fmax_nxv32f32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %red = call float @llvm.vector.reduce.fmax.nxv32f32(<vscale x 32 x float> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %red = call float @llvm.vector.reduce.fmax.nxv32f32(<vscale x 32 x float> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %red
;
; SIZE-LABEL: 'vreduce_fmax_nxv32f32'
@@ -822,7 +822,7 @@ define float @vreduce_fmax_nxv32f32(<vscale x 32 x float> %v) {
define double @vreduce_fmax_nxv1f64(<vscale x 1 x double> %v) {
; CHECK-LABEL: 'vreduce_fmax_nxv1f64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call double @llvm.vector.reduce.fmax.nxv1f64(<vscale x 1 x double> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call double @llvm.vector.reduce.fmax.nxv1f64(<vscale x 1 x double> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %red
;
; SIZE-LABEL: 'vreduce_fmax_nxv1f64'
@@ -835,7 +835,7 @@ define double @vreduce_fmax_nxv1f64(<vscale x 1 x double> %v) {
define double @vreduce_fmax_nxv1f64_nonans(<vscale x 1 x double> %v) {
; CHECK-LABEL: 'vreduce_fmax_nxv1f64_nonans'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call nnan double @llvm.vector.reduce.fmax.nxv1f64(<vscale x 1 x double> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call nnan double @llvm.vector.reduce.fmax.nxv1f64(<vscale x 1 x double> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %red
;
; SIZE-LABEL: 'vreduce_fmax_nxv1f64_nonans'
@@ -848,7 +848,7 @@ define double @vreduce_fmax_nxv1f64_nonans(<vscale x 1 x double> %v) {
define double @vreduce_fmax_nxv1f64_nonans_noinfs(<vscale x 1 x double> %v) {
; CHECK-LABEL: 'vreduce_fmax_nxv1f64_nonans_noinfs'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call nnan ninf double @llvm.vector.reduce.fmax.nxv1f64(<vscale x 1 x double> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call nnan ninf double @llvm.vector.reduce.fmax.nxv1f64(<vscale x 1 x double> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %red
;
; SIZE-LABEL: 'vreduce_fmax_nxv1f64_nonans_noinfs'
@@ -861,7 +861,7 @@ define double @vreduce_fmax_nxv1f64_nonans_noinfs(<vscale x 1 x double> %v) {
define double @vreduce_fmax_nxv2f64(<vscale x 2 x double> %v) {
; CHECK-LABEL: 'vreduce_fmax_nxv2f64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call double @llvm.vector.reduce.fmax.nxv2f64(<vscale x 2 x double> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call double @llvm.vector.reduce.fmax.nxv2f64(<vscale x 2 x double> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %red
;
; SIZE-LABEL: 'vreduce_fmax_nxv2f64'
@@ -874,7 +874,7 @@ define double @vreduce_fmax_nxv2f64(<vscale x 2 x double> %v) {
define double @vreduce_fmax_nxv4f64(<vscale x 4 x double> %v) {
; CHECK-LABEL: 'vreduce_fmax_nxv4f64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call double @llvm.vector.reduce.fmax.nxv4f64(<vscale x 4 x double> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call double @llvm.vector.reduce.fmax.nxv4f64(<vscale x 4 x double> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %red
;
; SIZE-LABEL: 'vreduce_fmax_nxv4f64'
@@ -887,7 +887,7 @@ define double @vreduce_fmax_nxv4f64(<vscale x 4 x double> %v) {
define double @vreduce_fmax_nxv16f64(<vscale x 16 x double> %v) {
; CHECK-LABEL: 'vreduce_fmax_nxv16f64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %red = call double @llvm.vector.reduce.fmax.nxv16f64(<vscale x 16 x double> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %red = call double @llvm.vector.reduce.fmax.nxv16f64(<vscale x 16 x double> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double %red
;
; SIZE-LABEL: 'vreduce_fmax_nxv16f64'
@@ -900,7 +900,7 @@ define double @vreduce_fmax_nxv16f64(<vscale x 16 x double> %v) {
define float @vreduce_nsz_fadd_nxv1f32(<vscale x 1 x float> %v, float %s) {
; CHECK-LABEL: 'vreduce_nsz_fadd_nxv1f32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call reassoc nsz float @llvm.vector.reduce.fadd.nxv1f32(float %s, <vscale x 1 x float> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call reassoc nsz float @llvm.vector.reduce.fadd.nxv1f32(float %s, <vscale x 1 x float> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret float %red
;
; SIZE-LABEL: 'vreduce_nsz_fadd_nxv1f32'
diff --git a/llvm/test/Analysis/CostModel/RISCV/reduce-scalable-int.ll b/llvm/test/Analysis/CostModel/RISCV/reduce-scalable-int.ll
index 131e89f2fc2a3..3444070d10dcc 100644
--- a/llvm/test/Analysis/CostModel/RISCV/reduce-scalable-int.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/reduce-scalable-int.ll
@@ -6,7 +6,7 @@
define signext i8 @vreduce_add_nxv1i8(<vscale x 1 x i8> %v) {
; CHECK-LABEL: 'vreduce_add_nxv1i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i8 @llvm.vector.reduce.add.nxv1i8(<vscale x 1 x i8> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i8 @llvm.vector.reduce.add.nxv1i8(<vscale x 1 x i8> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i8 %red
;
; SIZE-LABEL: 'vreduce_add_nxv1i8'
@@ -19,7 +19,7 @@ define signext i8 @vreduce_add_nxv1i8(<vscale x 1 x i8> %v) {
define signext i8 @vreduce_umax_nxv1i8(<vscale x 1 x i8> %v) {
; CHECK-LABEL: 'vreduce_umax_nxv1i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call i8 @llvm.vector.reduce.umax.nxv1i8(<vscale x 1 x i8> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i8 @llvm.vector.reduce.umax.nxv1i8(<vscale x 1 x i8> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i8 %red
;
; SIZE-LABEL: 'vreduce_umax_nxv1i8'
@@ -32,7 +32,7 @@ define signext i8 @vreduce_umax_nxv1i8(<vscale x 1 x i8> %v) {
define signext i8 @vreduce_smax_nxv1i8(<vscale x 1 x i8> %v) {
; CHECK-LABEL: 'vreduce_smax_nxv1i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call i8 @llvm.vector.reduce.smax.nxv1i8(<vscale x 1 x i8> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i8 @llvm.vector.reduce.smax.nxv1i8(<vscale x 1 x i8> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i8 %red
;
; SIZE-LABEL: 'vreduce_smax_nxv1i8'
@@ -45,7 +45,7 @@ define signext i8 @vreduce_smax_nxv1i8(<vscale x 1 x i8> %v) {
define signext i8 @vreduce_umin_nxv1i8(<vscale x 1 x i8> %v) {
; CHECK-LABEL: 'vreduce_umin_nxv1i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call i8 @llvm.vector.reduce.umin.nxv1i8(<vscale x 1 x i8> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i8 @llvm.vector.reduce.umin.nxv1i8(<vscale x 1 x i8> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i8 %red
;
; SIZE-LABEL: 'vreduce_umin_nxv1i8'
@@ -58,7 +58,7 @@ define signext i8 @vreduce_umin_nxv1i8(<vscale x 1 x i8> %v) {
define signext i8 @vreduce_smin_nxv1i8(<vscale x 1 x i8> %v) {
; CHECK-LABEL: 'vreduce_smin_nxv1i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call i8 @llvm.vector.reduce.smin.nxv1i8(<vscale x 1 x i8> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i8 @llvm.vector.reduce.smin.nxv1i8(<vscale x 1 x i8> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i8 %red
;
; SIZE-LABEL: 'vreduce_smin_nxv1i8'
@@ -71,7 +71,7 @@ define signext i8 @vreduce_smin_nxv1i8(<vscale x 1 x i8> %v) {
define signext i8 @vreduce_and_nxv1i8(<vscale x 1 x i8> %v) {
; CHECK-LABEL: 'vreduce_and_nxv1i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call i8 @llvm.vector.reduce.and.nxv1i8(<vscale x 1 x i8> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i8 @llvm.vector.reduce.and.nxv1i8(<vscale x 1 x i8> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i8 %red
;
; SIZE-LABEL: 'vreduce_and_nxv1i8'
@@ -84,7 +84,7 @@ define signext i8 @vreduce_and_nxv1i8(<vscale x 1 x i8> %v) {
define signext i8 @vreduce_or_nxv1i8(<vscale x 1 x i8> %v) {
; CHECK-LABEL: 'vreduce_or_nxv1i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call i8 @llvm.vector.reduce.or.nxv1i8(<vscale x 1 x i8> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i8 @llvm.vector.reduce.or.nxv1i8(<vscale x 1 x i8> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i8 %red
;
; SIZE-LABEL: 'vreduce_or_nxv1i8'
@@ -97,7 +97,7 @@ define signext i8 @vreduce_or_nxv1i8(<vscale x 1 x i8> %v) {
define signext i8 @vreduce_xor_nxv1i8(<vscale x 1 x i8> %v) {
; CHECK-LABEL: 'vreduce_xor_nxv1i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i8 @llvm.vector.reduce.xor.nxv1i8(<vscale x 1 x i8> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i8 @llvm.vector.reduce.xor.nxv1i8(<vscale x 1 x i8> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i8 %red
;
; SIZE-LABEL: 'vreduce_xor_nxv1i8'
@@ -110,7 +110,7 @@ define signext i8 @vreduce_xor_nxv1i8(<vscale x 1 x i8> %v) {
define signext i8 @vreduce_add_nxv2i8(<vscale x 2 x i8> %v) {
; CHECK-LABEL: 'vreduce_add_nxv2i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i8 @llvm.vector.reduce.add.nxv2i8(<vscale x 2 x i8> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i8 @llvm.vector.reduce.add.nxv2i8(<vscale x 2 x i8> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i8 %red
;
; SIZE-LABEL: 'vreduce_add_nxv2i8'
@@ -123,7 +123,7 @@ define signext i8 @vreduce_add_nxv2i8(<vscale x 2 x i8> %v) {
define signext i8 @vreduce_umax_nxv2i8(<vscale x 2 x i8> %v) {
; CHECK-LABEL: 'vreduce_umax_nxv2i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i8 @llvm.vector.reduce.umax.nxv2i8(<vscale x 2 x i8> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i8 @llvm.vector.reduce.umax.nxv2i8(<vscale x 2 x i8> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i8 %red
;
; SIZE-LABEL: 'vreduce_umax_nxv2i8'
@@ -136,7 +136,7 @@ define signext i8 @vreduce_umax_nxv2i8(<vscale x 2 x i8> %v) {
define signext i8 @vreduce_smax_nxv2i8(<vscale x 2 x i8> %v) {
; CHECK-LABEL: 'vreduce_smax_nxv2i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i8 @llvm.vector.reduce.smax.nxv2i8(<vscale x 2 x i8> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i8 @llvm.vector.reduce.smax.nxv2i8(<vscale x 2 x i8> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i8 %red
;
; SIZE-LABEL: 'vreduce_smax_nxv2i8'
@@ -149,7 +149,7 @@ define signext i8 @vreduce_smax_nxv2i8(<vscale x 2 x i8> %v) {
define signext i8 @vreduce_umin_nxv2i8(<vscale x 2 x i8> %v) {
; CHECK-LABEL: 'vreduce_umin_nxv2i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i8 @llvm.vector.reduce.umin.nxv2i8(<vscale x 2 x i8> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i8 @llvm.vector.reduce.umin.nxv2i8(<vscale x 2 x i8> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i8 %red
;
; SIZE-LABEL: 'vreduce_umin_nxv2i8'
@@ -162,7 +162,7 @@ define signext i8 @vreduce_umin_nxv2i8(<vscale x 2 x i8> %v) {
define signext i8 @vreduce_smin_nxv2i8(<vscale x 2 x i8> %v) {
; CHECK-LABEL: 'vreduce_smin_nxv2i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i8 @llvm.vector.reduce.smin.nxv2i8(<vscale x 2 x i8> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i8 @llvm.vector.reduce.smin.nxv2i8(<vscale x 2 x i8> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i8 %red
;
; SIZE-LABEL: 'vreduce_smin_nxv2i8'
@@ -175,7 +175,7 @@ define signext i8 @vreduce_smin_nxv2i8(<vscale x 2 x i8> %v) {
define signext i8 @vreduce_and_nxv2i8(<vscale x 2 x i8> %v) {
; CHECK-LABEL: 'vreduce_and_nxv2i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i8 @llvm.vector.reduce.and.nxv2i8(<vscale x 2 x i8> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i8 @llvm.vector.reduce.and.nxv2i8(<vscale x 2 x i8> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i8 %red
;
; SIZE-LABEL: 'vreduce_and_nxv2i8'
@@ -188,7 +188,7 @@ define signext i8 @vreduce_and_nxv2i8(<vscale x 2 x i8> %v) {
define signext i8 @vreduce_or_nxv2i8(<vscale x 2 x i8> %v) {
; CHECK-LABEL: 'vreduce_or_nxv2i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i8 @llvm.vector.reduce.or.nxv2i8(<vscale x 2 x i8> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i8 @llvm.vector.reduce.or.nxv2i8(<vscale x 2 x i8> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i8 %red
;
; SIZE-LABEL: 'vreduce_or_nxv2i8'
@@ -201,7 +201,7 @@ define signext i8 @vreduce_or_nxv2i8(<vscale x 2 x i8> %v) {
define signext i8 @vreduce_xor_nxv2i8(<vscale x 2 x i8> %v) {
; CHECK-LABEL: 'vreduce_xor_nxv2i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i8 @llvm.vector.reduce.xor.nxv2i8(<vscale x 2 x i8> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i8 @llvm.vector.reduce.xor.nxv2i8(<vscale x 2 x i8> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i8 %red
;
; SIZE-LABEL: 'vreduce_xor_nxv2i8'
@@ -214,7 +214,7 @@ define signext i8 @vreduce_xor_nxv2i8(<vscale x 2 x i8> %v) {
define signext i8 @vreduce_add_nxv4i8(<vscale x 4 x i8> %v) {
; CHECK-LABEL: 'vreduce_add_nxv4i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i8 @llvm.vector.reduce.add.nxv4i8(<vscale x 4 x i8> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %red = call i8 @llvm.vector.reduce.add.nxv4i8(<vscale x 4 x i8> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i8 %red
;
; SIZE-LABEL: 'vreduce_add_nxv4i8'
@@ -227,7 +227,7 @@ define signext i8 @vreduce_add_nxv4i8(<vscale x 4 x i8> %v) {
define signext i8 @vreduce_umax_nxv4i8(<vscale x 4 x i8> %v) {
; CHECK-LABEL: 'vreduce_umax_nxv4i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i8 @llvm.vector.reduce.umax.nxv4i8(<vscale x 4 x i8> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i8 @llvm.vector.reduce.umax.nxv4i8(<vscale x 4 x i8> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i8 %red
;
; SIZE-LABEL: 'vreduce_umax_nxv4i8'
@@ -240,7 +240,7 @@ define signext i8 @vreduce_umax_nxv4i8(<vscale x 4 x i8> %v) {
define signext i8 @vreduce_smax_nxv4i8(<vscale x 4 x i8> %v) {
; CHECK-LABEL: 'vreduce_smax_nxv4i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i8 @llvm.vector.reduce.smax.nxv4i8(<vscale x 4 x i8> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i8 @llvm.vector.reduce.smax.nxv4i8(<vscale x 4 x i8> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i8 %red
;
; SIZE-LABEL: 'vreduce_smax_nxv4i8'
@@ -253,7 +253,7 @@ define signext i8 @vreduce_smax_nxv4i8(<vscale x 4 x i8> %v) {
define signext i8 @vreduce_umin_nxv4i8(<vscale x 4 x i8> %v) {
; CHECK-LABEL: 'vreduce_umin_nxv4i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i8 @llvm.vector.reduce.umin.nxv4i8(<vscale x 4 x i8> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i8 @llvm.vector.reduce.umin.nxv4i8(<vscale x 4 x i8> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i8 %red
;
; SIZE-LABEL: 'vreduce_umin_nxv4i8'
@@ -266,7 +266,7 @@ define signext i8 @vreduce_umin_nxv4i8(<vscale x 4 x i8> %v) {
define signext i8 @vreduce_smin_nxv4i8(<vscale x 4 x i8> %v) {
; CHECK-LABEL: 'vreduce_smin_nxv4i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i8 @llvm.vector.reduce.smin.nxv4i8(<vscale x 4 x i8> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i8 @llvm.vector.reduce.smin.nxv4i8(<vscale x 4 x i8> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i8 %red
;
; SIZE-LABEL: 'vreduce_smin_nxv4i8'
@@ -279,7 +279,7 @@ define signext i8 @vreduce_smin_nxv4i8(<vscale x 4 x i8> %v) {
define signext i8 @vreduce_and_nxv4i8(<vscale x 4 x i8> %v) {
; CHECK-LABEL: 'vreduce_and_nxv4i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i8 @llvm.vector.reduce.and.nxv4i8(<vscale x 4 x i8> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i8 @llvm.vector.reduce.and.nxv4i8(<vscale x 4 x i8> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i8 %red
;
; SIZE-LABEL: 'vreduce_and_nxv4i8'
@@ -292,7 +292,7 @@ define signext i8 @vreduce_and_nxv4i8(<vscale x 4 x i8> %v) {
define signext i8 @vreduce_or_nxv4i8(<vscale x 4 x i8> %v) {
; CHECK-LABEL: 'vreduce_or_nxv4i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i8 @llvm.vector.reduce.or.nxv4i8(<vscale x 4 x i8> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i8 @llvm.vector.reduce.or.nxv4i8(<vscale x 4 x i8> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i8 %red
;
; SIZE-LABEL: 'vreduce_or_nxv4i8'
@@ -305,7 +305,7 @@ define signext i8 @vreduce_or_nxv4i8(<vscale x 4 x i8> %v) {
define signext i8 @vreduce_xor_nxv4i8(<vscale x 4 x i8> %v) {
; CHECK-LABEL: 'vreduce_xor_nxv4i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i8 @llvm.vector.reduce.xor.nxv4i8(<vscale x 4 x i8> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %red = call i8 @llvm.vector.reduce.xor.nxv4i8(<vscale x 4 x i8> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i8 %red
;
; SIZE-LABEL: 'vreduce_xor_nxv4i8'
@@ -318,7 +318,7 @@ define signext i8 @vreduce_xor_nxv4i8(<vscale x 4 x i8> %v) {
define signext i16 @vreduce_add_nxv1i16(<vscale x 1 x i16> %v) {
; CHECK-LABEL: 'vreduce_add_nxv1i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i16 @llvm.vector.reduce.add.nxv1i16(<vscale x 1 x i16> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i16 @llvm.vector.reduce.add.nxv1i16(<vscale x 1 x i16> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vreduce_add_nxv1i16'
@@ -332,7 +332,7 @@ define signext i16 @vreduce_add_nxv1i16(<vscale x 1 x i16> %v) {
define signext i16 @vwreduce_add_nxv1i8(<vscale x 1 x i8> %v) {
; CHECK-LABEL: 'vwreduce_add_nxv1i8'
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %e = sext <vscale x 1 x i8> %v to <vscale x 1 x i16>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i16 @llvm.vector.reduce.add.nxv1i16(<vscale x 1 x i16> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i16 @llvm.vector.reduce.add.nxv1i16(<vscale x 1 x i16> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vwreduce_add_nxv1i8'
@@ -348,7 +348,7 @@ define signext i16 @vwreduce_add_nxv1i8(<vscale x 1 x i8> %v) {
define signext i16 @vwreduce_uadd_nxv1i8(<vscale x 1 x i8> %v) {
; CHECK-LABEL: 'vwreduce_uadd_nxv1i8'
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %e = sext <vscale x 1 x i8> %v to <vscale x 1 x i16>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i16 @llvm.vector.reduce.add.nxv1i16(<vscale x 1 x i16> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i16 @llvm.vector.reduce.add.nxv1i16(<vscale x 1 x i16> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vwreduce_uadd_nxv1i8'
@@ -363,7 +363,7 @@ define signext i16 @vwreduce_uadd_nxv1i8(<vscale x 1 x i8> %v) {
define signext i16 @vreduce_umax_nxv1i16(<vscale x 1 x i16> %v) {
; CHECK-LABEL: 'vreduce_umax_nxv1i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call i16 @llvm.vector.reduce.umax.nxv1i16(<vscale x 1 x i16> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i16 @llvm.vector.reduce.umax.nxv1i16(<vscale x 1 x i16> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vreduce_umax_nxv1i16'
@@ -376,7 +376,7 @@ define signext i16 @vreduce_umax_nxv1i16(<vscale x 1 x i16> %v) {
define signext i16 @vreduce_smax_nxv1i16(<vscale x 1 x i16> %v) {
; CHECK-LABEL: 'vreduce_smax_nxv1i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call i16 @llvm.vector.reduce.smax.nxv1i16(<vscale x 1 x i16> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i16 @llvm.vector.reduce.smax.nxv1i16(<vscale x 1 x i16> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vreduce_smax_nxv1i16'
@@ -389,7 +389,7 @@ define signext i16 @vreduce_smax_nxv1i16(<vscale x 1 x i16> %v) {
define signext i16 @vreduce_umin_nxv1i16(<vscale x 1 x i16> %v) {
; CHECK-LABEL: 'vreduce_umin_nxv1i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call i16 @llvm.vector.reduce.umin.nxv1i16(<vscale x 1 x i16> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i16 @llvm.vector.reduce.umin.nxv1i16(<vscale x 1 x i16> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vreduce_umin_nxv1i16'
@@ -402,7 +402,7 @@ define signext i16 @vreduce_umin_nxv1i16(<vscale x 1 x i16> %v) {
define signext i16 @vreduce_smin_nxv1i16(<vscale x 1 x i16> %v) {
; CHECK-LABEL: 'vreduce_smin_nxv1i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call i16 @llvm.vector.reduce.smin.nxv1i16(<vscale x 1 x i16> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i16 @llvm.vector.reduce.smin.nxv1i16(<vscale x 1 x i16> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vreduce_smin_nxv1i16'
@@ -415,7 +415,7 @@ define signext i16 @vreduce_smin_nxv1i16(<vscale x 1 x i16> %v) {
define signext i16 @vreduce_and_nxv1i16(<vscale x 1 x i16> %v) {
; CHECK-LABEL: 'vreduce_and_nxv1i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call i16 @llvm.vector.reduce.and.nxv1i16(<vscale x 1 x i16> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i16 @llvm.vector.reduce.and.nxv1i16(<vscale x 1 x i16> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vreduce_and_nxv1i16'
@@ -428,7 +428,7 @@ define signext i16 @vreduce_and_nxv1i16(<vscale x 1 x i16> %v) {
define signext i16 @vreduce_or_nxv1i16(<vscale x 1 x i16> %v) {
; CHECK-LABEL: 'vreduce_or_nxv1i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call i16 @llvm.vector.reduce.or.nxv1i16(<vscale x 1 x i16> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i16 @llvm.vector.reduce.or.nxv1i16(<vscale x 1 x i16> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vreduce_or_nxv1i16'
@@ -441,7 +441,7 @@ define signext i16 @vreduce_or_nxv1i16(<vscale x 1 x i16> %v) {
define signext i16 @vreduce_xor_nxv1i16(<vscale x 1 x i16> %v) {
; CHECK-LABEL: 'vreduce_xor_nxv1i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i16 @llvm.vector.reduce.xor.nxv1i16(<vscale x 1 x i16> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i16 @llvm.vector.reduce.xor.nxv1i16(<vscale x 1 x i16> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vreduce_xor_nxv1i16'
@@ -454,7 +454,7 @@ define signext i16 @vreduce_xor_nxv1i16(<vscale x 1 x i16> %v) {
define signext i16 @vreduce_add_nxv2i16(<vscale x 2 x i16> %v) {
; CHECK-LABEL: 'vreduce_add_nxv2i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i16 @llvm.vector.reduce.add.nxv2i16(<vscale x 2 x i16> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i16 @llvm.vector.reduce.add.nxv2i16(<vscale x 2 x i16> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vreduce_add_nxv2i16'
@@ -468,7 +468,7 @@ define signext i16 @vreduce_add_nxv2i16(<vscale x 2 x i16> %v) {
define signext i16 @vwreduce_add_nxv2i8(<vscale x 2 x i8> %v) {
; CHECK-LABEL: 'vwreduce_add_nxv2i8'
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %e = sext <vscale x 2 x i8> %v to <vscale x 2 x i16>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i16 @llvm.vector.reduce.add.nxv2i16(<vscale x 2 x i16> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i16 @llvm.vector.reduce.add.nxv2i16(<vscale x 2 x i16> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vwreduce_add_nxv2i8'
@@ -484,7 +484,7 @@ define signext i16 @vwreduce_add_nxv2i8(<vscale x 2 x i8> %v) {
define signext i16 @vwreduce_uadd_nxv2i8(<vscale x 2 x i8> %v) {
; CHECK-LABEL: 'vwreduce_uadd_nxv2i8'
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %e = sext <vscale x 2 x i8> %v to <vscale x 2 x i16>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i16 @llvm.vector.reduce.add.nxv2i16(<vscale x 2 x i16> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i16 @llvm.vector.reduce.add.nxv2i16(<vscale x 2 x i16> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vwreduce_uadd_nxv2i8'
@@ -499,7 +499,7 @@ define signext i16 @vwreduce_uadd_nxv2i8(<vscale x 2 x i8> %v) {
define signext i16 @vreduce_umax_nxv2i16(<vscale x 2 x i16> %v) {
; CHECK-LABEL: 'vreduce_umax_nxv2i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i16 @llvm.vector.reduce.umax.nxv2i16(<vscale x 2 x i16> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i16 @llvm.vector.reduce.umax.nxv2i16(<vscale x 2 x i16> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vreduce_umax_nxv2i16'
@@ -512,7 +512,7 @@ define signext i16 @vreduce_umax_nxv2i16(<vscale x 2 x i16> %v) {
define signext i16 @vreduce_smax_nxv2i16(<vscale x 2 x i16> %v) {
; CHECK-LABEL: 'vreduce_smax_nxv2i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i16 @llvm.vector.reduce.smax.nxv2i16(<vscale x 2 x i16> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i16 @llvm.vector.reduce.smax.nxv2i16(<vscale x 2 x i16> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vreduce_smax_nxv2i16'
@@ -525,7 +525,7 @@ define signext i16 @vreduce_smax_nxv2i16(<vscale x 2 x i16> %v) {
define signext i16 @vreduce_umin_nxv2i16(<vscale x 2 x i16> %v) {
; CHECK-LABEL: 'vreduce_umin_nxv2i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i16 @llvm.vector.reduce.umin.nxv2i16(<vscale x 2 x i16> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i16 @llvm.vector.reduce.umin.nxv2i16(<vscale x 2 x i16> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vreduce_umin_nxv2i16'
@@ -538,7 +538,7 @@ define signext i16 @vreduce_umin_nxv2i16(<vscale x 2 x i16> %v) {
define signext i16 @vreduce_smin_nxv2i16(<vscale x 2 x i16> %v) {
; CHECK-LABEL: 'vreduce_smin_nxv2i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i16 @llvm.vector.reduce.smin.nxv2i16(<vscale x 2 x i16> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i16 @llvm.vector.reduce.smin.nxv2i16(<vscale x 2 x i16> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vreduce_smin_nxv2i16'
@@ -551,7 +551,7 @@ define signext i16 @vreduce_smin_nxv2i16(<vscale x 2 x i16> %v) {
define signext i16 @vreduce_and_nxv2i16(<vscale x 2 x i16> %v) {
; CHECK-LABEL: 'vreduce_and_nxv2i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i16 @llvm.vector.reduce.and.nxv2i16(<vscale x 2 x i16> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i16 @llvm.vector.reduce.and.nxv2i16(<vscale x 2 x i16> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vreduce_and_nxv2i16'
@@ -564,7 +564,7 @@ define signext i16 @vreduce_and_nxv2i16(<vscale x 2 x i16> %v) {
define signext i16 @vreduce_or_nxv2i16(<vscale x 2 x i16> %v) {
; CHECK-LABEL: 'vreduce_or_nxv2i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i16 @llvm.vector.reduce.or.nxv2i16(<vscale x 2 x i16> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i16 @llvm.vector.reduce.or.nxv2i16(<vscale x 2 x i16> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vreduce_or_nxv2i16'
@@ -577,7 +577,7 @@ define signext i16 @vreduce_or_nxv2i16(<vscale x 2 x i16> %v) {
define signext i16 @vreduce_xor_nxv2i16(<vscale x 2 x i16> %v) {
; CHECK-LABEL: 'vreduce_xor_nxv2i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i16 @llvm.vector.reduce.xor.nxv2i16(<vscale x 2 x i16> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i16 @llvm.vector.reduce.xor.nxv2i16(<vscale x 2 x i16> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vreduce_xor_nxv2i16'
@@ -590,7 +590,7 @@ define signext i16 @vreduce_xor_nxv2i16(<vscale x 2 x i16> %v) {
define signext i16 @vreduce_add_nxv4i16(<vscale x 4 x i16> %v) {
; CHECK-LABEL: 'vreduce_add_nxv4i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i16 @llvm.vector.reduce.add.nxv4i16(<vscale x 4 x i16> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %red = call i16 @llvm.vector.reduce.add.nxv4i16(<vscale x 4 x i16> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vreduce_add_nxv4i16'
@@ -604,7 +604,7 @@ define signext i16 @vreduce_add_nxv4i16(<vscale x 4 x i16> %v) {
define signext i16 @vwreduce_add_nxv4i8(<vscale x 4 x i8> %v) {
; CHECK-LABEL: 'vwreduce_add_nxv4i8'
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %e = sext <vscale x 4 x i8> %v to <vscale x 4 x i16>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i16 @llvm.vector.reduce.add.nxv4i16(<vscale x 4 x i16> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %red = call i16 @llvm.vector.reduce.add.nxv4i16(<vscale x 4 x i16> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vwreduce_add_nxv4i8'
@@ -620,7 +620,7 @@ define signext i16 @vwreduce_add_nxv4i8(<vscale x 4 x i8> %v) {
define signext i16 @vwreduce_uadd_nxv4i8(<vscale x 4 x i8> %v) {
; CHECK-LABEL: 'vwreduce_uadd_nxv4i8'
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %e = sext <vscale x 4 x i8> %v to <vscale x 4 x i16>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i16 @llvm.vector.reduce.add.nxv4i16(<vscale x 4 x i16> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %red = call i16 @llvm.vector.reduce.add.nxv4i16(<vscale x 4 x i16> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vwreduce_uadd_nxv4i8'
@@ -635,7 +635,7 @@ define signext i16 @vwreduce_uadd_nxv4i8(<vscale x 4 x i8> %v) {
define signext i16 @vreduce_umax_nxv4i16(<vscale x 4 x i16> %v) {
; CHECK-LABEL: 'vreduce_umax_nxv4i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i16 @llvm.vector.reduce.umax.nxv4i16(<vscale x 4 x i16> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i16 @llvm.vector.reduce.umax.nxv4i16(<vscale x 4 x i16> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vreduce_umax_nxv4i16'
@@ -648,7 +648,7 @@ define signext i16 @vreduce_umax_nxv4i16(<vscale x 4 x i16> %v) {
define signext i16 @vreduce_smax_nxv4i16(<vscale x 4 x i16> %v) {
; CHECK-LABEL: 'vreduce_smax_nxv4i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i16 @llvm.vector.reduce.smax.nxv4i16(<vscale x 4 x i16> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i16 @llvm.vector.reduce.smax.nxv4i16(<vscale x 4 x i16> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vreduce_smax_nxv4i16'
@@ -661,7 +661,7 @@ define signext i16 @vreduce_smax_nxv4i16(<vscale x 4 x i16> %v) {
define signext i16 @vreduce_umin_nxv4i16(<vscale x 4 x i16> %v) {
; CHECK-LABEL: 'vreduce_umin_nxv4i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i16 @llvm.vector.reduce.umin.nxv4i16(<vscale x 4 x i16> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i16 @llvm.vector.reduce.umin.nxv4i16(<vscale x 4 x i16> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vreduce_umin_nxv4i16'
@@ -674,7 +674,7 @@ define signext i16 @vreduce_umin_nxv4i16(<vscale x 4 x i16> %v) {
define signext i16 @vreduce_smin_nxv4i16(<vscale x 4 x i16> %v) {
; CHECK-LABEL: 'vreduce_smin_nxv4i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i16 @llvm.vector.reduce.smin.nxv4i16(<vscale x 4 x i16> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i16 @llvm.vector.reduce.smin.nxv4i16(<vscale x 4 x i16> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vreduce_smin_nxv4i16'
@@ -687,7 +687,7 @@ define signext i16 @vreduce_smin_nxv4i16(<vscale x 4 x i16> %v) {
define signext i16 @vreduce_and_nxv4i16(<vscale x 4 x i16> %v) {
; CHECK-LABEL: 'vreduce_and_nxv4i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i16 @llvm.vector.reduce.and.nxv4i16(<vscale x 4 x i16> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i16 @llvm.vector.reduce.and.nxv4i16(<vscale x 4 x i16> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vreduce_and_nxv4i16'
@@ -700,7 +700,7 @@ define signext i16 @vreduce_and_nxv4i16(<vscale x 4 x i16> %v) {
define signext i16 @vreduce_or_nxv4i16(<vscale x 4 x i16> %v) {
; CHECK-LABEL: 'vreduce_or_nxv4i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i16 @llvm.vector.reduce.or.nxv4i16(<vscale x 4 x i16> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i16 @llvm.vector.reduce.or.nxv4i16(<vscale x 4 x i16> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vreduce_or_nxv4i16'
@@ -713,7 +713,7 @@ define signext i16 @vreduce_or_nxv4i16(<vscale x 4 x i16> %v) {
define signext i16 @vreduce_xor_nxv4i16(<vscale x 4 x i16> %v) {
; CHECK-LABEL: 'vreduce_xor_nxv4i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i16 @llvm.vector.reduce.xor.nxv4i16(<vscale x 4 x i16> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %red = call i16 @llvm.vector.reduce.xor.nxv4i16(<vscale x 4 x i16> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i16 %red
;
; SIZE-LABEL: 'vreduce_xor_nxv4i16'
@@ -726,7 +726,7 @@ define signext i16 @vreduce_xor_nxv4i16(<vscale x 4 x i16> %v) {
define signext i32 @vreduce_add_nxv1i32(<vscale x 1 x i32> %v) {
; CHECK-LABEL: 'vreduce_add_nxv1i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i32 @llvm.vector.reduce.add.nxv1i32(<vscale x 1 x i32> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i32 @llvm.vector.reduce.add.nxv1i32(<vscale x 1 x i32> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vreduce_add_nxv1i32'
@@ -740,7 +740,7 @@ define signext i32 @vreduce_add_nxv1i32(<vscale x 1 x i32> %v) {
define signext i32 @vwreduce_add_nxv1i16(<vscale x 1 x i16> %v) {
; CHECK-LABEL: 'vwreduce_add_nxv1i16'
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %e = sext <vscale x 1 x i16> %v to <vscale x 1 x i32>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i32 @llvm.vector.reduce.add.nxv1i32(<vscale x 1 x i32> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i32 @llvm.vector.reduce.add.nxv1i32(<vscale x 1 x i32> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vwreduce_add_nxv1i16'
@@ -756,7 +756,7 @@ define signext i32 @vwreduce_add_nxv1i16(<vscale x 1 x i16> %v) {
define signext i32 @vwreduce_uadd_nxv1i16(<vscale x 1 x i16> %v) {
; CHECK-LABEL: 'vwreduce_uadd_nxv1i16'
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %e = zext <vscale x 1 x i16> %v to <vscale x 1 x i32>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i32 @llvm.vector.reduce.add.nxv1i32(<vscale x 1 x i32> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i32 @llvm.vector.reduce.add.nxv1i32(<vscale x 1 x i32> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vwreduce_uadd_nxv1i16'
@@ -771,7 +771,7 @@ define signext i32 @vwreduce_uadd_nxv1i16(<vscale x 1 x i16> %v) {
define signext i32 @vreduce_umax_nxv1i32(<vscale x 1 x i32> %v) {
; CHECK-LABEL: 'vreduce_umax_nxv1i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call i32 @llvm.vector.reduce.umax.nxv1i32(<vscale x 1 x i32> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i32 @llvm.vector.reduce.umax.nxv1i32(<vscale x 1 x i32> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vreduce_umax_nxv1i32'
@@ -784,7 +784,7 @@ define signext i32 @vreduce_umax_nxv1i32(<vscale x 1 x i32> %v) {
define signext i32 @vreduce_smax_nxv1i32(<vscale x 1 x i32> %v) {
; CHECK-LABEL: 'vreduce_smax_nxv1i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call i32 @llvm.vector.reduce.smax.nxv1i32(<vscale x 1 x i32> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i32 @llvm.vector.reduce.smax.nxv1i32(<vscale x 1 x i32> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vreduce_smax_nxv1i32'
@@ -797,7 +797,7 @@ define signext i32 @vreduce_smax_nxv1i32(<vscale x 1 x i32> %v) {
define signext i32 @vreduce_umin_nxv1i32(<vscale x 1 x i32> %v) {
; CHECK-LABEL: 'vreduce_umin_nxv1i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call i32 @llvm.vector.reduce.umin.nxv1i32(<vscale x 1 x i32> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i32 @llvm.vector.reduce.umin.nxv1i32(<vscale x 1 x i32> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vreduce_umin_nxv1i32'
@@ -810,7 +810,7 @@ define signext i32 @vreduce_umin_nxv1i32(<vscale x 1 x i32> %v) {
define signext i32 @vreduce_smin_nxv1i32(<vscale x 1 x i32> %v) {
; CHECK-LABEL: 'vreduce_smin_nxv1i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call i32 @llvm.vector.reduce.smin.nxv1i32(<vscale x 1 x i32> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i32 @llvm.vector.reduce.smin.nxv1i32(<vscale x 1 x i32> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vreduce_smin_nxv1i32'
@@ -823,7 +823,7 @@ define signext i32 @vreduce_smin_nxv1i32(<vscale x 1 x i32> %v) {
define signext i32 @vreduce_and_nxv1i32(<vscale x 1 x i32> %v) {
; CHECK-LABEL: 'vreduce_and_nxv1i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call i32 @llvm.vector.reduce.and.nxv1i32(<vscale x 1 x i32> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i32 @llvm.vector.reduce.and.nxv1i32(<vscale x 1 x i32> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vreduce_and_nxv1i32'
@@ -836,7 +836,7 @@ define signext i32 @vreduce_and_nxv1i32(<vscale x 1 x i32> %v) {
define signext i32 @vreduce_or_nxv1i32(<vscale x 1 x i32> %v) {
; CHECK-LABEL: 'vreduce_or_nxv1i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call i32 @llvm.vector.reduce.or.nxv1i32(<vscale x 1 x i32> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i32 @llvm.vector.reduce.or.nxv1i32(<vscale x 1 x i32> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vreduce_or_nxv1i32'
@@ -849,7 +849,7 @@ define signext i32 @vreduce_or_nxv1i32(<vscale x 1 x i32> %v) {
define signext i32 @vreduce_xor_nxv1i32(<vscale x 1 x i32> %v) {
; CHECK-LABEL: 'vreduce_xor_nxv1i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i32 @llvm.vector.reduce.xor.nxv1i32(<vscale x 1 x i32> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i32 @llvm.vector.reduce.xor.nxv1i32(<vscale x 1 x i32> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vreduce_xor_nxv1i32'
@@ -862,7 +862,7 @@ define signext i32 @vreduce_xor_nxv1i32(<vscale x 1 x i32> %v) {
define signext i32 @vreduce_add_nxv2i32(<vscale x 2 x i32> %v) {
; CHECK-LABEL: 'vreduce_add_nxv2i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i32 @llvm.vector.reduce.add.nxv2i32(<vscale x 2 x i32> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i32 @llvm.vector.reduce.add.nxv2i32(<vscale x 2 x i32> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vreduce_add_nxv2i32'
@@ -876,7 +876,7 @@ define signext i32 @vreduce_add_nxv2i32(<vscale x 2 x i32> %v) {
define signext i32 @vwreduce_add_nxv2i16(<vscale x 2 x i16> %v) {
; CHECK-LABEL: 'vwreduce_add_nxv2i16'
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %e = sext <vscale x 2 x i16> %v to <vscale x 2 x i32>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i32 @llvm.vector.reduce.add.nxv2i32(<vscale x 2 x i32> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i32 @llvm.vector.reduce.add.nxv2i32(<vscale x 2 x i32> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vwreduce_add_nxv2i16'
@@ -892,7 +892,7 @@ define signext i32 @vwreduce_add_nxv2i16(<vscale x 2 x i16> %v) {
define signext i32 @vwreduce_uadd_nxv2i16(<vscale x 2 x i16> %v) {
; CHECK-LABEL: 'vwreduce_uadd_nxv2i16'
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %e = zext <vscale x 2 x i16> %v to <vscale x 2 x i32>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i32 @llvm.vector.reduce.add.nxv2i32(<vscale x 2 x i32> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i32 @llvm.vector.reduce.add.nxv2i32(<vscale x 2 x i32> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vwreduce_uadd_nxv2i16'
@@ -907,7 +907,7 @@ define signext i32 @vwreduce_uadd_nxv2i16(<vscale x 2 x i16> %v) {
define signext i32 @vreduce_umax_nxv2i32(<vscale x 2 x i32> %v) {
; CHECK-LABEL: 'vreduce_umax_nxv2i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i32 @llvm.vector.reduce.umax.nxv2i32(<vscale x 2 x i32> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i32 @llvm.vector.reduce.umax.nxv2i32(<vscale x 2 x i32> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vreduce_umax_nxv2i32'
@@ -920,7 +920,7 @@ define signext i32 @vreduce_umax_nxv2i32(<vscale x 2 x i32> %v) {
define signext i32 @vreduce_smax_nxv2i32(<vscale x 2 x i32> %v) {
; CHECK-LABEL: 'vreduce_smax_nxv2i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i32 @llvm.vector.reduce.smax.nxv2i32(<vscale x 2 x i32> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i32 @llvm.vector.reduce.smax.nxv2i32(<vscale x 2 x i32> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vreduce_smax_nxv2i32'
@@ -933,7 +933,7 @@ define signext i32 @vreduce_smax_nxv2i32(<vscale x 2 x i32> %v) {
define signext i32 @vreduce_umin_nxv2i32(<vscale x 2 x i32> %v) {
; CHECK-LABEL: 'vreduce_umin_nxv2i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i32 @llvm.vector.reduce.umin.nxv2i32(<vscale x 2 x i32> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i32 @llvm.vector.reduce.umin.nxv2i32(<vscale x 2 x i32> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vreduce_umin_nxv2i32'
@@ -946,7 +946,7 @@ define signext i32 @vreduce_umin_nxv2i32(<vscale x 2 x i32> %v) {
define signext i32 @vreduce_smin_nxv2i32(<vscale x 2 x i32> %v) {
; CHECK-LABEL: 'vreduce_smin_nxv2i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i32 @llvm.vector.reduce.smin.nxv2i32(<vscale x 2 x i32> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i32 @llvm.vector.reduce.smin.nxv2i32(<vscale x 2 x i32> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vreduce_smin_nxv2i32'
@@ -959,7 +959,7 @@ define signext i32 @vreduce_smin_nxv2i32(<vscale x 2 x i32> %v) {
define signext i32 @vreduce_and_nxv2i32(<vscale x 2 x i32> %v) {
; CHECK-LABEL: 'vreduce_and_nxv2i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i32 @llvm.vector.reduce.and.nxv2i32(<vscale x 2 x i32> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i32 @llvm.vector.reduce.and.nxv2i32(<vscale x 2 x i32> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vreduce_and_nxv2i32'
@@ -972,7 +972,7 @@ define signext i32 @vreduce_and_nxv2i32(<vscale x 2 x i32> %v) {
define signext i32 @vreduce_or_nxv2i32(<vscale x 2 x i32> %v) {
; CHECK-LABEL: 'vreduce_or_nxv2i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i32 @llvm.vector.reduce.or.nxv2i32(<vscale x 2 x i32> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i32 @llvm.vector.reduce.or.nxv2i32(<vscale x 2 x i32> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vreduce_or_nxv2i32'
@@ -985,7 +985,7 @@ define signext i32 @vreduce_or_nxv2i32(<vscale x 2 x i32> %v) {
define signext i32 @vreduce_xor_nxv2i32(<vscale x 2 x i32> %v) {
; CHECK-LABEL: 'vreduce_xor_nxv2i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i32 @llvm.vector.reduce.xor.nxv2i32(<vscale x 2 x i32> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i32 @llvm.vector.reduce.xor.nxv2i32(<vscale x 2 x i32> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vreduce_xor_nxv2i32'
@@ -998,7 +998,7 @@ define signext i32 @vreduce_xor_nxv2i32(<vscale x 2 x i32> %v) {
define signext i32 @vreduce_add_nxv4i32(<vscale x 4 x i32> %v) {
; CHECK-LABEL: 'vreduce_add_nxv4i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i32 @llvm.vector.reduce.add.nxv4i32(<vscale x 4 x i32> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %red = call i32 @llvm.vector.reduce.add.nxv4i32(<vscale x 4 x i32> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vreduce_add_nxv4i32'
@@ -1012,7 +1012,7 @@ define signext i32 @vreduce_add_nxv4i32(<vscale x 4 x i32> %v) {
define signext i32 @vwreduce_add_nxv4i16(<vscale x 4 x i16> %v) {
; CHECK-LABEL: 'vwreduce_add_nxv4i16'
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %e = sext <vscale x 4 x i16> %v to <vscale x 4 x i32>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i32 @llvm.vector.reduce.add.nxv4i32(<vscale x 4 x i32> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %red = call i32 @llvm.vector.reduce.add.nxv4i32(<vscale x 4 x i32> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vwreduce_add_nxv4i16'
@@ -1028,7 +1028,7 @@ define signext i32 @vwreduce_add_nxv4i16(<vscale x 4 x i16> %v) {
define signext i32 @vwreduce_uadd_nxv4i16(<vscale x 4 x i16> %v) {
; CHECK-LABEL: 'vwreduce_uadd_nxv4i16'
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %e = zext <vscale x 4 x i16> %v to <vscale x 4 x i32>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i32 @llvm.vector.reduce.add.nxv4i32(<vscale x 4 x i32> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %red = call i32 @llvm.vector.reduce.add.nxv4i32(<vscale x 4 x i32> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vwreduce_uadd_nxv4i16'
@@ -1043,7 +1043,7 @@ define signext i32 @vwreduce_uadd_nxv4i16(<vscale x 4 x i16> %v) {
define signext i32 @vreduce_umax_nxv4i32(<vscale x 4 x i32> %v) {
; CHECK-LABEL: 'vreduce_umax_nxv4i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i32 @llvm.vector.reduce.umax.nxv4i32(<vscale x 4 x i32> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i32 @llvm.vector.reduce.umax.nxv4i32(<vscale x 4 x i32> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vreduce_umax_nxv4i32'
@@ -1056,7 +1056,7 @@ define signext i32 @vreduce_umax_nxv4i32(<vscale x 4 x i32> %v) {
define signext i32 @vreduce_smax_nxv4i32(<vscale x 4 x i32> %v) {
; CHECK-LABEL: 'vreduce_smax_nxv4i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i32 @llvm.vector.reduce.smax.nxv4i32(<vscale x 4 x i32> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i32 @llvm.vector.reduce.smax.nxv4i32(<vscale x 4 x i32> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vreduce_smax_nxv4i32'
@@ -1069,7 +1069,7 @@ define signext i32 @vreduce_smax_nxv4i32(<vscale x 4 x i32> %v) {
define signext i32 @vreduce_umin_nxv4i32(<vscale x 4 x i32> %v) {
; CHECK-LABEL: 'vreduce_umin_nxv4i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i32 @llvm.vector.reduce.umin.nxv4i32(<vscale x 4 x i32> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i32 @llvm.vector.reduce.umin.nxv4i32(<vscale x 4 x i32> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vreduce_umin_nxv4i32'
@@ -1082,7 +1082,7 @@ define signext i32 @vreduce_umin_nxv4i32(<vscale x 4 x i32> %v) {
define signext i32 @vreduce_smin_nxv4i32(<vscale x 4 x i32> %v) {
; CHECK-LABEL: 'vreduce_smin_nxv4i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i32 @llvm.vector.reduce.smin.nxv4i32(<vscale x 4 x i32> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i32 @llvm.vector.reduce.smin.nxv4i32(<vscale x 4 x i32> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vreduce_smin_nxv4i32'
@@ -1095,7 +1095,7 @@ define signext i32 @vreduce_smin_nxv4i32(<vscale x 4 x i32> %v) {
define signext i32 @vreduce_and_nxv4i32(<vscale x 4 x i32> %v) {
; CHECK-LABEL: 'vreduce_and_nxv4i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i32 @llvm.vector.reduce.and.nxv4i32(<vscale x 4 x i32> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i32 @llvm.vector.reduce.and.nxv4i32(<vscale x 4 x i32> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vreduce_and_nxv4i32'
@@ -1108,7 +1108,7 @@ define signext i32 @vreduce_and_nxv4i32(<vscale x 4 x i32> %v) {
define signext i32 @vreduce_or_nxv4i32(<vscale x 4 x i32> %v) {
; CHECK-LABEL: 'vreduce_or_nxv4i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i32 @llvm.vector.reduce.or.nxv4i32(<vscale x 4 x i32> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i32 @llvm.vector.reduce.or.nxv4i32(<vscale x 4 x i32> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vreduce_or_nxv4i32'
@@ -1121,7 +1121,7 @@ define signext i32 @vreduce_or_nxv4i32(<vscale x 4 x i32> %v) {
define signext i32 @vreduce_xor_nxv4i32(<vscale x 4 x i32> %v) {
; CHECK-LABEL: 'vreduce_xor_nxv4i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i32 @llvm.vector.reduce.xor.nxv4i32(<vscale x 4 x i32> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %red = call i32 @llvm.vector.reduce.xor.nxv4i32(<vscale x 4 x i32> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 %red
;
; SIZE-LABEL: 'vreduce_xor_nxv4i32'
@@ -1134,7 +1134,7 @@ define signext i32 @vreduce_xor_nxv4i32(<vscale x 4 x i32> %v) {
define i64 @vreduce_add_nxv1i64(<vscale x 1 x i64> %v) {
; CHECK-LABEL: 'vreduce_add_nxv1i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i64 @llvm.vector.reduce.add.nxv1i64(<vscale x 1 x i64> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i64 @llvm.vector.reduce.add.nxv1i64(<vscale x 1 x i64> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vreduce_add_nxv1i64'
@@ -1148,7 +1148,7 @@ define i64 @vreduce_add_nxv1i64(<vscale x 1 x i64> %v) {
define i64 @vwreduce_add_nxv1i32(<vscale x 1 x i32> %v) {
; CHECK-LABEL: 'vwreduce_add_nxv1i32'
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %e = sext <vscale x 1 x i32> %v to <vscale x 1 x i64>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i64 @llvm.vector.reduce.add.nxv1i64(<vscale x 1 x i64> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i64 @llvm.vector.reduce.add.nxv1i64(<vscale x 1 x i64> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vwreduce_add_nxv1i32'
@@ -1164,7 +1164,7 @@ define i64 @vwreduce_add_nxv1i32(<vscale x 1 x i32> %v) {
define i64 @vwreduce_uadd_nxv1i32(<vscale x 1 x i32> %v) {
; CHECK-LABEL: 'vwreduce_uadd_nxv1i32'
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %e = zext <vscale x 1 x i32> %v to <vscale x 1 x i64>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i64 @llvm.vector.reduce.add.nxv1i64(<vscale x 1 x i64> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i64 @llvm.vector.reduce.add.nxv1i64(<vscale x 1 x i64> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vwreduce_uadd_nxv1i32'
@@ -1179,7 +1179,7 @@ define i64 @vwreduce_uadd_nxv1i32(<vscale x 1 x i32> %v) {
define i64 @vreduce_umax_nxv1i64(<vscale x 1 x i64> %v) {
; CHECK-LABEL: 'vreduce_umax_nxv1i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call i64 @llvm.vector.reduce.umax.nxv1i64(<vscale x 1 x i64> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i64 @llvm.vector.reduce.umax.nxv1i64(<vscale x 1 x i64> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vreduce_umax_nxv1i64'
@@ -1192,7 +1192,7 @@ define i64 @vreduce_umax_nxv1i64(<vscale x 1 x i64> %v) {
define i64 @vreduce_smax_nxv1i64(<vscale x 1 x i64> %v) {
; CHECK-LABEL: 'vreduce_smax_nxv1i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call i64 @llvm.vector.reduce.smax.nxv1i64(<vscale x 1 x i64> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i64 @llvm.vector.reduce.smax.nxv1i64(<vscale x 1 x i64> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vreduce_smax_nxv1i64'
@@ -1205,7 +1205,7 @@ define i64 @vreduce_smax_nxv1i64(<vscale x 1 x i64> %v) {
define i64 @vreduce_umin_nxv1i64(<vscale x 1 x i64> %v) {
; CHECK-LABEL: 'vreduce_umin_nxv1i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call i64 @llvm.vector.reduce.umin.nxv1i64(<vscale x 1 x i64> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i64 @llvm.vector.reduce.umin.nxv1i64(<vscale x 1 x i64> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vreduce_umin_nxv1i64'
@@ -1218,7 +1218,7 @@ define i64 @vreduce_umin_nxv1i64(<vscale x 1 x i64> %v) {
define i64 @vreduce_smin_nxv1i64(<vscale x 1 x i64> %v) {
; CHECK-LABEL: 'vreduce_smin_nxv1i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call i64 @llvm.vector.reduce.smin.nxv1i64(<vscale x 1 x i64> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i64 @llvm.vector.reduce.smin.nxv1i64(<vscale x 1 x i64> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vreduce_smin_nxv1i64'
@@ -1231,7 +1231,7 @@ define i64 @vreduce_smin_nxv1i64(<vscale x 1 x i64> %v) {
define i64 @vreduce_and_nxv1i64(<vscale x 1 x i64> %v) {
; CHECK-LABEL: 'vreduce_and_nxv1i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call i64 @llvm.vector.reduce.and.nxv1i64(<vscale x 1 x i64> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i64 @llvm.vector.reduce.and.nxv1i64(<vscale x 1 x i64> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vreduce_and_nxv1i64'
@@ -1244,7 +1244,7 @@ define i64 @vreduce_and_nxv1i64(<vscale x 1 x i64> %v) {
define i64 @vreduce_or_nxv1i64(<vscale x 1 x i64> %v) {
; CHECK-LABEL: 'vreduce_or_nxv1i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %red = call i64 @llvm.vector.reduce.or.nxv1i64(<vscale x 1 x i64> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i64 @llvm.vector.reduce.or.nxv1i64(<vscale x 1 x i64> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vreduce_or_nxv1i64'
@@ -1257,7 +1257,7 @@ define i64 @vreduce_or_nxv1i64(<vscale x 1 x i64> %v) {
define i64 @vreduce_xor_nxv1i64(<vscale x 1 x i64> %v) {
; CHECK-LABEL: 'vreduce_xor_nxv1i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i64 @llvm.vector.reduce.xor.nxv1i64(<vscale x 1 x i64> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i64 @llvm.vector.reduce.xor.nxv1i64(<vscale x 1 x i64> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vreduce_xor_nxv1i64'
@@ -1270,7 +1270,7 @@ define i64 @vreduce_xor_nxv1i64(<vscale x 1 x i64> %v) {
define i64 @vreduce_add_nxv2i64(<vscale x 2 x i64> %v) {
; CHECK-LABEL: 'vreduce_add_nxv2i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i64 @llvm.vector.reduce.add.nxv2i64(<vscale x 2 x i64> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i64 @llvm.vector.reduce.add.nxv2i64(<vscale x 2 x i64> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vreduce_add_nxv2i64'
@@ -1284,7 +1284,7 @@ define i64 @vreduce_add_nxv2i64(<vscale x 2 x i64> %v) {
define i64 @vwreduce_add_nxv2i32(<vscale x 2 x i32> %v) {
; CHECK-LABEL: 'vwreduce_add_nxv2i32'
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %e = sext <vscale x 2 x i32> %v to <vscale x 2 x i64>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i64 @llvm.vector.reduce.add.nxv2i64(<vscale x 2 x i64> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i64 @llvm.vector.reduce.add.nxv2i64(<vscale x 2 x i64> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vwreduce_add_nxv2i32'
@@ -1300,7 +1300,7 @@ define i64 @vwreduce_add_nxv2i32(<vscale x 2 x i32> %v) {
define i64 @vwreduce_uadd_nxv2i32(<vscale x 2 x i32> %v) {
; CHECK-LABEL: 'vwreduce_uadd_nxv2i32'
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %e = zext <vscale x 2 x i32> %v to <vscale x 2 x i64>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i64 @llvm.vector.reduce.add.nxv2i64(<vscale x 2 x i64> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i64 @llvm.vector.reduce.add.nxv2i64(<vscale x 2 x i64> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vwreduce_uadd_nxv2i32'
@@ -1315,7 +1315,7 @@ define i64 @vwreduce_uadd_nxv2i32(<vscale x 2 x i32> %v) {
define i64 @vreduce_umax_nxv2i64(<vscale x 2 x i64> %v) {
; CHECK-LABEL: 'vreduce_umax_nxv2i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i64 @llvm.vector.reduce.umax.nxv2i64(<vscale x 2 x i64> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i64 @llvm.vector.reduce.umax.nxv2i64(<vscale x 2 x i64> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vreduce_umax_nxv2i64'
@@ -1328,7 +1328,7 @@ define i64 @vreduce_umax_nxv2i64(<vscale x 2 x i64> %v) {
define i64 @vreduce_smax_nxv2i64(<vscale x 2 x i64> %v) {
; CHECK-LABEL: 'vreduce_smax_nxv2i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i64 @llvm.vector.reduce.smax.nxv2i64(<vscale x 2 x i64> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i64 @llvm.vector.reduce.smax.nxv2i64(<vscale x 2 x i64> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vreduce_smax_nxv2i64'
@@ -1341,7 +1341,7 @@ define i64 @vreduce_smax_nxv2i64(<vscale x 2 x i64> %v) {
define i64 @vreduce_umin_nxv2i64(<vscale x 2 x i64> %v) {
; CHECK-LABEL: 'vreduce_umin_nxv2i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i64 @llvm.vector.reduce.umin.nxv2i64(<vscale x 2 x i64> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i64 @llvm.vector.reduce.umin.nxv2i64(<vscale x 2 x i64> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vreduce_umin_nxv2i64'
@@ -1354,7 +1354,7 @@ define i64 @vreduce_umin_nxv2i64(<vscale x 2 x i64> %v) {
define i64 @vreduce_smin_nxv2i64(<vscale x 2 x i64> %v) {
; CHECK-LABEL: 'vreduce_smin_nxv2i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i64 @llvm.vector.reduce.smin.nxv2i64(<vscale x 2 x i64> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i64 @llvm.vector.reduce.smin.nxv2i64(<vscale x 2 x i64> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vreduce_smin_nxv2i64'
@@ -1367,7 +1367,7 @@ define i64 @vreduce_smin_nxv2i64(<vscale x 2 x i64> %v) {
define i64 @vreduce_and_nxv2i64(<vscale x 2 x i64> %v) {
; CHECK-LABEL: 'vreduce_and_nxv2i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i64 @llvm.vector.reduce.and.nxv2i64(<vscale x 2 x i64> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i64 @llvm.vector.reduce.and.nxv2i64(<vscale x 2 x i64> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vreduce_and_nxv2i64'
@@ -1380,7 +1380,7 @@ define i64 @vreduce_and_nxv2i64(<vscale x 2 x i64> %v) {
define i64 @vreduce_or_nxv2i64(<vscale x 2 x i64> %v) {
; CHECK-LABEL: 'vreduce_or_nxv2i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %red = call i64 @llvm.vector.reduce.or.nxv2i64(<vscale x 2 x i64> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i64 @llvm.vector.reduce.or.nxv2i64(<vscale x 2 x i64> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vreduce_or_nxv2i64'
@@ -1393,7 +1393,7 @@ define i64 @vreduce_or_nxv2i64(<vscale x 2 x i64> %v) {
define i64 @vreduce_xor_nxv2i64(<vscale x 2 x i64> %v) {
; CHECK-LABEL: 'vreduce_xor_nxv2i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i64 @llvm.vector.reduce.xor.nxv2i64(<vscale x 2 x i64> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i64 @llvm.vector.reduce.xor.nxv2i64(<vscale x 2 x i64> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vreduce_xor_nxv2i64'
@@ -1406,7 +1406,7 @@ define i64 @vreduce_xor_nxv2i64(<vscale x 2 x i64> %v) {
define i64 @vreduce_add_nxv4i64(<vscale x 4 x i64> %v) {
; CHECK-LABEL: 'vreduce_add_nxv4i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i64 @llvm.vector.reduce.add.nxv4i64(<vscale x 4 x i64> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %red = call i64 @llvm.vector.reduce.add.nxv4i64(<vscale x 4 x i64> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vreduce_add_nxv4i64'
@@ -1420,7 +1420,7 @@ define i64 @vreduce_add_nxv4i64(<vscale x 4 x i64> %v) {
define i64 @vwreduce_add_nxv4i32(<vscale x 4 x i32> %v) {
; CHECK-LABEL: 'vwreduce_add_nxv4i32'
; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %e = sext <vscale x 4 x i32> %v to <vscale x 4 x i64>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i64 @llvm.vector.reduce.add.nxv4i64(<vscale x 4 x i64> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %red = call i64 @llvm.vector.reduce.add.nxv4i64(<vscale x 4 x i64> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vwreduce_add_nxv4i32'
@@ -1436,7 +1436,7 @@ define i64 @vwreduce_add_nxv4i32(<vscale x 4 x i32> %v) {
define i64 @vwreduce_uadd_nxv4i32(<vscale x 4 x i32> %v) {
; CHECK-LABEL: 'vwreduce_uadd_nxv4i32'
; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %e = zext <vscale x 4 x i32> %v to <vscale x 4 x i64>
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i64 @llvm.vector.reduce.add.nxv4i64(<vscale x 4 x i64> %e)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %red = call i64 @llvm.vector.reduce.add.nxv4i64(<vscale x 4 x i64> %e)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vwreduce_uadd_nxv4i32'
@@ -1451,7 +1451,7 @@ define i64 @vwreduce_uadd_nxv4i32(<vscale x 4 x i32> %v) {
define i64 @vreduce_umax_nxv4i64(<vscale x 4 x i64> %v) {
; CHECK-LABEL: 'vreduce_umax_nxv4i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i64 @llvm.vector.reduce.umax.nxv4i64(<vscale x 4 x i64> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i64 @llvm.vector.reduce.umax.nxv4i64(<vscale x 4 x i64> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vreduce_umax_nxv4i64'
@@ -1464,7 +1464,7 @@ define i64 @vreduce_umax_nxv4i64(<vscale x 4 x i64> %v) {
define i64 @vreduce_smax_nxv4i64(<vscale x 4 x i64> %v) {
; CHECK-LABEL: 'vreduce_smax_nxv4i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i64 @llvm.vector.reduce.smax.nxv4i64(<vscale x 4 x i64> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i64 @llvm.vector.reduce.smax.nxv4i64(<vscale x 4 x i64> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vreduce_smax_nxv4i64'
@@ -1477,7 +1477,7 @@ define i64 @vreduce_smax_nxv4i64(<vscale x 4 x i64> %v) {
define i64 @vreduce_umin_nxv4i64(<vscale x 4 x i64> %v) {
; CHECK-LABEL: 'vreduce_umin_nxv4i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i64 @llvm.vector.reduce.umin.nxv4i64(<vscale x 4 x i64> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i64 @llvm.vector.reduce.umin.nxv4i64(<vscale x 4 x i64> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vreduce_umin_nxv4i64'
@@ -1490,7 +1490,7 @@ define i64 @vreduce_umin_nxv4i64(<vscale x 4 x i64> %v) {
define i64 @vreduce_smin_nxv4i64(<vscale x 4 x i64> %v) {
; CHECK-LABEL: 'vreduce_smin_nxv4i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i64 @llvm.vector.reduce.smin.nxv4i64(<vscale x 4 x i64> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i64 @llvm.vector.reduce.smin.nxv4i64(<vscale x 4 x i64> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vreduce_smin_nxv4i64'
@@ -1503,7 +1503,7 @@ define i64 @vreduce_smin_nxv4i64(<vscale x 4 x i64> %v) {
define i64 @vreduce_and_nxv4i64(<vscale x 4 x i64> %v) {
; CHECK-LABEL: 'vreduce_and_nxv4i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i64 @llvm.vector.reduce.and.nxv4i64(<vscale x 4 x i64> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i64 @llvm.vector.reduce.and.nxv4i64(<vscale x 4 x i64> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vreduce_and_nxv4i64'
@@ -1516,7 +1516,7 @@ define i64 @vreduce_and_nxv4i64(<vscale x 4 x i64> %v) {
define i64 @vreduce_or_nxv4i64(<vscale x 4 x i64> %v) {
; CHECK-LABEL: 'vreduce_or_nxv4i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %red = call i64 @llvm.vector.reduce.or.nxv4i64(<vscale x 4 x i64> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i64 @llvm.vector.reduce.or.nxv4i64(<vscale x 4 x i64> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vreduce_or_nxv4i64'
@@ -1529,7 +1529,7 @@ define i64 @vreduce_or_nxv4i64(<vscale x 4 x i64> %v) {
define i64 @vreduce_xor_nxv4i64(<vscale x 4 x i64> %v) {
; CHECK-LABEL: 'vreduce_xor_nxv4i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %red = call i64 @llvm.vector.reduce.xor.nxv4i64(<vscale x 4 x i64> %v)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %red = call i64 @llvm.vector.reduce.xor.nxv4i64(<vscale x 4 x i64> %v)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i64 %red
;
; SIZE-LABEL: 'vreduce_xor_nxv4i64'
diff --git a/llvm/test/Analysis/CostModel/RISCV/reduce-xor.ll b/llvm/test/Analysis/CostModel/RISCV/reduce-xor.ll
index abc5335afbc98..92062f4c5f02c 100644
--- a/llvm/test/Analysis/CostModel/RISCV/reduce-xor.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/reduce-xor.ll
@@ -40,14 +40,14 @@ define i32 @reduce_i1(i32 %arg) {
define i32 @reduce_i8(i32 %arg) {
; CHECK-LABEL: 'reduce_i8'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i8 @llvm.vector.reduce.xor.v1i8(<1 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.vector.reduce.xor.v2i8(<2 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i8 @llvm.vector.reduce.xor.v4i8(<4 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i8 @llvm.vector.reduce.xor.v8i8(<8 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i8 @llvm.vector.reduce.xor.v16i8(<16 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i8 @llvm.vector.reduce.xor.v32i8(<32 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i8 @llvm.vector.reduce.xor.v64i8(<64 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V128 = call i8 @llvm.vector.reduce.xor.v128i8(<128 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1 = call i8 @llvm.vector.reduce.xor.v1i8(<1 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i8 @llvm.vector.reduce.xor.v2i8(<2 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i8 @llvm.vector.reduce.xor.v4i8(<4 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i8 @llvm.vector.reduce.xor.v8i8(<8 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V16 = call i8 @llvm.vector.reduce.xor.v16i8(<16 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i8 @llvm.vector.reduce.xor.v32i8(<32 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V64 = call i8 @llvm.vector.reduce.xor.v64i8(<64 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %V128 = call i8 @llvm.vector.reduce.xor.v128i8(<128 x i8> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_i8'
@@ -74,14 +74,14 @@ define i32 @reduce_i8(i32 %arg) {
define i32 @reduce_i16(i32 %arg) {
; CHECK-LABEL: 'reduce_i16'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i16 @llvm.vector.reduce.xor.v1i16(<1 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.vector.reduce.xor.v2i16(<2 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.vector.reduce.xor.v4i16(<4 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i16 @llvm.vector.reduce.xor.v8i16(<8 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i16 @llvm.vector.reduce.xor.v16i16(<16 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i16 @llvm.vector.reduce.xor.v32i16(<32 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.vector.reduce.xor.v64i16(<64 x i16> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i16 @llvm.vector.reduce.xor.v128i16(<128 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1 = call i16 @llvm.vector.reduce.xor.v1i16(<1 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i16 @llvm.vector.reduce.xor.v2i16(<2 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i16 @llvm.vector.reduce.xor.v4i16(<4 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i16 @llvm.vector.reduce.xor.v8i16(<8 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V16 = call i16 @llvm.vector.reduce.xor.v16i16(<16 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i16 @llvm.vector.reduce.xor.v32i16(<32 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V64 = call i16 @llvm.vector.reduce.xor.v64i16(<64 x i16> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %V128 = call i16 @llvm.vector.reduce.xor.v128i16(<128 x i16> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_i16'
@@ -108,14 +108,14 @@ define i32 @reduce_i16(i32 %arg) {
define i32 @reduce_i32(i32 %arg) {
; CHECK-LABEL: 'reduce_i32'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i32 @llvm.vector.reduce.xor.v1i32(<1 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.vector.reduce.xor.v2i32(<2 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i32 @llvm.vector.reduce.xor.v4i32(<4 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i32 @llvm.vector.reduce.xor.v8i32(<8 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i32 @llvm.vector.reduce.xor.v16i32(<16 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i32 @llvm.vector.reduce.xor.v32i32(<32 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V64 = call i32 @llvm.vector.reduce.xor.v64i32(<64 x i32> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V128 = call i32 @llvm.vector.reduce.xor.v128i32(<128 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1 = call i32 @llvm.vector.reduce.xor.v1i32(<1 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i32 @llvm.vector.reduce.xor.v2i32(<2 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i32 @llvm.vector.reduce.xor.v4i32(<4 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.vector.reduce.xor.v8i32(<8 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V16 = call i32 @llvm.vector.reduce.xor.v16i32(<16 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V32 = call i32 @llvm.vector.reduce.xor.v32i32(<32 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V64 = call i32 @llvm.vector.reduce.xor.v64i32(<64 x i32> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 32 for instruction: %V128 = call i32 @llvm.vector.reduce.xor.v128i32(<128 x i32> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_i32'
@@ -142,14 +142,14 @@ define i32 @reduce_i32(i32 %arg) {
define i32 @reduce_i64(i32 %arg) {
; CHECK-LABEL: 'reduce_i64'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i64 @llvm.vector.reduce.xor.v1i64(<1 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.vector.reduce.xor.v2i64(<2 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.vector.reduce.xor.v4i64(<4 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i64 @llvm.vector.reduce.xor.v8i64(<8 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i64 @llvm.vector.reduce.xor.v16i64(<16 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V32 = call i64 @llvm.vector.reduce.xor.v32i64(<32 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %V64 = call i64 @llvm.vector.reduce.xor.v64i64(<64 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 62 for instruction: %V128 = call i64 @llvm.vector.reduce.xor.v128i64(<128 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V1 = call i64 @llvm.vector.reduce.xor.v1i64(<1 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2 = call i64 @llvm.vector.reduce.xor.v2i64(<2 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V4 = call i64 @llvm.vector.reduce.xor.v4i64(<4 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i64 @llvm.vector.reduce.xor.v8i64(<8 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V16 = call i64 @llvm.vector.reduce.xor.v16i64(<16 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V32 = call i64 @llvm.vector.reduce.xor.v32i64(<32 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 31 for instruction: %V64 = call i64 @llvm.vector.reduce.xor.v64i64(<64 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 63 for instruction: %V128 = call i64 @llvm.vector.reduce.xor.v128i64(<128 x i64> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; SIZE-LABEL: 'reduce_i64'
diff --git a/llvm/test/Analysis/CostModel/RISCV/rvv-expandload-compressstore.ll b/llvm/test/Analysis/CostModel/RISCV/rvv-expandload-compressstore.ll
index 2d43b48b3b1b3..5d39f64fecd95 100644
--- a/llvm/test/Analysis/CostModel/RISCV/rvv-expandload-compressstore.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/rvv-expandload-compressstore.ll
@@ -12,10 +12,10 @@ define void @expand_load() {
; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %t6 = call <4 x i64> @llvm.masked.expandload.v4i64.p0(ptr align 8 poison, <4 x i1> poison, <4 x i64> poison)
; CHECK-NEXT: Cost Model: Found an estimated cost of 32 for instruction: %t7 = call <8 x i64> @llvm.masked.expandload.v8i64.p0(ptr align 8 poison, <8 x i1> poison, <8 x i64> poison)
; CHECK-NEXT: Cost Model: Found an estimated cost of 96 for instruction: %t8 = call <16 x i64> @llvm.masked.expandload.v16i64.p0(ptr align 8 poison, <16 x i1> poison, <16 x i64> poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %t9 = call <2 x i64> @llvm.masked.expandload.v2i64.p0(ptr poison, <2 x i1> poison, <2 x i64> poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %t10 = call <4 x i64> @llvm.masked.expandload.v4i64.p0(ptr poison, <4 x i1> poison, <4 x i64> poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 38 for instruction: %t11 = call <8 x i64> @llvm.masked.expandload.v8i64.p0(ptr poison, <8 x i1> poison, <8 x i64> poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 78 for instruction: %t12 = call <16 x i64> @llvm.masked.expandload.v16i64.p0(ptr poison, <16 x i1> poison, <16 x i64> poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %t9 = call <2 x i64> @llvm.masked.expandload.v2i64.p0(ptr poison, <2 x i1> poison, <2 x i64> poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %t10 = call <4 x i64> @llvm.masked.expandload.v4i64.p0(ptr poison, <4 x i1> poison, <4 x i64> poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 46 for instruction: %t11 = call <8 x i64> @llvm.masked.expandload.v8i64.p0(ptr poison, <8 x i1> poison, <8 x i64> poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 94 for instruction: %t12 = call <16 x i64> @llvm.masked.expandload.v16i64.p0(ptr poison, <16 x i1> poison, <16 x i64> poison)
; CHECK-NEXT: Cost Model: Invalid cost for instruction: %t13 = call <vscale x 2 x i8> @llvm.masked.expandload.nxv2i8.p0(ptr poison, <vscale x 2 x i1> poison, <vscale x 2 x i8> poison)
; CHECK-NEXT: Cost Model: Invalid cost for instruction: %t14 = call <vscale x 4 x i8> @llvm.masked.expandload.nxv4i8.p0(ptr poison, <vscale x 4 x i1> poison, <vscale x 4 x i8> poison)
; CHECK-NEXT: Cost Model: Invalid cost for instruction: %t15 = call <vscale x 8 x i8> @llvm.masked.expandload.nxv8i8.p0(ptr poison, <vscale x 8 x i1> poison, <vscale x 8 x i8> poison)
@@ -27,18 +27,18 @@ define void @expand_load() {
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; TYPEBASED-LABEL: 'expand_load'
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %t1 = call <2 x i8> @llvm.masked.expandload.v2i8.p0(ptr poison, <2 x i1> poison, <2 x i8> poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %t2 = call <4 x i8> @llvm.masked.expandload.v4i8.p0(ptr poison, <4 x i1> poison, <4 x i8> poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 62 for instruction: %t3 = call <8 x i8> @llvm.masked.expandload.v8i8.p0(ptr poison, <8 x i1> poison, <8 x i8> poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 126 for instruction: %t4 = call <16 x i8> @llvm.masked.expandload.v16i8.p0(ptr poison, <16 x i1> poison, <16 x i8> poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %t5 = call <2 x i64> @llvm.masked.expandload.v2i64.p0(ptr align 8 poison, <2 x i1> poison, <2 x i64> poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 33 for instruction: %t6 = call <4 x i64> @llvm.masked.expandload.v4i64.p0(ptr align 8 poison, <4 x i1> poison, <4 x i64> poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 69 for instruction: %t7 = call <8 x i64> @llvm.masked.expandload.v8i64.p0(ptr align 8 poison, <8 x i1> poison, <8 x i64> poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 141 for instruction: %t8 = call <16 x i64> @llvm.masked.expandload.v16i64.p0(ptr align 8 poison, <16 x i1> poison, <16 x i64> poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %t9 = call <2 x i64> @llvm.masked.expandload.v2i64.p0(ptr poison, <2 x i1> poison, <2 x i64> poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 33 for instruction: %t10 = call <4 x i64> @llvm.masked.expandload.v4i64.p0(ptr poison, <4 x i1> poison, <4 x i64> poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 69 for instruction: %t11 = call <8 x i64> @llvm.masked.expandload.v8i64.p0(ptr poison, <8 x i1> poison, <8 x i64> poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 141 for instruction: %t12 = call <16 x i64> @llvm.masked.expandload.v16i64.p0(ptr poison, <16 x i1> poison, <16 x i64> poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %t1 = call <2 x i8> @llvm.masked.expandload.v2i8.p0(ptr poison, <2 x i1> poison, <2 x i8> poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 38 for instruction: %t2 = call <4 x i8> @llvm.masked.expandload.v4i8.p0(ptr poison, <4 x i1> poison, <4 x i8> poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 78 for instruction: %t3 = call <8 x i8> @llvm.masked.expandload.v8i8.p0(ptr poison, <8 x i1> poison, <8 x i8> poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 158 for instruction: %t4 = call <16 x i8> @llvm.masked.expandload.v16i8.p0(ptr poison, <16 x i1> poison, <16 x i8> poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %t5 = call <2 x i64> @llvm.masked.expandload.v2i64.p0(ptr align 8 poison, <2 x i1> poison, <2 x i64> poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 41 for instruction: %t6 = call <4 x i64> @llvm.masked.expandload.v4i64.p0(ptr align 8 poison, <4 x i1> poison, <4 x i64> poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 85 for instruction: %t7 = call <8 x i64> @llvm.masked.expandload.v8i64.p0(ptr align 8 poison, <8 x i1> poison, <8 x i64> poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 173 for instruction: %t8 = call <16 x i64> @llvm.masked.expandload.v16i64.p0(ptr align 8 poison, <16 x i1> poison, <16 x i64> poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %t9 = call <2 x i64> @llvm.masked.expandload.v2i64.p0(ptr poison, <2 x i1> poison, <2 x i64> poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 41 for instruction: %t10 = call <4 x i64> @llvm.masked.expandload.v4i64.p0(ptr poison, <4 x i1> poison, <4 x i64> poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 85 for instruction: %t11 = call <8 x i64> @llvm.masked.expandload.v8i64.p0(ptr poison, <8 x i1> poison, <8 x i64> poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 173 for instruction: %t12 = call <16 x i64> @llvm.masked.expandload.v16i64.p0(ptr poison, <16 x i1> poison, <16 x i64> poison)
; TYPEBASED-NEXT: Cost Model: Invalid cost for instruction: %t13 = call <vscale x 2 x i8> @llvm.masked.expandload.nxv2i8.p0(ptr poison, <vscale x 2 x i1> poison, <vscale x 2 x i8> poison)
; TYPEBASED-NEXT: Cost Model: Invalid cost for instruction: %t14 = call <vscale x 4 x i8> @llvm.masked.expandload.nxv4i8.p0(ptr poison, <vscale x 4 x i1> poison, <vscale x 4 x i8> poison)
; TYPEBASED-NEXT: Cost Model: Invalid cost for instruction: %t15 = call <vscale x 8 x i8> @llvm.masked.expandload.nxv8i8.p0(ptr poison, <vscale x 8 x i1> poison, <vscale x 8 x i8> poison)
@@ -82,10 +82,10 @@ define void @compress_store() {
; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: call void @llvm.masked.compressstore.v4i64.p0(<4 x i64> poison, ptr align 8 poison, <4 x i1> poison)
; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: call void @llvm.masked.compressstore.v8i64.p0(<8 x i64> poison, ptr align 8 poison, <8 x i1> poison)
; CHECK-NEXT: Cost Model: Found an estimated cost of 24 for instruction: call void @llvm.masked.compressstore.v16i64.p0(<16 x i64> poison, ptr align 8 poison, <16 x i1> poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: call void @llvm.masked.compressstore.v2i64.p0(<2 x i64> poison, ptr poison, <2 x i1> poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: call void @llvm.masked.compressstore.v4i64.p0(<4 x i64> poison, ptr poison, <4 x i1> poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 38 for instruction: call void @llvm.masked.compressstore.v8i64.p0(<8 x i64> poison, ptr poison, <8 x i1> poison)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 78 for instruction: call void @llvm.masked.compressstore.v16i64.p0(<16 x i64> poison, ptr poison, <16 x i1> poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: call void @llvm.masked.compressstore.v2i64.p0(<2 x i64> poison, ptr poison, <2 x i1> poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 26 for instruction: call void @llvm.masked.compressstore.v4i64.p0(<4 x i64> poison, ptr poison, <4 x i1> poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 54 for instruction: call void @llvm.masked.compressstore.v8i64.p0(<8 x i64> poison, ptr poison, <8 x i1> poison)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 110 for instruction: call void @llvm.masked.compressstore.v16i64.p0(<16 x i64> poison, ptr poison, <16 x i1> poison)
; CHECK-NEXT: Cost Model: Invalid cost for instruction: call void @llvm.masked.compressstore.nxv2i8.p0(<vscale x 2 x i8> poison, ptr poison, <vscale x 2 x i1> poison)
; CHECK-NEXT: Cost Model: Invalid cost for instruction: call void @llvm.masked.compressstore.nxv4i8.p0(<vscale x 4 x i8> poison, ptr poison, <vscale x 4 x i1> poison)
; CHECK-NEXT: Cost Model: Invalid cost for instruction: call void @llvm.masked.compressstore.nxv8i8.p0(<vscale x 8 x i8> poison, ptr poison, <vscale x 8 x i1> poison)
@@ -97,18 +97,18 @@ define void @compress_store() {
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; TYPEBASED-LABEL: 'compress_store'
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 12 for instruction: call void @llvm.masked.compressstore.v2i8.p0(<2 x i8> poison, ptr poison, <2 x i1> poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 26 for instruction: call void @llvm.masked.compressstore.v4i8.p0(<4 x i8> poison, ptr poison, <4 x i1> poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 54 for instruction: call void @llvm.masked.compressstore.v8i8.p0(<8 x i8> poison, ptr poison, <8 x i1> poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 110 for instruction: call void @llvm.masked.compressstore.v16i8.p0(<16 x i8> poison, ptr poison, <16 x i1> poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 12 for instruction: call void @llvm.masked.compressstore.v2i64.p0(<2 x i64> poison, ptr align 8 poison, <2 x i1> poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 26 for instruction: call void @llvm.masked.compressstore.v4i64.p0(<4 x i64> poison, ptr align 8 poison, <4 x i1> poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 54 for instruction: call void @llvm.masked.compressstore.v8i64.p0(<8 x i64> poison, ptr align 8 poison, <8 x i1> poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 110 for instruction: call void @llvm.masked.compressstore.v16i64.p0(<16 x i64> poison, ptr align 8 poison, <16 x i1> poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 12 for instruction: call void @llvm.masked.compressstore.v2i64.p0(<2 x i64> poison, ptr poison, <2 x i1> poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 26 for instruction: call void @llvm.masked.compressstore.v4i64.p0(<4 x i64> poison, ptr poison, <4 x i1> poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 54 for instruction: call void @llvm.masked.compressstore.v8i64.p0(<8 x i64> poison, ptr poison, <8 x i1> poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 110 for instruction: call void @llvm.masked.compressstore.v16i64.p0(<16 x i64> poison, ptr poison, <16 x i1> poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 16 for instruction: call void @llvm.masked.compressstore.v2i8.p0(<2 x i8> poison, ptr poison, <2 x i1> poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 34 for instruction: call void @llvm.masked.compressstore.v4i8.p0(<4 x i8> poison, ptr poison, <4 x i1> poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 70 for instruction: call void @llvm.masked.compressstore.v8i8.p0(<8 x i8> poison, ptr poison, <8 x i1> poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 142 for instruction: call void @llvm.masked.compressstore.v16i8.p0(<16 x i8> poison, ptr poison, <16 x i1> poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 16 for instruction: call void @llvm.masked.compressstore.v2i64.p0(<2 x i64> poison, ptr align 8 poison, <2 x i1> poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 34 for instruction: call void @llvm.masked.compressstore.v4i64.p0(<4 x i64> poison, ptr align 8 poison, <4 x i1> poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 70 for instruction: call void @llvm.masked.compressstore.v8i64.p0(<8 x i64> poison, ptr align 8 poison, <8 x i1> poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 142 for instruction: call void @llvm.masked.compressstore.v16i64.p0(<16 x i64> poison, ptr align 8 poison, <16 x i1> poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 16 for instruction: call void @llvm.masked.compressstore.v2i64.p0(<2 x i64> poison, ptr poison, <2 x i1> poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 34 for instruction: call void @llvm.masked.compressstore.v4i64.p0(<4 x i64> poison, ptr poison, <4 x i1> poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 70 for instruction: call void @llvm.masked.compressstore.v8i64.p0(<8 x i64> poison, ptr poison, <8 x i1> poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 142 for instruction: call void @llvm.masked.compressstore.v16i64.p0(<16 x i64> poison, ptr poison, <16 x i1> poison)
; TYPEBASED-NEXT: Cost Model: Invalid cost for instruction: call void @llvm.masked.compressstore.nxv2i8.p0(<vscale x 2 x i8> poison, ptr poison, <vscale x 2 x i1> poison)
; TYPEBASED-NEXT: Cost Model: Invalid cost for instruction: call void @llvm.masked.compressstore.nxv4i8.p0(<vscale x 4 x i8> poison, ptr poison, <vscale x 4 x i1> poison)
; TYPEBASED-NEXT: Cost Model: Invalid cost for instruction: call void @llvm.masked.compressstore.nxv8i8.p0(<vscale x 8 x i8> poison, ptr poison, <vscale x 8 x i1> poison)
diff --git a/llvm/test/Analysis/CostModel/RISCV/rvv-extractelement.ll b/llvm/test/Analysis/CostModel/RISCV/rvv-extractelement.ll
index f3cbe25655a89..50704350e17fa 100644
--- a/llvm/test/Analysis/CostModel/RISCV/rvv-extractelement.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/rvv-extractelement.ll
@@ -8,52 +8,52 @@
define void @extractelement_int(i32 %x) {
; RV32V-LABEL: 'extractelement_int'
-; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i1_0 = extractelement <2 x i1> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i1_0 = extractelement <4 x i1> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i1_0 = extractelement <8 x i1> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i1_0 = extractelement <16 x i1> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v32i1_0 = extractelement <32 x i1> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i1_0 = extractelement <vscale x 2 x i1> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i1_0 = extractelement <vscale x 4 x i1> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i1_0 = extractelement <vscale x 8 x i1> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv16i1_0 = extractelement <vscale x 16 x i1> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %nxv32i1_0 = extractelement <vscale x 32 x i1> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v2i8_0 = extractelement <2 x i8> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v4i8_0 = extractelement <4 x i8> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v8i8_0 = extractelement <8 x i8> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v16i8_0 = extractelement <16 x i8> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i8_0 = extractelement <32 x i8> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v64i8_0 = extractelement <64 x i8> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v128i8_0 = extractelement <128 x i8> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv2i8_0 = extractelement <vscale x 2 x i8> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv4i8_0 = extractelement <vscale x 4 x i8> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv8i8_0 = extractelement <vscale x 8 x i8> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv16i8_0 = extractelement <vscale x 16 x i8> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv32i8_0 = extractelement <vscale x 32 x i8> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv64i8_0 = extractelement <vscale x 64 x i8> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv128i8_0 = extractelement <vscale x 128 x i8> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v2i16_0 = extractelement <2 x i16> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v4i16_0 = extractelement <4 x i16> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v8i16_0 = extractelement <8 x i16> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v16i16_0 = extractelement <16 x i16> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i16_0 = extractelement <32 x i16> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v64i16_0 = extractelement <64 x i16> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv2i16_0 = extractelement <vscale x 2 x i16> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv4i16_0 = extractelement <vscale x 4 x i16> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv8i16_0 = extractelement <vscale x 8 x i16> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv16i16_0 = extractelement <vscale x 16 x i16> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv32i16_0 = extractelement <vscale x 32 x i16> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv64i16_0 = extractelement <vscale x 64 x i16> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v2i32_0 = extractelement <2 x i32> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v4i32_0 = extractelement <4 x i32> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v8i32_0 = extractelement <8 x i32> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v16i32_0 = extractelement <16 x i32> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i32_0 = extractelement <32 x i32> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv2i32_0 = extractelement <vscale x 2 x i32> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv4i32_0 = extractelement <vscale x 4 x i32> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv8i32_0 = extractelement <vscale x 8 x i32> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv16i32_0 = extractelement <vscale x 16 x i32> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv32i32_0 = extractelement <vscale x 32 x i32> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v2i1_0 = extractelement <2 x i1> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v4i1_0 = extractelement <4 x i1> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v8i1_0 = extractelement <8 x i1> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v16i1_0 = extractelement <16 x i1> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v32i1_0 = extractelement <32 x i1> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv2i1_0 = extractelement <vscale x 2 x i1> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv4i1_0 = extractelement <vscale x 4 x i1> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv8i1_0 = extractelement <vscale x 8 x i1> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %nxv16i1_0 = extractelement <vscale x 16 x i1> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %nxv32i1_0 = extractelement <vscale x 32 x i1> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i8_0 = extractelement <2 x i8> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i8_0 = extractelement <4 x i8> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i8_0 = extractelement <8 x i8> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i8_0 = extractelement <16 x i8> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i8_0 = extractelement <32 x i8> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i8_0 = extractelement <64 x i8> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v128i8_0 = extractelement <128 x i8> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i8_0 = extractelement <vscale x 2 x i8> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i8_0 = extractelement <vscale x 4 x i8> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i8_0 = extractelement <vscale x 8 x i8> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i8_0 = extractelement <vscale x 16 x i8> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i8_0 = extractelement <vscale x 32 x i8> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv64i8_0 = extractelement <vscale x 64 x i8> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv128i8_0 = extractelement <vscale x 128 x i8> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i16_0 = extractelement <2 x i16> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i16_0 = extractelement <4 x i16> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i16_0 = extractelement <8 x i16> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i16_0 = extractelement <16 x i16> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i16_0 = extractelement <32 x i16> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i16_0 = extractelement <64 x i16> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i16_0 = extractelement <vscale x 2 x i16> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i16_0 = extractelement <vscale x 4 x i16> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i16_0 = extractelement <vscale x 8 x i16> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i16_0 = extractelement <vscale x 16 x i16> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i16_0 = extractelement <vscale x 32 x i16> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv64i16_0 = extractelement <vscale x 64 x i16> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i32_0 = extractelement <2 x i32> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i32_0 = extractelement <4 x i32> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i32_0 = extractelement <8 x i32> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i32_0 = extractelement <16 x i32> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_0 = extractelement <32 x i32> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i32_0 = extractelement <vscale x 2 x i32> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i32_0 = extractelement <vscale x 4 x i32> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i32_0 = extractelement <vscale x 8 x i32> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i32_0 = extractelement <vscale x 16 x i32> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i32_0 = extractelement <vscale x 32 x i32> undef, i32 0
; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v2i64_0 = extractelement <2 x i64> undef, i32 0
; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v4i64_0 = extractelement <4 x i64> undef, i32 0
; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v8i64_0 = extractelement <8 x i64> undef, i32 0
@@ -62,52 +62,52 @@ define void @extractelement_int(i32 %x) {
; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv4i64_0 = extractelement <vscale x 4 x i64> undef, i32 0
; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv8i64_0 = extractelement <vscale x 8 x i64> undef, i32 0
; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv16i64_0 = extractelement <vscale x 16 x i64> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v2i1_1 = extractelement <2 x i1> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v4i1_1 = extractelement <4 x i1> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v8i1_1 = extractelement <8 x i1> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v16i1_1 = extractelement <16 x i1> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v32i1_1 = extractelement <32 x i1> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv2i1_1 = extractelement <vscale x 2 x i1> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv4i1_1 = extractelement <vscale x 4 x i1> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv8i1_1 = extractelement <vscale x 8 x i1> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %nxv16i1_1 = extractelement <vscale x 16 x i1> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %nxv32i1_1 = extractelement <vscale x 32 x i1> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i8_1 = extractelement <2 x i8> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i8_1 = extractelement <4 x i8> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i8_1 = extractelement <8 x i8> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i8_1 = extractelement <16 x i8> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i8_1 = extractelement <32 x i8> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i8_1 = extractelement <64 x i8> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v128i8_1 = extractelement <128 x i8> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i8_1 = extractelement <vscale x 2 x i8> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i8_1 = extractelement <vscale x 4 x i8> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i8_1 = extractelement <vscale x 8 x i8> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i8_1 = extractelement <vscale x 16 x i8> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i8_1 = extractelement <vscale x 32 x i8> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv64i8_1 = extractelement <vscale x 64 x i8> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv128i8_1 = extractelement <vscale x 128 x i8> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i16_1 = extractelement <2 x i16> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i16_1 = extractelement <4 x i16> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i16_1 = extractelement <8 x i16> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i16_1 = extractelement <16 x i16> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i16_1 = extractelement <32 x i16> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i16_1 = extractelement <64 x i16> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i16_1 = extractelement <vscale x 2 x i16> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i16_1 = extractelement <vscale x 4 x i16> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i16_1 = extractelement <vscale x 8 x i16> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i16_1 = extractelement <vscale x 16 x i16> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i16_1 = extractelement <vscale x 32 x i16> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv64i16_1 = extractelement <vscale x 64 x i16> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i32_1 = extractelement <2 x i32> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i32_1 = extractelement <4 x i32> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i32_1 = extractelement <8 x i32> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i32_1 = extractelement <16 x i32> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_1 = extractelement <32 x i32> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i32_1 = extractelement <vscale x 2 x i32> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i32_1 = extractelement <vscale x 4 x i32> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i32_1 = extractelement <vscale x 8 x i32> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i32_1 = extractelement <vscale x 16 x i32> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i32_1 = extractelement <vscale x 32 x i32> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v2i1_1 = extractelement <2 x i1> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v4i1_1 = extractelement <4 x i1> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v8i1_1 = extractelement <8 x i1> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v16i1_1 = extractelement <16 x i1> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %v32i1_1 = extractelement <32 x i1> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv2i1_1 = extractelement <vscale x 2 x i1> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv4i1_1 = extractelement <vscale x 4 x i1> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv8i1_1 = extractelement <vscale x 8 x i1> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %nxv16i1_1 = extractelement <vscale x 16 x i1> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %nxv32i1_1 = extractelement <vscale x 32 x i1> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i8_1 = extractelement <2 x i8> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i8_1 = extractelement <4 x i8> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i8_1 = extractelement <8 x i8> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i8_1 = extractelement <16 x i8> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i8_1 = extractelement <32 x i8> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v64i8_1 = extractelement <64 x i8> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v128i8_1 = extractelement <128 x i8> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i8_1 = extractelement <vscale x 2 x i8> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i8_1 = extractelement <vscale x 4 x i8> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i8_1 = extractelement <vscale x 8 x i8> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i8_1 = extractelement <vscale x 16 x i8> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32i8_1 = extractelement <vscale x 32 x i8> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv64i8_1 = extractelement <vscale x 64 x i8> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv128i8_1 = extractelement <vscale x 128 x i8> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i16_1 = extractelement <2 x i16> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i16_1 = extractelement <4 x i16> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i16_1 = extractelement <8 x i16> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i16_1 = extractelement <16 x i16> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i16_1 = extractelement <32 x i16> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v64i16_1 = extractelement <64 x i16> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i16_1 = extractelement <vscale x 2 x i16> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i16_1 = extractelement <vscale x 4 x i16> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i16_1 = extractelement <vscale x 8 x i16> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i16_1 = extractelement <vscale x 16 x i16> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32i16_1 = extractelement <vscale x 32 x i16> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv64i16_1 = extractelement <vscale x 64 x i16> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i32_1 = extractelement <2 x i32> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i32_1 = extractelement <4 x i32> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i32_1 = extractelement <8 x i32> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i32_1 = extractelement <16 x i32> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_1 = extractelement <32 x i32> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i32_1 = extractelement <vscale x 2 x i32> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i32_1 = extractelement <vscale x 4 x i32> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i32_1 = extractelement <vscale x 8 x i32> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i32_1 = extractelement <vscale x 16 x i32> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32i32_1 = extractelement <vscale x 32 x i32> undef, i32 1
; RV32V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v2i64_1 = extractelement <2 x i64> undef, i32 1
; RV32V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v4i64_1 = extractelement <4 x i64> undef, i32 1
; RV32V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v8i64_1 = extractelement <8 x i64> undef, i32 1
@@ -116,51 +116,51 @@ define void @extractelement_int(i32 %x) {
; RV32V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv4i64_1 = extractelement <vscale x 4 x i64> undef, i32 1
; RV32V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv8i64_1 = extractelement <vscale x 8 x i64> undef, i32 1
; RV32V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv16i64_1 = extractelement <vscale x 16 x i64> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v2i1_x = extractelement <2 x i1> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v4i1_x = extractelement <4 x i1> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v8i1_x = extractelement <8 x i1> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v16i1_x = extractelement <16 x i1> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v32i1_x = extractelement <32 x i1> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv2i1_x = extractelement <vscale x 2 x i1> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv4i1_x = extractelement <vscale x 4 x i1> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv8i1_x = extractelement <vscale x 8 x i1> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %nxv16i1_x = extractelement <vscale x 16 x i1> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %nxv32i1_x = extractelement <vscale x 32 x i1> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i8_x = extractelement <2 x i8> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i8_x = extractelement <4 x i8> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i8_x = extractelement <8 x i8> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i8_x = extractelement <16 x i8> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i8_x = extractelement <32 x i8> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i8_x = extractelement <64 x i8> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v128i8_x = extractelement <128 x i8> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i8_x = extractelement <vscale x 2 x i8> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i8_x = extractelement <vscale x 4 x i8> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i8_x = extractelement <vscale x 8 x i8> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i8_x = extractelement <vscale x 16 x i8> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i8_x = extractelement <vscale x 32 x i8> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv64i8_x = extractelement <vscale x 64 x i8> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v2i1_x = extractelement <2 x i1> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v4i1_x = extractelement <4 x i1> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v8i1_x = extractelement <8 x i1> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v16i1_x = extractelement <16 x i1> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %v32i1_x = extractelement <32 x i1> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv2i1_x = extractelement <vscale x 2 x i1> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv4i1_x = extractelement <vscale x 4 x i1> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv8i1_x = extractelement <vscale x 8 x i1> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %nxv16i1_x = extractelement <vscale x 16 x i1> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %nxv32i1_x = extractelement <vscale x 32 x i1> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i8_x = extractelement <2 x i8> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i8_x = extractelement <4 x i8> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i8_x = extractelement <8 x i8> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i8_x = extractelement <16 x i8> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i8_x = extractelement <32 x i8> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v64i8_x = extractelement <64 x i8> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v128i8_x = extractelement <128 x i8> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i8_x = extractelement <vscale x 2 x i8> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i8_x = extractelement <vscale x 4 x i8> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i8_x = extractelement <vscale x 8 x i8> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i8_x = extractelement <vscale x 16 x i8> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32i8_x = extractelement <vscale x 32 x i8> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv64i8_x = extractelement <vscale x 64 x i8> undef, i32 %x
; RV32V-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %nxv128i8_x = extractelement <vscale x 128 x i8> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i16_x = extractelement <2 x i16> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i16_x = extractelement <4 x i16> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i16_x = extractelement <8 x i16> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i16_x = extractelement <16 x i16> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i16_x = extractelement <32 x i16> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i16_x = extractelement <64 x i16> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i16_x = extractelement <vscale x 2 x i16> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i16_x = extractelement <vscale x 4 x i16> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i16_x = extractelement <vscale x 8 x i16> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i16_x = extractelement <vscale x 16 x i16> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i16_x = extractelement <vscale x 32 x i16> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i16_x = extractelement <2 x i16> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i16_x = extractelement <4 x i16> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i16_x = extractelement <8 x i16> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i16_x = extractelement <16 x i16> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i16_x = extractelement <32 x i16> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v64i16_x = extractelement <64 x i16> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i16_x = extractelement <vscale x 2 x i16> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i16_x = extractelement <vscale x 4 x i16> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i16_x = extractelement <vscale x 8 x i16> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i16_x = extractelement <vscale x 16 x i16> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32i16_x = extractelement <vscale x 32 x i16> undef, i32 %x
; RV32V-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %nxv64i16_x = extractelement <vscale x 64 x i16> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i32_x = extractelement <2 x i32> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i32_x = extractelement <4 x i32> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i32_x = extractelement <8 x i32> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i32_x = extractelement <16 x i32> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_x = extractelement <32 x i32> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i32_x = extractelement <vscale x 2 x i32> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i32_x = extractelement <vscale x 4 x i32> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i32_x = extractelement <vscale x 8 x i32> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i32_x = extractelement <vscale x 16 x i32> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i32_x = extractelement <2 x i32> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i32_x = extractelement <4 x i32> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i32_x = extractelement <8 x i32> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i32_x = extractelement <16 x i32> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_x = extractelement <32 x i32> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i32_x = extractelement <vscale x 2 x i32> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i32_x = extractelement <vscale x 4 x i32> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i32_x = extractelement <vscale x 8 x i32> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i32_x = extractelement <vscale x 16 x i32> undef, i32 %x
; RV32V-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %nxv32i32_x = extractelement <vscale x 32 x i32> undef, i32 %x
; RV32V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v2i64_x = extractelement <2 x i64> undef, i32 %x
; RV32V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v4i64_x = extractelement <4 x i64> undef, i32 %x
@@ -173,218 +173,217 @@ define void @extractelement_int(i32 %x) {
; RV32V-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; RV64V-LABEL: 'extractelement_int'
-; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i1_0 = extractelement <2 x i1> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i1_0 = extractelement <4 x i1> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i1_0 = extractelement <8 x i1> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i1_0 = extractelement <16 x i1> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v32i1_0 = extractelement <32 x i1> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i1_0 = extractelement <vscale x 2 x i1> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i1_0 = extractelement <vscale x 4 x i1> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i1_0 = extractelement <vscale x 8 x i1> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv16i1_0 = extractelement <vscale x 16 x i1> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %nxv32i1_0 = extractelement <vscale x 32 x i1> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v2i8_0 = extractelement <2 x i8> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v4i8_0 = extractelement <4 x i8> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v8i8_0 = extractelement <8 x i8> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v16i8_0 = extractelement <16 x i8> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i8_0 = extractelement <32 x i8> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v64i8_0 = extractelement <64 x i8> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v128i8_0 = extractelement <128 x i8> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv2i8_0 = extractelement <vscale x 2 x i8> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv4i8_0 = extractelement <vscale x 4 x i8> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv8i8_0 = extractelement <vscale x 8 x i8> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv16i8_0 = extractelement <vscale x 16 x i8> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv32i8_0 = extractelement <vscale x 32 x i8> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv64i8_0 = extractelement <vscale x 64 x i8> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv128i8_0 = extractelement <vscale x 128 x i8> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v2i16_0 = extractelement <2 x i16> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v4i16_0 = extractelement <4 x i16> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v8i16_0 = extractelement <8 x i16> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v16i16_0 = extractelement <16 x i16> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i16_0 = extractelement <32 x i16> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v64i16_0 = extractelement <64 x i16> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv2i16_0 = extractelement <vscale x 2 x i16> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv4i16_0 = extractelement <vscale x 4 x i16> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv8i16_0 = extractelement <vscale x 8 x i16> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv16i16_0 = extractelement <vscale x 16 x i16> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv32i16_0 = extractelement <vscale x 32 x i16> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv64i16_0 = extractelement <vscale x 64 x i16> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v2i32_0 = extractelement <2 x i32> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v4i32_0 = extractelement <4 x i32> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v8i32_0 = extractelement <8 x i32> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v16i32_0 = extractelement <16 x i32> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i32_0 = extractelement <32 x i32> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv2i32_0 = extractelement <vscale x 2 x i32> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv4i32_0 = extractelement <vscale x 4 x i32> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv8i32_0 = extractelement <vscale x 8 x i32> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv16i32_0 = extractelement <vscale x 16 x i32> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv32i32_0 = extractelement <vscale x 32 x i32> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v2i64_0 = extractelement <2 x i64> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v4i64_0 = extractelement <4 x i64> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v8i64_0 = extractelement <8 x i64> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v16i64_0 = extractelement <16 x i64> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv2i64_0 = extractelement <vscale x 2 x i64> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv4i64_0 = extractelement <vscale x 4 x i64> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv8i64_0 = extractelement <vscale x 8 x i64> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv16i64_0 = extractelement <vscale x 16 x i64> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v2i1_1 = extractelement <2 x i1> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v4i1_1 = extractelement <4 x i1> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v8i1_1 = extractelement <8 x i1> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v16i1_1 = extractelement <16 x i1> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v32i1_1 = extractelement <32 x i1> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv2i1_1 = extractelement <vscale x 2 x i1> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv4i1_1 = extractelement <vscale x 4 x i1> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv8i1_1 = extractelement <vscale x 8 x i1> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %nxv16i1_1 = extractelement <vscale x 16 x i1> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %nxv32i1_1 = extractelement <vscale x 32 x i1> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i8_1 = extractelement <2 x i8> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i8_1 = extractelement <4 x i8> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i8_1 = extractelement <8 x i8> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i8_1 = extractelement <16 x i8> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i8_1 = extractelement <32 x i8> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i8_1 = extractelement <64 x i8> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v128i8_1 = extractelement <128 x i8> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i8_1 = extractelement <vscale x 2 x i8> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i8_1 = extractelement <vscale x 4 x i8> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i8_1 = extractelement <vscale x 8 x i8> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i8_1 = extractelement <vscale x 16 x i8> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i8_1 = extractelement <vscale x 32 x i8> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv64i8_1 = extractelement <vscale x 64 x i8> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv128i8_1 = extractelement <vscale x 128 x i8> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i16_1 = extractelement <2 x i16> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i16_1 = extractelement <4 x i16> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i16_1 = extractelement <8 x i16> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i16_1 = extractelement <16 x i16> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i16_1 = extractelement <32 x i16> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i16_1 = extractelement <64 x i16> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i16_1 = extractelement <vscale x 2 x i16> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i16_1 = extractelement <vscale x 4 x i16> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i16_1 = extractelement <vscale x 8 x i16> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i16_1 = extractelement <vscale x 16 x i16> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i16_1 = extractelement <vscale x 32 x i16> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv64i16_1 = extractelement <vscale x 64 x i16> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i32_1 = extractelement <2 x i32> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i32_1 = extractelement <4 x i32> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i32_1 = extractelement <8 x i32> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i32_1 = extractelement <16 x i32> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_1 = extractelement <32 x i32> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i32_1 = extractelement <vscale x 2 x i32> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i32_1 = extractelement <vscale x 4 x i32> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i32_1 = extractelement <vscale x 8 x i32> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i32_1 = extractelement <vscale x 16 x i32> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i32_1 = extractelement <vscale x 32 x i32> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i64_1 = extractelement <2 x i64> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i64_1 = extractelement <4 x i64> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i64_1 = extractelement <8 x i64> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i64_1 = extractelement <16 x i64> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i64_1 = extractelement <vscale x 2 x i64> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i64_1 = extractelement <vscale x 4 x i64> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i64_1 = extractelement <vscale x 8 x i64> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i64_1 = extractelement <vscale x 16 x i64> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v2i1_x = extractelement <2 x i1> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v4i1_x = extractelement <4 x i1> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v8i1_x = extractelement <8 x i1> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v16i1_x = extractelement <16 x i1> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v32i1_x = extractelement <32 x i1> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv2i1_x = extractelement <vscale x 2 x i1> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv4i1_x = extractelement <vscale x 4 x i1> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv8i1_x = extractelement <vscale x 8 x i1> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %nxv16i1_x = extractelement <vscale x 16 x i1> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %nxv32i1_x = extractelement <vscale x 32 x i1> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i8_x = extractelement <2 x i8> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i8_x = extractelement <4 x i8> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i8_x = extractelement <8 x i8> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i8_x = extractelement <16 x i8> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i8_x = extractelement <32 x i8> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i8_x = extractelement <64 x i8> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v128i8_x = extractelement <128 x i8> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i8_x = extractelement <vscale x 2 x i8> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i8_x = extractelement <vscale x 4 x i8> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i8_x = extractelement <vscale x 8 x i8> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i8_x = extractelement <vscale x 16 x i8> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i8_x = extractelement <vscale x 32 x i8> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv64i8_x = extractelement <vscale x 64 x i8> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v2i1_0 = extractelement <2 x i1> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v4i1_0 = extractelement <4 x i1> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v8i1_0 = extractelement <8 x i1> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v16i1_0 = extractelement <16 x i1> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v32i1_0 = extractelement <32 x i1> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv2i1_0 = extractelement <vscale x 2 x i1> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv4i1_0 = extractelement <vscale x 4 x i1> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv8i1_0 = extractelement <vscale x 8 x i1> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %nxv16i1_0 = extractelement <vscale x 16 x i1> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %nxv32i1_0 = extractelement <vscale x 32 x i1> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i8_0 = extractelement <2 x i8> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i8_0 = extractelement <4 x i8> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i8_0 = extractelement <8 x i8> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i8_0 = extractelement <16 x i8> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i8_0 = extractelement <32 x i8> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i8_0 = extractelement <64 x i8> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v128i8_0 = extractelement <128 x i8> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i8_0 = extractelement <vscale x 2 x i8> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i8_0 = extractelement <vscale x 4 x i8> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i8_0 = extractelement <vscale x 8 x i8> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i8_0 = extractelement <vscale x 16 x i8> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i8_0 = extractelement <vscale x 32 x i8> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv64i8_0 = extractelement <vscale x 64 x i8> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv128i8_0 = extractelement <vscale x 128 x i8> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i16_0 = extractelement <2 x i16> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i16_0 = extractelement <4 x i16> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i16_0 = extractelement <8 x i16> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i16_0 = extractelement <16 x i16> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i16_0 = extractelement <32 x i16> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i16_0 = extractelement <64 x i16> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i16_0 = extractelement <vscale x 2 x i16> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i16_0 = extractelement <vscale x 4 x i16> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i16_0 = extractelement <vscale x 8 x i16> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i16_0 = extractelement <vscale x 16 x i16> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i16_0 = extractelement <vscale x 32 x i16> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv64i16_0 = extractelement <vscale x 64 x i16> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i32_0 = extractelement <2 x i32> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i32_0 = extractelement <4 x i32> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i32_0 = extractelement <8 x i32> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i32_0 = extractelement <16 x i32> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_0 = extractelement <32 x i32> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i32_0 = extractelement <vscale x 2 x i32> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i32_0 = extractelement <vscale x 4 x i32> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i32_0 = extractelement <vscale x 8 x i32> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i32_0 = extractelement <vscale x 16 x i32> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i32_0 = extractelement <vscale x 32 x i32> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i64_0 = extractelement <2 x i64> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i64_0 = extractelement <4 x i64> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i64_0 = extractelement <8 x i64> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i64_0 = extractelement <16 x i64> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i64_0 = extractelement <vscale x 2 x i64> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i64_0 = extractelement <vscale x 4 x i64> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i64_0 = extractelement <vscale x 8 x i64> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i64_0 = extractelement <vscale x 16 x i64> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v2i1_1 = extractelement <2 x i1> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v4i1_1 = extractelement <4 x i1> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v8i1_1 = extractelement <8 x i1> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v16i1_1 = extractelement <16 x i1> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %v32i1_1 = extractelement <32 x i1> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv2i1_1 = extractelement <vscale x 2 x i1> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv4i1_1 = extractelement <vscale x 4 x i1> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv8i1_1 = extractelement <vscale x 8 x i1> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %nxv16i1_1 = extractelement <vscale x 16 x i1> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %nxv32i1_1 = extractelement <vscale x 32 x i1> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i8_1 = extractelement <2 x i8> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i8_1 = extractelement <4 x i8> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i8_1 = extractelement <8 x i8> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i8_1 = extractelement <16 x i8> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i8_1 = extractelement <32 x i8> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v64i8_1 = extractelement <64 x i8> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v128i8_1 = extractelement <128 x i8> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i8_1 = extractelement <vscale x 2 x i8> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i8_1 = extractelement <vscale x 4 x i8> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i8_1 = extractelement <vscale x 8 x i8> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i8_1 = extractelement <vscale x 16 x i8> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32i8_1 = extractelement <vscale x 32 x i8> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv64i8_1 = extractelement <vscale x 64 x i8> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv128i8_1 = extractelement <vscale x 128 x i8> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i16_1 = extractelement <2 x i16> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i16_1 = extractelement <4 x i16> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i16_1 = extractelement <8 x i16> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i16_1 = extractelement <16 x i16> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i16_1 = extractelement <32 x i16> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v64i16_1 = extractelement <64 x i16> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i16_1 = extractelement <vscale x 2 x i16> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i16_1 = extractelement <vscale x 4 x i16> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i16_1 = extractelement <vscale x 8 x i16> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i16_1 = extractelement <vscale x 16 x i16> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32i16_1 = extractelement <vscale x 32 x i16> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv64i16_1 = extractelement <vscale x 64 x i16> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i32_1 = extractelement <2 x i32> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i32_1 = extractelement <4 x i32> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i32_1 = extractelement <8 x i32> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i32_1 = extractelement <16 x i32> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_1 = extractelement <32 x i32> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i32_1 = extractelement <vscale x 2 x i32> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i32_1 = extractelement <vscale x 4 x i32> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i32_1 = extractelement <vscale x 8 x i32> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i32_1 = extractelement <vscale x 16 x i32> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32i32_1 = extractelement <vscale x 32 x i32> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i64_1 = extractelement <2 x i64> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i64_1 = extractelement <4 x i64> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i64_1 = extractelement <8 x i64> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i64_1 = extractelement <16 x i64> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i64_1 = extractelement <vscale x 2 x i64> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i64_1 = extractelement <vscale x 4 x i64> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i64_1 = extractelement <vscale x 8 x i64> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i64_1 = extractelement <vscale x 16 x i64> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v2i1_x = extractelement <2 x i1> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v4i1_x = extractelement <4 x i1> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v8i1_x = extractelement <8 x i1> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v16i1_x = extractelement <16 x i1> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %v32i1_x = extractelement <32 x i1> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv2i1_x = extractelement <vscale x 2 x i1> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv4i1_x = extractelement <vscale x 4 x i1> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv8i1_x = extractelement <vscale x 8 x i1> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %nxv16i1_x = extractelement <vscale x 16 x i1> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %nxv32i1_x = extractelement <vscale x 32 x i1> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i8_x = extractelement <2 x i8> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i8_x = extractelement <4 x i8> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i8_x = extractelement <8 x i8> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i8_x = extractelement <16 x i8> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i8_x = extractelement <32 x i8> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v64i8_x = extractelement <64 x i8> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v128i8_x = extractelement <128 x i8> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i8_x = extractelement <vscale x 2 x i8> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i8_x = extractelement <vscale x 4 x i8> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i8_x = extractelement <vscale x 8 x i8> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i8_x = extractelement <vscale x 16 x i8> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32i8_x = extractelement <vscale x 32 x i8> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv64i8_x = extractelement <vscale x 64 x i8> undef, i32 %x
; RV64V-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %nxv128i8_x = extractelement <vscale x 128 x i8> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i16_x = extractelement <2 x i16> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i16_x = extractelement <4 x i16> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i16_x = extractelement <8 x i16> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i16_x = extractelement <16 x i16> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i16_x = extractelement <32 x i16> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i16_x = extractelement <64 x i16> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i16_x = extractelement <vscale x 2 x i16> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i16_x = extractelement <vscale x 4 x i16> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i16_x = extractelement <vscale x 8 x i16> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i16_x = extractelement <vscale x 16 x i16> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i16_x = extractelement <vscale x 32 x i16> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i16_x = extractelement <2 x i16> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i16_x = extractelement <4 x i16> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i16_x = extractelement <8 x i16> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i16_x = extractelement <16 x i16> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i16_x = extractelement <32 x i16> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v64i16_x = extractelement <64 x i16> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i16_x = extractelement <vscale x 2 x i16> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i16_x = extractelement <vscale x 4 x i16> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i16_x = extractelement <vscale x 8 x i16> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i16_x = extractelement <vscale x 16 x i16> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32i16_x = extractelement <vscale x 32 x i16> undef, i32 %x
; RV64V-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %nxv64i16_x = extractelement <vscale x 64 x i16> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i32_x = extractelement <2 x i32> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i32_x = extractelement <4 x i32> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i32_x = extractelement <8 x i32> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i32_x = extractelement <16 x i32> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_x = extractelement <32 x i32> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i32_x = extractelement <vscale x 2 x i32> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i32_x = extractelement <vscale x 4 x i32> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i32_x = extractelement <vscale x 8 x i32> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i32_x = extractelement <vscale x 16 x i32> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i32_x = extractelement <2 x i32> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i32_x = extractelement <4 x i32> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i32_x = extractelement <8 x i32> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i32_x = extractelement <16 x i32> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_x = extractelement <32 x i32> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i32_x = extractelement <vscale x 2 x i32> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i32_x = extractelement <vscale x 4 x i32> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i32_x = extractelement <vscale x 8 x i32> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i32_x = extractelement <vscale x 16 x i32> undef, i32 %x
; RV64V-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %nxv32i32_x = extractelement <vscale x 32 x i32> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i64_x = extractelement <2 x i64> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i64_x = extractelement <4 x i64> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i64_x = extractelement <8 x i64> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i64_x = extractelement <16 x i64> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i64_x = extractelement <vscale x 2 x i64> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i64_x = extractelement <vscale x 4 x i64> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i64_x = extractelement <vscale x 8 x i64> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i64_x = extractelement <2 x i64> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i64_x = extractelement <4 x i64> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i64_x = extractelement <8 x i64> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i64_x = extractelement <16 x i64> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i64_x = extractelement <vscale x 2 x i64> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i64_x = extractelement <vscale x 4 x i64> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i64_x = extractelement <vscale x 8 x i64> undef, i32 %x
; RV64V-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %nxv16i64_x = extractelement <vscale x 16 x i64> undef, i32 %x
; RV64V-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
-;
; RV32ZVE64X-LABEL: 'extractelement_int'
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i1_0 = extractelement <2 x i1> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i1_0 = extractelement <4 x i1> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i1_0 = extractelement <8 x i1> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v16i1_0 = extractelement <16 x i1> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v32i1_0 = extractelement <32 x i1> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i1_0 = extractelement <vscale x 2 x i1> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i1_0 = extractelement <vscale x 4 x i1> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i1_0 = extractelement <vscale x 8 x i1> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv16i1_0 = extractelement <vscale x 16 x i1> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %nxv32i1_0 = extractelement <vscale x 32 x i1> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v2i8_0 = extractelement <2 x i8> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v4i8_0 = extractelement <4 x i8> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v8i8_0 = extractelement <8 x i8> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v16i8_0 = extractelement <16 x i8> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i8_0 = extractelement <32 x i8> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v64i8_0 = extractelement <64 x i8> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v128i8_0 = extractelement <128 x i8> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv2i8_0 = extractelement <vscale x 2 x i8> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv4i8_0 = extractelement <vscale x 4 x i8> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv8i8_0 = extractelement <vscale x 8 x i8> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv16i8_0 = extractelement <vscale x 16 x i8> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv32i8_0 = extractelement <vscale x 32 x i8> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv64i8_0 = extractelement <vscale x 64 x i8> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv128i8_0 = extractelement <vscale x 128 x i8> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v2i16_0 = extractelement <2 x i16> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v4i16_0 = extractelement <4 x i16> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v8i16_0 = extractelement <8 x i16> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v16i16_0 = extractelement <16 x i16> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i16_0 = extractelement <32 x i16> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v64i16_0 = extractelement <64 x i16> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv2i16_0 = extractelement <vscale x 2 x i16> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv4i16_0 = extractelement <vscale x 4 x i16> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv8i16_0 = extractelement <vscale x 8 x i16> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv16i16_0 = extractelement <vscale x 16 x i16> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv32i16_0 = extractelement <vscale x 32 x i16> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv64i16_0 = extractelement <vscale x 64 x i16> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v2i32_0 = extractelement <2 x i32> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v4i32_0 = extractelement <4 x i32> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v8i32_0 = extractelement <8 x i32> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v16i32_0 = extractelement <16 x i32> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i32_0 = extractelement <32 x i32> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv2i32_0 = extractelement <vscale x 2 x i32> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv4i32_0 = extractelement <vscale x 4 x i32> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv8i32_0 = extractelement <vscale x 8 x i32> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv16i32_0 = extractelement <vscale x 16 x i32> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv32i32_0 = extractelement <vscale x 32 x i32> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v2i1_0 = extractelement <2 x i1> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v4i1_0 = extractelement <4 x i1> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v8i1_0 = extractelement <8 x i1> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v16i1_0 = extractelement <16 x i1> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %v32i1_0 = extractelement <32 x i1> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv2i1_0 = extractelement <vscale x 2 x i1> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv4i1_0 = extractelement <vscale x 4 x i1> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv8i1_0 = extractelement <vscale x 8 x i1> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %nxv16i1_0 = extractelement <vscale x 16 x i1> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %nxv32i1_0 = extractelement <vscale x 32 x i1> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i8_0 = extractelement <2 x i8> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i8_0 = extractelement <4 x i8> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i8_0 = extractelement <8 x i8> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i8_0 = extractelement <16 x i8> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i8_0 = extractelement <32 x i8> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i8_0 = extractelement <64 x i8> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v128i8_0 = extractelement <128 x i8> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i8_0 = extractelement <vscale x 2 x i8> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i8_0 = extractelement <vscale x 4 x i8> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i8_0 = extractelement <vscale x 8 x i8> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i8_0 = extractelement <vscale x 16 x i8> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i8_0 = extractelement <vscale x 32 x i8> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv64i8_0 = extractelement <vscale x 64 x i8> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv128i8_0 = extractelement <vscale x 128 x i8> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i16_0 = extractelement <2 x i16> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i16_0 = extractelement <4 x i16> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i16_0 = extractelement <8 x i16> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i16_0 = extractelement <16 x i16> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i16_0 = extractelement <32 x i16> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i16_0 = extractelement <64 x i16> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i16_0 = extractelement <vscale x 2 x i16> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i16_0 = extractelement <vscale x 4 x i16> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i16_0 = extractelement <vscale x 8 x i16> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i16_0 = extractelement <vscale x 16 x i16> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i16_0 = extractelement <vscale x 32 x i16> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv64i16_0 = extractelement <vscale x 64 x i16> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i32_0 = extractelement <2 x i32> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i32_0 = extractelement <4 x i32> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i32_0 = extractelement <8 x i32> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i32_0 = extractelement <16 x i32> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_0 = extractelement <32 x i32> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i32_0 = extractelement <vscale x 2 x i32> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i32_0 = extractelement <vscale x 4 x i32> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i32_0 = extractelement <vscale x 8 x i32> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i32_0 = extractelement <vscale x 16 x i32> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i32_0 = extractelement <vscale x 32 x i32> undef, i32 0
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v2i64_0 = extractelement <2 x i64> undef, i32 0
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v4i64_0 = extractelement <4 x i64> undef, i32 0
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v8i64_0 = extractelement <8 x i64> undef, i32 0
@@ -393,52 +392,52 @@ define void @extractelement_int(i32 %x) {
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv4i64_0 = extractelement <vscale x 4 x i64> undef, i32 0
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv8i64_0 = extractelement <vscale x 8 x i64> undef, i32 0
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv16i64_0 = extractelement <vscale x 16 x i64> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v2i1_1 = extractelement <2 x i1> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v4i1_1 = extractelement <4 x i1> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v8i1_1 = extractelement <8 x i1> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v16i1_1 = extractelement <16 x i1> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %v32i1_1 = extractelement <32 x i1> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv2i1_1 = extractelement <vscale x 2 x i1> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv4i1_1 = extractelement <vscale x 4 x i1> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv8i1_1 = extractelement <vscale x 8 x i1> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %nxv16i1_1 = extractelement <vscale x 16 x i1> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %nxv32i1_1 = extractelement <vscale x 32 x i1> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i8_1 = extractelement <2 x i8> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i8_1 = extractelement <4 x i8> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i8_1 = extractelement <8 x i8> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i8_1 = extractelement <16 x i8> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i8_1 = extractelement <32 x i8> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i8_1 = extractelement <64 x i8> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v128i8_1 = extractelement <128 x i8> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i8_1 = extractelement <vscale x 2 x i8> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i8_1 = extractelement <vscale x 4 x i8> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i8_1 = extractelement <vscale x 8 x i8> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i8_1 = extractelement <vscale x 16 x i8> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i8_1 = extractelement <vscale x 32 x i8> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv64i8_1 = extractelement <vscale x 64 x i8> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv128i8_1 = extractelement <vscale x 128 x i8> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i16_1 = extractelement <2 x i16> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i16_1 = extractelement <4 x i16> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i16_1 = extractelement <8 x i16> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i16_1 = extractelement <16 x i16> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i16_1 = extractelement <32 x i16> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i16_1 = extractelement <64 x i16> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i16_1 = extractelement <vscale x 2 x i16> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i16_1 = extractelement <vscale x 4 x i16> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i16_1 = extractelement <vscale x 8 x i16> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i16_1 = extractelement <vscale x 16 x i16> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i16_1 = extractelement <vscale x 32 x i16> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv64i16_1 = extractelement <vscale x 64 x i16> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i32_1 = extractelement <2 x i32> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i32_1 = extractelement <4 x i32> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i32_1 = extractelement <8 x i32> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i32_1 = extractelement <16 x i32> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_1 = extractelement <32 x i32> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i32_1 = extractelement <vscale x 2 x i32> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i32_1 = extractelement <vscale x 4 x i32> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i32_1 = extractelement <vscale x 8 x i32> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i32_1 = extractelement <vscale x 16 x i32> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i32_1 = extractelement <vscale x 32 x i32> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v2i1_1 = extractelement <2 x i1> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v4i1_1 = extractelement <4 x i1> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v8i1_1 = extractelement <8 x i1> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %v16i1_1 = extractelement <16 x i1> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %v32i1_1 = extractelement <32 x i1> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv2i1_1 = extractelement <vscale x 2 x i1> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv4i1_1 = extractelement <vscale x 4 x i1> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv8i1_1 = extractelement <vscale x 8 x i1> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %nxv16i1_1 = extractelement <vscale x 16 x i1> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %nxv32i1_1 = extractelement <vscale x 32 x i1> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i8_1 = extractelement <2 x i8> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i8_1 = extractelement <4 x i8> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i8_1 = extractelement <8 x i8> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i8_1 = extractelement <16 x i8> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i8_1 = extractelement <32 x i8> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v64i8_1 = extractelement <64 x i8> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v128i8_1 = extractelement <128 x i8> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i8_1 = extractelement <vscale x 2 x i8> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i8_1 = extractelement <vscale x 4 x i8> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i8_1 = extractelement <vscale x 8 x i8> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i8_1 = extractelement <vscale x 16 x i8> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32i8_1 = extractelement <vscale x 32 x i8> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv64i8_1 = extractelement <vscale x 64 x i8> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv128i8_1 = extractelement <vscale x 128 x i8> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i16_1 = extractelement <2 x i16> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i16_1 = extractelement <4 x i16> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i16_1 = extractelement <8 x i16> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i16_1 = extractelement <16 x i16> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i16_1 = extractelement <32 x i16> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v64i16_1 = extractelement <64 x i16> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i16_1 = extractelement <vscale x 2 x i16> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i16_1 = extractelement <vscale x 4 x i16> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i16_1 = extractelement <vscale x 8 x i16> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i16_1 = extractelement <vscale x 16 x i16> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32i16_1 = extractelement <vscale x 32 x i16> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv64i16_1 = extractelement <vscale x 64 x i16> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i32_1 = extractelement <2 x i32> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i32_1 = extractelement <4 x i32> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i32_1 = extractelement <8 x i32> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i32_1 = extractelement <16 x i32> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_1 = extractelement <32 x i32> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i32_1 = extractelement <vscale x 2 x i32> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i32_1 = extractelement <vscale x 4 x i32> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i32_1 = extractelement <vscale x 8 x i32> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i32_1 = extractelement <vscale x 16 x i32> undef, i32 1
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32i32_1 = extractelement <vscale x 32 x i32> undef, i32 1
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v2i64_1 = extractelement <2 x i64> undef, i32 1
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v4i64_1 = extractelement <4 x i64> undef, i32 1
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v8i64_1 = extractelement <8 x i64> undef, i32 1
@@ -447,51 +446,51 @@ define void @extractelement_int(i32 %x) {
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv4i64_1 = extractelement <vscale x 4 x i64> undef, i32 1
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv8i64_1 = extractelement <vscale x 8 x i64> undef, i32 1
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv16i64_1 = extractelement <vscale x 16 x i64> undef, i32 1
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v2i1_x = extractelement <2 x i1> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v4i1_x = extractelement <4 x i1> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v8i1_x = extractelement <8 x i1> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v16i1_x = extractelement <16 x i1> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %v32i1_x = extractelement <32 x i1> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv2i1_x = extractelement <vscale x 2 x i1> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv4i1_x = extractelement <vscale x 4 x i1> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv8i1_x = extractelement <vscale x 8 x i1> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %nxv16i1_x = extractelement <vscale x 16 x i1> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %nxv32i1_x = extractelement <vscale x 32 x i1> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i8_x = extractelement <2 x i8> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i8_x = extractelement <4 x i8> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i8_x = extractelement <8 x i8> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i8_x = extractelement <16 x i8> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i8_x = extractelement <32 x i8> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i8_x = extractelement <64 x i8> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v2i1_x = extractelement <2 x i1> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v4i1_x = extractelement <4 x i1> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v8i1_x = extractelement <8 x i1> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %v16i1_x = extractelement <16 x i1> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %v32i1_x = extractelement <32 x i1> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv2i1_x = extractelement <vscale x 2 x i1> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv4i1_x = extractelement <vscale x 4 x i1> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv8i1_x = extractelement <vscale x 8 x i1> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %nxv16i1_x = extractelement <vscale x 16 x i1> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %nxv32i1_x = extractelement <vscale x 32 x i1> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i8_x = extractelement <2 x i8> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i8_x = extractelement <4 x i8> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i8_x = extractelement <8 x i8> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i8_x = extractelement <16 x i8> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i8_x = extractelement <32 x i8> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v64i8_x = extractelement <64 x i8> undef, i32 %x
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %v128i8_x = extractelement <128 x i8> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i8_x = extractelement <vscale x 2 x i8> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i8_x = extractelement <vscale x 4 x i8> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i8_x = extractelement <vscale x 8 x i8> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i8_x = extractelement <vscale x 16 x i8> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i8_x = extractelement <vscale x 32 x i8> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv64i8_x = extractelement <vscale x 64 x i8> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i8_x = extractelement <vscale x 2 x i8> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i8_x = extractelement <vscale x 4 x i8> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i8_x = extractelement <vscale x 8 x i8> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i8_x = extractelement <vscale x 16 x i8> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32i8_x = extractelement <vscale x 32 x i8> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv64i8_x = extractelement <vscale x 64 x i8> undef, i32 %x
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %nxv128i8_x = extractelement <vscale x 128 x i8> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i16_x = extractelement <2 x i16> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i16_x = extractelement <4 x i16> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i16_x = extractelement <8 x i16> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i16_x = extractelement <16 x i16> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i16_x = extractelement <32 x i16> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i16_x = extractelement <2 x i16> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i16_x = extractelement <4 x i16> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i16_x = extractelement <8 x i16> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i16_x = extractelement <16 x i16> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i16_x = extractelement <32 x i16> undef, i32 %x
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %v64i16_x = extractelement <64 x i16> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i16_x = extractelement <vscale x 2 x i16> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i16_x = extractelement <vscale x 4 x i16> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i16_x = extractelement <vscale x 8 x i16> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i16_x = extractelement <vscale x 16 x i16> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i16_x = extractelement <vscale x 32 x i16> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i16_x = extractelement <vscale x 2 x i16> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i16_x = extractelement <vscale x 4 x i16> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i16_x = extractelement <vscale x 8 x i16> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i16_x = extractelement <vscale x 16 x i16> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32i16_x = extractelement <vscale x 32 x i16> undef, i32 %x
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %nxv64i16_x = extractelement <vscale x 64 x i16> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i32_x = extractelement <2 x i32> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i32_x = extractelement <4 x i32> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i32_x = extractelement <8 x i32> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i32_x = extractelement <16 x i32> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i32_x = extractelement <2 x i32> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i32_x = extractelement <4 x i32> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i32_x = extractelement <8 x i32> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i32_x = extractelement <16 x i32> undef, i32 %x
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %v32i32_x = extractelement <32 x i32> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i32_x = extractelement <vscale x 2 x i32> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i32_x = extractelement <vscale x 4 x i32> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i32_x = extractelement <vscale x 8 x i32> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i32_x = extractelement <vscale x 16 x i32> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i32_x = extractelement <vscale x 2 x i32> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i32_x = extractelement <vscale x 4 x i32> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i32_x = extractelement <vscale x 8 x i32> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i32_x = extractelement <vscale x 16 x i32> undef, i32 %x
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %nxv32i32_x = extractelement <vscale x 32 x i32> undef, i32 %x
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v2i64_x = extractelement <2 x i64> undef, i32 %x
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v4i64_x = extractelement <4 x i64> undef, i32 %x
@@ -504,167 +503,167 @@ define void @extractelement_int(i32 %x) {
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; RV64ZVE64X-LABEL: 'extractelement_int'
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i1_0 = extractelement <2 x i1> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i1_0 = extractelement <4 x i1> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i1_0 = extractelement <8 x i1> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v16i1_0 = extractelement <16 x i1> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %v32i1_0 = extractelement <32 x i1> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i1_0 = extractelement <vscale x 2 x i1> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i1_0 = extractelement <vscale x 4 x i1> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i1_0 = extractelement <vscale x 8 x i1> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv16i1_0 = extractelement <vscale x 16 x i1> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %nxv32i1_0 = extractelement <vscale x 32 x i1> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v2i8_0 = extractelement <2 x i8> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v4i8_0 = extractelement <4 x i8> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v8i8_0 = extractelement <8 x i8> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v16i8_0 = extractelement <16 x i8> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i8_0 = extractelement <32 x i8> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v64i8_0 = extractelement <64 x i8> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v128i8_0 = extractelement <128 x i8> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv2i8_0 = extractelement <vscale x 2 x i8> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv4i8_0 = extractelement <vscale x 4 x i8> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv8i8_0 = extractelement <vscale x 8 x i8> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv16i8_0 = extractelement <vscale x 16 x i8> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv32i8_0 = extractelement <vscale x 32 x i8> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv64i8_0 = extractelement <vscale x 64 x i8> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv128i8_0 = extractelement <vscale x 128 x i8> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v2i16_0 = extractelement <2 x i16> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v4i16_0 = extractelement <4 x i16> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v8i16_0 = extractelement <8 x i16> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v16i16_0 = extractelement <16 x i16> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i16_0 = extractelement <32 x i16> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v64i16_0 = extractelement <64 x i16> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv2i16_0 = extractelement <vscale x 2 x i16> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv4i16_0 = extractelement <vscale x 4 x i16> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv8i16_0 = extractelement <vscale x 8 x i16> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv16i16_0 = extractelement <vscale x 16 x i16> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv32i16_0 = extractelement <vscale x 32 x i16> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv64i16_0 = extractelement <vscale x 64 x i16> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v2i32_0 = extractelement <2 x i32> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v4i32_0 = extractelement <4 x i32> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v8i32_0 = extractelement <8 x i32> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v16i32_0 = extractelement <16 x i32> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i32_0 = extractelement <32 x i32> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv2i32_0 = extractelement <vscale x 2 x i32> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv4i32_0 = extractelement <vscale x 4 x i32> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv8i32_0 = extractelement <vscale x 8 x i32> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv16i32_0 = extractelement <vscale x 16 x i32> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv32i32_0 = extractelement <vscale x 32 x i32> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v2i64_0 = extractelement <2 x i64> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v4i64_0 = extractelement <4 x i64> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v8i64_0 = extractelement <8 x i64> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v16i64_0 = extractelement <16 x i64> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv2i64_0 = extractelement <vscale x 2 x i64> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv4i64_0 = extractelement <vscale x 4 x i64> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv8i64_0 = extractelement <vscale x 8 x i64> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv16i64_0 = extractelement <vscale x 16 x i64> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v2i1_1 = extractelement <2 x i1> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v4i1_1 = extractelement <4 x i1> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v8i1_1 = extractelement <8 x i1> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v16i1_1 = extractelement <16 x i1> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %v32i1_1 = extractelement <32 x i1> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv2i1_1 = extractelement <vscale x 2 x i1> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv4i1_1 = extractelement <vscale x 4 x i1> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv8i1_1 = extractelement <vscale x 8 x i1> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %nxv16i1_1 = extractelement <vscale x 16 x i1> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %nxv32i1_1 = extractelement <vscale x 32 x i1> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i8_1 = extractelement <2 x i8> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i8_1 = extractelement <4 x i8> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i8_1 = extractelement <8 x i8> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i8_1 = extractelement <16 x i8> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i8_1 = extractelement <32 x i8> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i8_1 = extractelement <64 x i8> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v128i8_1 = extractelement <128 x i8> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i8_1 = extractelement <vscale x 2 x i8> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i8_1 = extractelement <vscale x 4 x i8> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i8_1 = extractelement <vscale x 8 x i8> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i8_1 = extractelement <vscale x 16 x i8> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i8_1 = extractelement <vscale x 32 x i8> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv64i8_1 = extractelement <vscale x 64 x i8> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv128i8_1 = extractelement <vscale x 128 x i8> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i16_1 = extractelement <2 x i16> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i16_1 = extractelement <4 x i16> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i16_1 = extractelement <8 x i16> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i16_1 = extractelement <16 x i16> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i16_1 = extractelement <32 x i16> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i16_1 = extractelement <64 x i16> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i16_1 = extractelement <vscale x 2 x i16> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i16_1 = extractelement <vscale x 4 x i16> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i16_1 = extractelement <vscale x 8 x i16> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i16_1 = extractelement <vscale x 16 x i16> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i16_1 = extractelement <vscale x 32 x i16> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv64i16_1 = extractelement <vscale x 64 x i16> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i32_1 = extractelement <2 x i32> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i32_1 = extractelement <4 x i32> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i32_1 = extractelement <8 x i32> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i32_1 = extractelement <16 x i32> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_1 = extractelement <32 x i32> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i32_1 = extractelement <vscale x 2 x i32> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i32_1 = extractelement <vscale x 4 x i32> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i32_1 = extractelement <vscale x 8 x i32> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i32_1 = extractelement <vscale x 16 x i32> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i32_1 = extractelement <vscale x 32 x i32> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i64_1 = extractelement <2 x i64> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i64_1 = extractelement <4 x i64> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i64_1 = extractelement <8 x i64> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i64_1 = extractelement <16 x i64> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i64_1 = extractelement <vscale x 2 x i64> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i64_1 = extractelement <vscale x 4 x i64> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i64_1 = extractelement <vscale x 8 x i64> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i64_1 = extractelement <vscale x 16 x i64> undef, i32 1
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v2i1_x = extractelement <2 x i1> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v4i1_x = extractelement <4 x i1> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v8i1_x = extractelement <8 x i1> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v16i1_x = extractelement <16 x i1> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %v32i1_x = extractelement <32 x i1> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv2i1_x = extractelement <vscale x 2 x i1> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv4i1_x = extractelement <vscale x 4 x i1> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv8i1_x = extractelement <vscale x 8 x i1> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %nxv16i1_x = extractelement <vscale x 16 x i1> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %nxv32i1_x = extractelement <vscale x 32 x i1> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i8_x = extractelement <2 x i8> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i8_x = extractelement <4 x i8> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i8_x = extractelement <8 x i8> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i8_x = extractelement <16 x i8> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i8_x = extractelement <32 x i8> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i8_x = extractelement <64 x i8> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v2i1_0 = extractelement <2 x i1> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v4i1_0 = extractelement <4 x i1> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %v8i1_0 = extractelement <8 x i1> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %v16i1_0 = extractelement <16 x i1> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %v32i1_0 = extractelement <32 x i1> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv2i1_0 = extractelement <vscale x 2 x i1> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv4i1_0 = extractelement <vscale x 4 x i1> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %nxv8i1_0 = extractelement <vscale x 8 x i1> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %nxv16i1_0 = extractelement <vscale x 16 x i1> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %nxv32i1_0 = extractelement <vscale x 32 x i1> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i8_0 = extractelement <2 x i8> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i8_0 = extractelement <4 x i8> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i8_0 = extractelement <8 x i8> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i8_0 = extractelement <16 x i8> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i8_0 = extractelement <32 x i8> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i8_0 = extractelement <64 x i8> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v128i8_0 = extractelement <128 x i8> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i8_0 = extractelement <vscale x 2 x i8> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i8_0 = extractelement <vscale x 4 x i8> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i8_0 = extractelement <vscale x 8 x i8> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i8_0 = extractelement <vscale x 16 x i8> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i8_0 = extractelement <vscale x 32 x i8> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv64i8_0 = extractelement <vscale x 64 x i8> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv128i8_0 = extractelement <vscale x 128 x i8> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i16_0 = extractelement <2 x i16> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i16_0 = extractelement <4 x i16> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i16_0 = extractelement <8 x i16> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i16_0 = extractelement <16 x i16> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i16_0 = extractelement <32 x i16> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i16_0 = extractelement <64 x i16> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i16_0 = extractelement <vscale x 2 x i16> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i16_0 = extractelement <vscale x 4 x i16> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i16_0 = extractelement <vscale x 8 x i16> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i16_0 = extractelement <vscale x 16 x i16> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i16_0 = extractelement <vscale x 32 x i16> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv64i16_0 = extractelement <vscale x 64 x i16> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i32_0 = extractelement <2 x i32> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i32_0 = extractelement <4 x i32> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i32_0 = extractelement <8 x i32> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i32_0 = extractelement <16 x i32> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_0 = extractelement <32 x i32> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i32_0 = extractelement <vscale x 2 x i32> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i32_0 = extractelement <vscale x 4 x i32> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i32_0 = extractelement <vscale x 8 x i32> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i32_0 = extractelement <vscale x 16 x i32> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i32_0 = extractelement <vscale x 32 x i32> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i64_0 = extractelement <2 x i64> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i64_0 = extractelement <4 x i64> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i64_0 = extractelement <8 x i64> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i64_0 = extractelement <16 x i64> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i64_0 = extractelement <vscale x 2 x i64> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i64_0 = extractelement <vscale x 4 x i64> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i64_0 = extractelement <vscale x 8 x i64> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i64_0 = extractelement <vscale x 16 x i64> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v2i1_1 = extractelement <2 x i1> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v4i1_1 = extractelement <4 x i1> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v8i1_1 = extractelement <8 x i1> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %v16i1_1 = extractelement <16 x i1> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %v32i1_1 = extractelement <32 x i1> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv2i1_1 = extractelement <vscale x 2 x i1> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv4i1_1 = extractelement <vscale x 4 x i1> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv8i1_1 = extractelement <vscale x 8 x i1> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %nxv16i1_1 = extractelement <vscale x 16 x i1> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %nxv32i1_1 = extractelement <vscale x 32 x i1> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i8_1 = extractelement <2 x i8> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i8_1 = extractelement <4 x i8> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i8_1 = extractelement <8 x i8> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i8_1 = extractelement <16 x i8> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i8_1 = extractelement <32 x i8> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v64i8_1 = extractelement <64 x i8> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v128i8_1 = extractelement <128 x i8> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i8_1 = extractelement <vscale x 2 x i8> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i8_1 = extractelement <vscale x 4 x i8> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i8_1 = extractelement <vscale x 8 x i8> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i8_1 = extractelement <vscale x 16 x i8> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32i8_1 = extractelement <vscale x 32 x i8> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv64i8_1 = extractelement <vscale x 64 x i8> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv128i8_1 = extractelement <vscale x 128 x i8> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i16_1 = extractelement <2 x i16> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i16_1 = extractelement <4 x i16> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i16_1 = extractelement <8 x i16> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i16_1 = extractelement <16 x i16> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i16_1 = extractelement <32 x i16> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v64i16_1 = extractelement <64 x i16> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i16_1 = extractelement <vscale x 2 x i16> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i16_1 = extractelement <vscale x 4 x i16> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i16_1 = extractelement <vscale x 8 x i16> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i16_1 = extractelement <vscale x 16 x i16> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32i16_1 = extractelement <vscale x 32 x i16> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv64i16_1 = extractelement <vscale x 64 x i16> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i32_1 = extractelement <2 x i32> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i32_1 = extractelement <4 x i32> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i32_1 = extractelement <8 x i32> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i32_1 = extractelement <16 x i32> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_1 = extractelement <32 x i32> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i32_1 = extractelement <vscale x 2 x i32> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i32_1 = extractelement <vscale x 4 x i32> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i32_1 = extractelement <vscale x 8 x i32> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i32_1 = extractelement <vscale x 16 x i32> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32i32_1 = extractelement <vscale x 32 x i32> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i64_1 = extractelement <2 x i64> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i64_1 = extractelement <4 x i64> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i64_1 = extractelement <8 x i64> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i64_1 = extractelement <16 x i64> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i64_1 = extractelement <vscale x 2 x i64> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i64_1 = extractelement <vscale x 4 x i64> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i64_1 = extractelement <vscale x 8 x i64> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i64_1 = extractelement <vscale x 16 x i64> undef, i32 1
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v2i1_x = extractelement <2 x i1> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v4i1_x = extractelement <4 x i1> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %v8i1_x = extractelement <8 x i1> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %v16i1_x = extractelement <16 x i1> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %v32i1_x = extractelement <32 x i1> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv2i1_x = extractelement <vscale x 2 x i1> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv4i1_x = extractelement <vscale x 4 x i1> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %nxv8i1_x = extractelement <vscale x 8 x i1> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %nxv16i1_x = extractelement <vscale x 16 x i1> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %nxv32i1_x = extractelement <vscale x 32 x i1> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i8_x = extractelement <2 x i8> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i8_x = extractelement <4 x i8> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i8_x = extractelement <8 x i8> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i8_x = extractelement <16 x i8> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i8_x = extractelement <32 x i8> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v64i8_x = extractelement <64 x i8> undef, i32 %x
; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %v128i8_x = extractelement <128 x i8> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i8_x = extractelement <vscale x 2 x i8> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i8_x = extractelement <vscale x 4 x i8> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i8_x = extractelement <vscale x 8 x i8> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i8_x = extractelement <vscale x 16 x i8> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i8_x = extractelement <vscale x 32 x i8> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv64i8_x = extractelement <vscale x 64 x i8> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i8_x = extractelement <vscale x 2 x i8> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i8_x = extractelement <vscale x 4 x i8> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i8_x = extractelement <vscale x 8 x i8> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i8_x = extractelement <vscale x 16 x i8> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32i8_x = extractelement <vscale x 32 x i8> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv64i8_x = extractelement <vscale x 64 x i8> undef, i32 %x
; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %nxv128i8_x = extractelement <vscale x 128 x i8> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i16_x = extractelement <2 x i16> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i16_x = extractelement <4 x i16> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i16_x = extractelement <8 x i16> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i16_x = extractelement <16 x i16> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i16_x = extractelement <32 x i16> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i16_x = extractelement <2 x i16> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i16_x = extractelement <4 x i16> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i16_x = extractelement <8 x i16> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i16_x = extractelement <16 x i16> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i16_x = extractelement <32 x i16> undef, i32 %x
; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %v64i16_x = extractelement <64 x i16> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i16_x = extractelement <vscale x 2 x i16> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i16_x = extractelement <vscale x 4 x i16> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i16_x = extractelement <vscale x 8 x i16> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i16_x = extractelement <vscale x 16 x i16> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32i16_x = extractelement <vscale x 32 x i16> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i16_x = extractelement <vscale x 2 x i16> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i16_x = extractelement <vscale x 4 x i16> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i16_x = extractelement <vscale x 8 x i16> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i16_x = extractelement <vscale x 16 x i16> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32i16_x = extractelement <vscale x 32 x i16> undef, i32 %x
; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %nxv64i16_x = extractelement <vscale x 64 x i16> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i32_x = extractelement <2 x i32> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i32_x = extractelement <4 x i32> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i32_x = extractelement <8 x i32> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i32_x = extractelement <16 x i32> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i32_x = extractelement <2 x i32> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i32_x = extractelement <4 x i32> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i32_x = extractelement <8 x i32> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16i32_x = extractelement <16 x i32> undef, i32 %x
; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %v32i32_x = extractelement <32 x i32> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i32_x = extractelement <vscale x 2 x i32> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i32_x = extractelement <vscale x 4 x i32> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i32_x = extractelement <vscale x 8 x i32> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16i32_x = extractelement <vscale x 16 x i32> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i32_x = extractelement <vscale x 2 x i32> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i32_x = extractelement <vscale x 4 x i32> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i32_x = extractelement <vscale x 8 x i32> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16i32_x = extractelement <vscale x 16 x i32> undef, i32 %x
; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %nxv32i32_x = extractelement <vscale x 32 x i32> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2i64_x = extractelement <2 x i64> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i64_x = extractelement <4 x i64> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i64_x = extractelement <8 x i64> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2i64_x = extractelement <2 x i64> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4i64_x = extractelement <4 x i64> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8i64_x = extractelement <8 x i64> undef, i32 %x
; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %v16i64_x = extractelement <16 x i64> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2i64_x = extractelement <vscale x 2 x i64> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4i64_x = extractelement <vscale x 4 x i64> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8i64_x = extractelement <vscale x 8 x i64> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2i64_x = extractelement <vscale x 2 x i64> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4i64_x = extractelement <vscale x 4 x i64> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8i64_x = extractelement <vscale x 8 x i64> undef, i32 %x
; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %nxv16i64_x = extractelement <vscale x 16 x i64> undef, i32 %x
; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
@@ -866,51 +865,50 @@ define void @extractelement_int(i32 %x) {
; Cover high lmul extracts and types which require splitting
define void @extractelement_int_lmul(i32 %x) {
; RV32V-LABEL: 'extractelement_int_lmul'
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v128i8_31 = extractelement <128 x i8> undef, i32 31
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v128i8_63 = extractelement <128 x i8> undef, i32 63
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v128i8_127 = extractelement <128 x i8> undef, i32 127
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v256i8_127 = extractelement <256 x i8> undef, i32 127
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v256i8_255 = extractelement <256 x i8> undef, i32 255
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_3 = extractelement <32 x i32> undef, i32 3
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_7 = extractelement <32 x i32> undef, i32 7
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_15 = extractelement <32 x i32> undef, i32 15
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_31 = extractelement <32 x i32> undef, i32 31
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i32_63 = extractelement <64 x i32> undef, i32 63
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v128i8 = extractelement <128 x i8> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v128i8_31 = extractelement <128 x i8> undef, i32 31
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v128i8_63 = extractelement <128 x i8> undef, i32 63
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v128i8_127 = extractelement <128 x i8> undef, i32 127
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v256i8_127 = extractelement <256 x i8> undef, i32 127
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v256i8_255 = extractelement <256 x i8> undef, i32 255
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_3 = extractelement <32 x i32> undef, i32 3
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_7 = extractelement <32 x i32> undef, i32 7
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_15 = extractelement <32 x i32> undef, i32 15
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_31 = extractelement <32 x i32> undef, i32 31
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v64i32_63 = extractelement <64 x i32> undef, i32 63
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v128i8 = extractelement <128 x i8> undef, i32 %x
; RV32V-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %v256i8 = extractelement <256 x i8> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32 = extractelement <32 x i32> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32 = extractelement <32 x i32> undef, i32 %x
; RV32V-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %v64i32 = extractelement <64 x i32> undef, i32 %x
; RV32V-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; RV64V-LABEL: 'extractelement_int_lmul'
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v128i8_31 = extractelement <128 x i8> undef, i32 31
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v128i8_63 = extractelement <128 x i8> undef, i32 63
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v128i8_127 = extractelement <128 x i8> undef, i32 127
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v256i8_127 = extractelement <256 x i8> undef, i32 127
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v256i8_255 = extractelement <256 x i8> undef, i32 255
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_3 = extractelement <32 x i32> undef, i32 3
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_7 = extractelement <32 x i32> undef, i32 7
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_15 = extractelement <32 x i32> undef, i32 15
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_31 = extractelement <32 x i32> undef, i32 31
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i32_63 = extractelement <64 x i32> undef, i32 63
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v128i8 = extractelement <128 x i8> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v128i8_31 = extractelement <128 x i8> undef, i32 31
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v128i8_63 = extractelement <128 x i8> undef, i32 63
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v128i8_127 = extractelement <128 x i8> undef, i32 127
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v256i8_127 = extractelement <256 x i8> undef, i32 127
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v256i8_255 = extractelement <256 x i8> undef, i32 255
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_3 = extractelement <32 x i32> undef, i32 3
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_7 = extractelement <32 x i32> undef, i32 7
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_15 = extractelement <32 x i32> undef, i32 15
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_31 = extractelement <32 x i32> undef, i32 31
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v64i32_63 = extractelement <64 x i32> undef, i32 63
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v128i8 = extractelement <128 x i8> undef, i32 %x
; RV64V-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %v256i8 = extractelement <256 x i8> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32 = extractelement <32 x i32> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32 = extractelement <32 x i32> undef, i32 %x
; RV64V-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %v64i32 = extractelement <64 x i32> undef, i32 %x
; RV64V-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
-;
; RV32ZVE64X-LABEL: 'extractelement_int_lmul'
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v128i8_31 = extractelement <128 x i8> undef, i32 31
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v128i8_63 = extractelement <128 x i8> undef, i32 63
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v128i8_127 = extractelement <128 x i8> undef, i32 127
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v256i8_127 = extractelement <256 x i8> undef, i32 127
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v256i8_255 = extractelement <256 x i8> undef, i32 255
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_3 = extractelement <32 x i32> undef, i32 3
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_7 = extractelement <32 x i32> undef, i32 7
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_15 = extractelement <32 x i32> undef, i32 15
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_31 = extractelement <32 x i32> undef, i32 31
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i32_63 = extractelement <64 x i32> undef, i32 63
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v128i8_31 = extractelement <128 x i8> undef, i32 31
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v128i8_63 = extractelement <128 x i8> undef, i32 63
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v128i8_127 = extractelement <128 x i8> undef, i32 127
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v256i8_127 = extractelement <256 x i8> undef, i32 127
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v256i8_255 = extractelement <256 x i8> undef, i32 255
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_3 = extractelement <32 x i32> undef, i32 3
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_7 = extractelement <32 x i32> undef, i32 7
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_15 = extractelement <32 x i32> undef, i32 15
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_31 = extractelement <32 x i32> undef, i32 31
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v64i32_63 = extractelement <64 x i32> undef, i32 63
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %v128i8 = extractelement <128 x i8> undef, i32 %x
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 34 for instruction: %v256i8 = extractelement <256 x i8> undef, i32 %x
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %v32i32 = extractelement <32 x i32> undef, i32 %x
@@ -918,16 +916,16 @@ define void @extractelement_int_lmul(i32 %x) {
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; RV64ZVE64X-LABEL: 'extractelement_int_lmul'
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v128i8_31 = extractelement <128 x i8> undef, i32 31
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v128i8_63 = extractelement <128 x i8> undef, i32 63
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v128i8_127 = extractelement <128 x i8> undef, i32 127
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v256i8_127 = extractelement <256 x i8> undef, i32 127
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v256i8_255 = extractelement <256 x i8> undef, i32 255
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_3 = extractelement <32 x i32> undef, i32 3
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_7 = extractelement <32 x i32> undef, i32 7
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_15 = extractelement <32 x i32> undef, i32 15
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_31 = extractelement <32 x i32> undef, i32 31
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64i32_63 = extractelement <64 x i32> undef, i32 63
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v128i8_31 = extractelement <128 x i8> undef, i32 31
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v128i8_63 = extractelement <128 x i8> undef, i32 63
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v128i8_127 = extractelement <128 x i8> undef, i32 127
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v256i8_127 = extractelement <256 x i8> undef, i32 127
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v256i8_255 = extractelement <256 x i8> undef, i32 255
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_3 = extractelement <32 x i32> undef, i32 3
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_7 = extractelement <32 x i32> undef, i32 7
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_15 = extractelement <32 x i32> undef, i32 15
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_31 = extractelement <32 x i32> undef, i32 31
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v64i32_63 = extractelement <64 x i32> undef, i32 63
; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %v128i8 = extractelement <128 x i8> undef, i32 %x
; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 34 for instruction: %v256i8 = extractelement <256 x i8> undef, i32 %x
; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %v32i32 = extractelement <32 x i32> undef, i32 %x
@@ -958,192 +956,191 @@ define void @extractelement_int_lmul(i32 %x) {
define void @extractelement_fp(i32 %x) {
; RV32V-LABEL: 'extractelement_fp'
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v2f16_0 = extractelement <2 x half> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v4f16_0 = extractelement <4 x half> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v8f16_0 = extractelement <8 x half> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v16f16_0 = extractelement <16 x half> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32f16_0 = extractelement <32 x half> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v64f16_0 = extractelement <64 x half> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv2f16_0 = extractelement <vscale x 2 x half> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv4f16_0 = extractelement <vscale x 4 x half> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv8f16_0 = extractelement <vscale x 8 x half> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv16f16_0 = extractelement <vscale x 16 x half> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv32f16_0 = extractelement <vscale x 32 x half> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv64f16_0 = extractelement <vscale x 64 x half> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v2f32_0 = extractelement <2 x float> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v4f32_0 = extractelement <4 x float> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v8f32_0 = extractelement <8 x float> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v16f32_0 = extractelement <16 x float> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32f32_0 = extractelement <32 x float> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv2f32_0 = extractelement <vscale x 2 x float> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv4f32_0 = extractelement <vscale x 4 x float> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv8f32_0 = extractelement <vscale x 8 x float> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv16f32_0 = extractelement <vscale x 16 x float> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv32f32_0 = extractelement <vscale x 32 x float> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v2f64_0 = extractelement <2 x double> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v4f64_0 = extractelement <4 x double> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v8f64_0 = extractelement <8 x double> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v16f64_0 = extractelement <16 x double> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv2f64_0 = extractelement <vscale x 2 x double> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv4f64_0 = extractelement <vscale x 4 x double> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv8f64_0 = extractelement <vscale x 8 x double> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv16f64_0 = extractelement <vscale x 16 x double> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2f16_1 = extractelement <2 x half> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4f16_1 = extractelement <4 x half> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8f16_1 = extractelement <8 x half> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16f16_1 = extractelement <16 x half> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32f16_1 = extractelement <32 x half> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64f16_1 = extractelement <64 x half> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2f16_1 = extractelement <vscale x 2 x half> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4f16_1 = extractelement <vscale x 4 x half> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8f16_1 = extractelement <vscale x 8 x half> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16f16_1 = extractelement <vscale x 16 x half> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32f16_1 = extractelement <vscale x 32 x half> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv64f16_1 = extractelement <vscale x 64 x half> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2f32_1 = extractelement <2 x float> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4f32_1 = extractelement <4 x float> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8f32_1 = extractelement <8 x float> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16f32_1 = extractelement <16 x float> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32f32_1 = extractelement <32 x float> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2f32_1 = extractelement <vscale x 2 x float> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4f32_1 = extractelement <vscale x 4 x float> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8f32_1 = extractelement <vscale x 8 x float> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16f32_1 = extractelement <vscale x 16 x float> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32f32_1 = extractelement <vscale x 32 x float> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2f64_1 = extractelement <2 x double> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4f64_1 = extractelement <4 x double> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8f64_1 = extractelement <8 x double> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16f64_1 = extractelement <16 x double> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2f64_1 = extractelement <vscale x 2 x double> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4f64_1 = extractelement <vscale x 4 x double> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8f64_1 = extractelement <vscale x 8 x double> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16f64_1 = extractelement <vscale x 16 x double> undef, i32 1
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2f16_x = extractelement <2 x half> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4f16_x = extractelement <4 x half> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8f16_x = extractelement <8 x half> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16f16_x = extractelement <16 x half> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32f16_x = extractelement <32 x half> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64f16_x = extractelement <64 x half> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2f16_x = extractelement <vscale x 2 x half> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4f16_x = extractelement <vscale x 4 x half> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8f16_x = extractelement <vscale x 8 x half> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16f16_x = extractelement <vscale x 16 x half> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32f16_x = extractelement <vscale x 32 x half> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2f16_0 = extractelement <2 x half> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4f16_0 = extractelement <4 x half> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8f16_0 = extractelement <8 x half> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16f16_0 = extractelement <16 x half> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32f16_0 = extractelement <32 x half> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64f16_0 = extractelement <64 x half> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2f16_0 = extractelement <vscale x 2 x half> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4f16_0 = extractelement <vscale x 4 x half> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8f16_0 = extractelement <vscale x 8 x half> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16f16_0 = extractelement <vscale x 16 x half> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32f16_0 = extractelement <vscale x 32 x half> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv64f16_0 = extractelement <vscale x 64 x half> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2f32_0 = extractelement <2 x float> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4f32_0 = extractelement <4 x float> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8f32_0 = extractelement <8 x float> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16f32_0 = extractelement <16 x float> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32f32_0 = extractelement <32 x float> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2f32_0 = extractelement <vscale x 2 x float> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4f32_0 = extractelement <vscale x 4 x float> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8f32_0 = extractelement <vscale x 8 x float> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16f32_0 = extractelement <vscale x 16 x float> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32f32_0 = extractelement <vscale x 32 x float> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2f64_0 = extractelement <2 x double> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4f64_0 = extractelement <4 x double> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8f64_0 = extractelement <8 x double> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16f64_0 = extractelement <16 x double> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2f64_0 = extractelement <vscale x 2 x double> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4f64_0 = extractelement <vscale x 4 x double> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8f64_0 = extractelement <vscale x 8 x double> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16f64_0 = extractelement <vscale x 16 x double> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2f16_1 = extractelement <2 x half> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4f16_1 = extractelement <4 x half> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8f16_1 = extractelement <8 x half> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16f16_1 = extractelement <16 x half> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32f16_1 = extractelement <32 x half> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v64f16_1 = extractelement <64 x half> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2f16_1 = extractelement <vscale x 2 x half> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4f16_1 = extractelement <vscale x 4 x half> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8f16_1 = extractelement <vscale x 8 x half> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16f16_1 = extractelement <vscale x 16 x half> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32f16_1 = extractelement <vscale x 32 x half> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv64f16_1 = extractelement <vscale x 64 x half> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2f32_1 = extractelement <2 x float> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4f32_1 = extractelement <4 x float> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8f32_1 = extractelement <8 x float> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16f32_1 = extractelement <16 x float> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32f32_1 = extractelement <32 x float> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2f32_1 = extractelement <vscale x 2 x float> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4f32_1 = extractelement <vscale x 4 x float> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8f32_1 = extractelement <vscale x 8 x float> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16f32_1 = extractelement <vscale x 16 x float> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32f32_1 = extractelement <vscale x 32 x float> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2f64_1 = extractelement <2 x double> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4f64_1 = extractelement <4 x double> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8f64_1 = extractelement <8 x double> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16f64_1 = extractelement <16 x double> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2f64_1 = extractelement <vscale x 2 x double> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4f64_1 = extractelement <vscale x 4 x double> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8f64_1 = extractelement <vscale x 8 x double> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16f64_1 = extractelement <vscale x 16 x double> undef, i32 1
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2f16_x = extractelement <2 x half> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4f16_x = extractelement <4 x half> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8f16_x = extractelement <8 x half> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16f16_x = extractelement <16 x half> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32f16_x = extractelement <32 x half> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v64f16_x = extractelement <64 x half> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2f16_x = extractelement <vscale x 2 x half> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4f16_x = extractelement <vscale x 4 x half> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8f16_x = extractelement <vscale x 8 x half> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16f16_x = extractelement <vscale x 16 x half> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32f16_x = extractelement <vscale x 32 x half> undef, i32 %x
; RV32V-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %nxv64f16_x = extractelement <vscale x 64 x half> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2f32_x = extractelement <2 x float> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4f32_x = extractelement <4 x float> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8f32_x = extractelement <8 x float> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16f32_x = extractelement <16 x float> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32f32_x = extractelement <32 x float> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2f32_x = extractelement <vscale x 2 x float> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4f32_x = extractelement <vscale x 4 x float> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8f32_x = extractelement <vscale x 8 x float> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16f32_x = extractelement <vscale x 16 x float> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2f32_x = extractelement <2 x float> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4f32_x = extractelement <4 x float> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8f32_x = extractelement <8 x float> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16f32_x = extractelement <16 x float> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32f32_x = extractelement <32 x float> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2f32_x = extractelement <vscale x 2 x float> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4f32_x = extractelement <vscale x 4 x float> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8f32_x = extractelement <vscale x 8 x float> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16f32_x = extractelement <vscale x 16 x float> undef, i32 %x
; RV32V-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %nxv32f32_x = extractelement <vscale x 32 x float> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2f64_x = extractelement <2 x double> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4f64_x = extractelement <4 x double> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8f64_x = extractelement <8 x double> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16f64_x = extractelement <16 x double> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2f64_x = extractelement <vscale x 2 x double> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4f64_x = extractelement <vscale x 4 x double> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8f64_x = extractelement <vscale x 8 x double> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2f64_x = extractelement <2 x double> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4f64_x = extractelement <4 x double> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8f64_x = extractelement <8 x double> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16f64_x = extractelement <16 x double> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2f64_x = extractelement <vscale x 2 x double> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4f64_x = extractelement <vscale x 4 x double> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8f64_x = extractelement <vscale x 8 x double> undef, i32 %x
; RV32V-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %nxv16f64_x = extractelement <vscale x 16 x double> undef, i32 %x
; RV32V-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; RV64V-LABEL: 'extractelement_fp'
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v2f16_0 = extractelement <2 x half> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v4f16_0 = extractelement <4 x half> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v8f16_0 = extractelement <8 x half> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v16f16_0 = extractelement <16 x half> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32f16_0 = extractelement <32 x half> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v64f16_0 = extractelement <64 x half> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv2f16_0 = extractelement <vscale x 2 x half> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv4f16_0 = extractelement <vscale x 4 x half> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv8f16_0 = extractelement <vscale x 8 x half> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv16f16_0 = extractelement <vscale x 16 x half> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv32f16_0 = extractelement <vscale x 32 x half> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv64f16_0 = extractelement <vscale x 64 x half> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v2f32_0 = extractelement <2 x float> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v4f32_0 = extractelement <4 x float> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v8f32_0 = extractelement <8 x float> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v16f32_0 = extractelement <16 x float> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32f32_0 = extractelement <32 x float> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv2f32_0 = extractelement <vscale x 2 x float> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv4f32_0 = extractelement <vscale x 4 x float> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv8f32_0 = extractelement <vscale x 8 x float> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv16f32_0 = extractelement <vscale x 16 x float> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv32f32_0 = extractelement <vscale x 32 x float> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v2f64_0 = extractelement <2 x double> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v4f64_0 = extractelement <4 x double> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v8f64_0 = extractelement <8 x double> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v16f64_0 = extractelement <16 x double> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv2f64_0 = extractelement <vscale x 2 x double> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv4f64_0 = extractelement <vscale x 4 x double> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv8f64_0 = extractelement <vscale x 8 x double> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %nxv16f64_0 = extractelement <vscale x 16 x double> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2f16_1 = extractelement <2 x half> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4f16_1 = extractelement <4 x half> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8f16_1 = extractelement <8 x half> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16f16_1 = extractelement <16 x half> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32f16_1 = extractelement <32 x half> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64f16_1 = extractelement <64 x half> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2f16_1 = extractelement <vscale x 2 x half> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4f16_1 = extractelement <vscale x 4 x half> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8f16_1 = extractelement <vscale x 8 x half> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16f16_1 = extractelement <vscale x 16 x half> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32f16_1 = extractelement <vscale x 32 x half> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv64f16_1 = extractelement <vscale x 64 x half> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2f32_1 = extractelement <2 x float> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4f32_1 = extractelement <4 x float> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8f32_1 = extractelement <8 x float> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16f32_1 = extractelement <16 x float> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32f32_1 = extractelement <32 x float> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2f32_1 = extractelement <vscale x 2 x float> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4f32_1 = extractelement <vscale x 4 x float> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8f32_1 = extractelement <vscale x 8 x float> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16f32_1 = extractelement <vscale x 16 x float> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32f32_1 = extractelement <vscale x 32 x float> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2f64_1 = extractelement <2 x double> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4f64_1 = extractelement <4 x double> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8f64_1 = extractelement <8 x double> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16f64_1 = extractelement <16 x double> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2f64_1 = extractelement <vscale x 2 x double> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4f64_1 = extractelement <vscale x 4 x double> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8f64_1 = extractelement <vscale x 8 x double> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16f64_1 = extractelement <vscale x 16 x double> undef, i32 1
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2f16_x = extractelement <2 x half> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4f16_x = extractelement <4 x half> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8f16_x = extractelement <8 x half> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16f16_x = extractelement <16 x half> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32f16_x = extractelement <32 x half> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64f16_x = extractelement <64 x half> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2f16_x = extractelement <vscale x 2 x half> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4f16_x = extractelement <vscale x 4 x half> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8f16_x = extractelement <vscale x 8 x half> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16f16_x = extractelement <vscale x 16 x half> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32f16_x = extractelement <vscale x 32 x half> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2f16_0 = extractelement <2 x half> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4f16_0 = extractelement <4 x half> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8f16_0 = extractelement <8 x half> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16f16_0 = extractelement <16 x half> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32f16_0 = extractelement <32 x half> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v64f16_0 = extractelement <64 x half> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2f16_0 = extractelement <vscale x 2 x half> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4f16_0 = extractelement <vscale x 4 x half> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8f16_0 = extractelement <vscale x 8 x half> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16f16_0 = extractelement <vscale x 16 x half> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32f16_0 = extractelement <vscale x 32 x half> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv64f16_0 = extractelement <vscale x 64 x half> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2f32_0 = extractelement <2 x float> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4f32_0 = extractelement <4 x float> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8f32_0 = extractelement <8 x float> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16f32_0 = extractelement <16 x float> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32f32_0 = extractelement <32 x float> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2f32_0 = extractelement <vscale x 2 x float> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4f32_0 = extractelement <vscale x 4 x float> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8f32_0 = extractelement <vscale x 8 x float> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16f32_0 = extractelement <vscale x 16 x float> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv32f32_0 = extractelement <vscale x 32 x float> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2f64_0 = extractelement <2 x double> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4f64_0 = extractelement <4 x double> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8f64_0 = extractelement <8 x double> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16f64_0 = extractelement <16 x double> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2f64_0 = extractelement <vscale x 2 x double> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4f64_0 = extractelement <vscale x 4 x double> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8f64_0 = extractelement <vscale x 8 x double> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16f64_0 = extractelement <vscale x 16 x double> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2f16_1 = extractelement <2 x half> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4f16_1 = extractelement <4 x half> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8f16_1 = extractelement <8 x half> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16f16_1 = extractelement <16 x half> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32f16_1 = extractelement <32 x half> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v64f16_1 = extractelement <64 x half> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2f16_1 = extractelement <vscale x 2 x half> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4f16_1 = extractelement <vscale x 4 x half> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8f16_1 = extractelement <vscale x 8 x half> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16f16_1 = extractelement <vscale x 16 x half> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32f16_1 = extractelement <vscale x 32 x half> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv64f16_1 = extractelement <vscale x 64 x half> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2f32_1 = extractelement <2 x float> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4f32_1 = extractelement <4 x float> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8f32_1 = extractelement <8 x float> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16f32_1 = extractelement <16 x float> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32f32_1 = extractelement <32 x float> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2f32_1 = extractelement <vscale x 2 x float> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4f32_1 = extractelement <vscale x 4 x float> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8f32_1 = extractelement <vscale x 8 x float> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16f32_1 = extractelement <vscale x 16 x float> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32f32_1 = extractelement <vscale x 32 x float> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2f64_1 = extractelement <2 x double> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4f64_1 = extractelement <4 x double> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8f64_1 = extractelement <8 x double> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16f64_1 = extractelement <16 x double> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2f64_1 = extractelement <vscale x 2 x double> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4f64_1 = extractelement <vscale x 4 x double> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8f64_1 = extractelement <vscale x 8 x double> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16f64_1 = extractelement <vscale x 16 x double> undef, i32 1
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2f16_x = extractelement <2 x half> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4f16_x = extractelement <4 x half> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8f16_x = extractelement <8 x half> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16f16_x = extractelement <16 x half> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32f16_x = extractelement <32 x half> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v64f16_x = extractelement <64 x half> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2f16_x = extractelement <vscale x 2 x half> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4f16_x = extractelement <vscale x 4 x half> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8f16_x = extractelement <vscale x 8 x half> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16f16_x = extractelement <vscale x 16 x half> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv32f16_x = extractelement <vscale x 32 x half> undef, i32 %x
; RV64V-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %nxv64f16_x = extractelement <vscale x 64 x half> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2f32_x = extractelement <2 x float> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4f32_x = extractelement <4 x float> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8f32_x = extractelement <8 x float> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16f32_x = extractelement <16 x float> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32f32_x = extractelement <32 x float> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2f32_x = extractelement <vscale x 2 x float> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4f32_x = extractelement <vscale x 4 x float> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8f32_x = extractelement <vscale x 8 x float> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv16f32_x = extractelement <vscale x 16 x float> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2f32_x = extractelement <2 x float> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4f32_x = extractelement <4 x float> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8f32_x = extractelement <8 x float> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16f32_x = extractelement <16 x float> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32f32_x = extractelement <32 x float> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2f32_x = extractelement <vscale x 2 x float> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4f32_x = extractelement <vscale x 4 x float> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8f32_x = extractelement <vscale x 8 x float> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv16f32_x = extractelement <vscale x 16 x float> undef, i32 %x
; RV64V-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %nxv32f32_x = extractelement <vscale x 32 x float> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2f64_x = extractelement <2 x double> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4f64_x = extractelement <4 x double> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8f64_x = extractelement <8 x double> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16f64_x = extractelement <16 x double> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv2f64_x = extractelement <vscale x 2 x double> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv4f64_x = extractelement <vscale x 4 x double> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %nxv8f64_x = extractelement <vscale x 8 x double> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v2f64_x = extractelement <2 x double> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v4f64_x = extractelement <4 x double> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v8f64_x = extractelement <8 x double> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v16f64_x = extractelement <16 x double> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv2f64_x = extractelement <vscale x 2 x double> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv4f64_x = extractelement <vscale x 4 x double> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %nxv8f64_x = extractelement <vscale x 8 x double> undef, i32 %x
; RV64V-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %nxv16f64_x = extractelement <vscale x 16 x double> undef, i32 %x
; RV64V-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
-;
; RV32ZVE64X-LABEL: 'extractelement_fp'
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %v2f16_0 = extractelement <2 x half> undef, i32 0
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %v4f16_0 = extractelement <4 x half> undef, i32 0
@@ -1444,52 +1441,51 @@ define void @extractelement_fp(i32 %x) {
define void @extractelement_int_nonpoweroftwo(i32 %x) {
; RV32V-LABEL: 'extractelement_int_nonpoweroftwo'
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v3i8 = extractelement <3 x i8> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v7i8 = extractelement <7 x i8> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v15i8 = extractelement <15 x i8> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v31i8 = extractelement <31 x i8> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v3i32 = extractelement <3 x i32> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v7i32 = extractelement <7 x i32> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v15i32 = extractelement <15 x i32> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v31i32 = extractelement <31 x i32> undef, i32 %x
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v31i32_0 = extractelement <31 x i32> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v3i8 = extractelement <3 x i8> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v7i8 = extractelement <7 x i8> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v15i8 = extractelement <15 x i8> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v31i8 = extractelement <31 x i8> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v3i32 = extractelement <3 x i32> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v7i32 = extractelement <7 x i32> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v15i32 = extractelement <15 x i32> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v31i32 = extractelement <31 x i32> undef, i32 %x
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v31i32_0 = extractelement <31 x i32> undef, i32 0
; RV32V-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; RV64V-LABEL: 'extractelement_int_nonpoweroftwo'
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v3i8 = extractelement <3 x i8> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v7i8 = extractelement <7 x i8> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v15i8 = extractelement <15 x i8> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v31i8 = extractelement <31 x i8> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v3i32 = extractelement <3 x i32> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v7i32 = extractelement <7 x i32> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v15i32 = extractelement <15 x i32> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v31i32 = extractelement <31 x i32> undef, i32 %x
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v31i32_0 = extractelement <31 x i32> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v3i8 = extractelement <3 x i8> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v7i8 = extractelement <7 x i8> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v15i8 = extractelement <15 x i8> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v31i8 = extractelement <31 x i8> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v3i32 = extractelement <3 x i32> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v7i32 = extractelement <7 x i32> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v15i32 = extractelement <15 x i32> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v31i32 = extractelement <31 x i32> undef, i32 %x
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v31i32_0 = extractelement <31 x i32> undef, i32 0
; RV64V-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
-;
; RV32ZVE64X-LABEL: 'extractelement_int_nonpoweroftwo'
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v3i8 = extractelement <3 x i8> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v7i8 = extractelement <7 x i8> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v15i8 = extractelement <15 x i8> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v31i8 = extractelement <31 x i8> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v3i32 = extractelement <3 x i32> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v7i32 = extractelement <7 x i32> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v15i32 = extractelement <15 x i32> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v3i8 = extractelement <3 x i8> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v7i8 = extractelement <7 x i8> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v15i8 = extractelement <15 x i8> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v31i8 = extractelement <31 x i8> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v3i32 = extractelement <3 x i32> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v7i32 = extractelement <7 x i32> undef, i32 %x
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v15i32 = extractelement <15 x i32> undef, i32 %x
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %v31i32 = extractelement <31 x i32> undef, i32 %x
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v31i32_0 = extractelement <31 x i32> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v31i32_0 = extractelement <31 x i32> undef, i32 0
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; RV64ZVE64X-LABEL: 'extractelement_int_nonpoweroftwo'
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v3i8 = extractelement <3 x i8> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v7i8 = extractelement <7 x i8> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v15i8 = extractelement <15 x i8> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v31i8 = extractelement <31 x i8> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v3i32 = extractelement <3 x i32> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v7i32 = extractelement <7 x i32> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v15i32 = extractelement <15 x i32> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v3i8 = extractelement <3 x i8> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v7i8 = extractelement <7 x i8> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v15i8 = extractelement <15 x i8> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v31i8 = extractelement <31 x i8> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v3i32 = extractelement <3 x i32> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v7i32 = extractelement <7 x i32> undef, i32 %x
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v15i32 = extractelement <15 x i32> undef, i32 %x
; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %v31i32 = extractelement <31 x i32> undef, i32 %x
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v31i32_0 = extractelement <31 x i32> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v31i32_0 = extractelement <31 x i32> undef, i32 0
; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%v3i8 = extractelement <3 x i8> undef, i32 %x
@@ -1508,44 +1504,43 @@ define void @extractelement_int_nonpoweroftwo(i32 %x) {
define void @extractelement_vls(i32 %x) vscale_range(2,2) {
; RV32V-LABEL: 'extractelement_vls'
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i32_0 = extractelement <32 x i32> undef, i32 0
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i32_4 = extractelement <32 x i32> undef, i32 4
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_5 = extractelement <32 x i32> undef, i32 5
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i32_8 = extractelement <32 x i32> undef, i32 8
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_9 = extractelement <32 x i32> undef, i32 9
-; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_11 = extractelement <32 x i32> undef, i32 11
-; RV32V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i32_12 = extractelement <32 x i32> undef, i32 12
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_0 = extractelement <32 x i32> undef, i32 0
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_4 = extractelement <32 x i32> undef, i32 4
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_5 = extractelement <32 x i32> undef, i32 5
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_8 = extractelement <32 x i32> undef, i32 8
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_9 = extractelement <32 x i32> undef, i32 9
+; RV32V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_11 = extractelement <32 x i32> undef, i32 11
+; RV32V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_12 = extractelement <32 x i32> undef, i32 12
; RV32V-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; RV64V-LABEL: 'extractelement_vls'
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i32_0 = extractelement <32 x i32> undef, i32 0
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i32_4 = extractelement <32 x i32> undef, i32 4
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_5 = extractelement <32 x i32> undef, i32 5
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i32_8 = extractelement <32 x i32> undef, i32 8
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_9 = extractelement <32 x i32> undef, i32 9
-; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_11 = extractelement <32 x i32> undef, i32 11
-; RV64V-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i32_12 = extractelement <32 x i32> undef, i32 12
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_0 = extractelement <32 x i32> undef, i32 0
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_4 = extractelement <32 x i32> undef, i32 4
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_5 = extractelement <32 x i32> undef, i32 5
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_8 = extractelement <32 x i32> undef, i32 8
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_9 = extractelement <32 x i32> undef, i32 9
+; RV64V-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_11 = extractelement <32 x i32> undef, i32 11
+; RV64V-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_12 = extractelement <32 x i32> undef, i32 12
; RV64V-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
-;
; RV32ZVE64X-LABEL: 'extractelement_vls'
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i32_0 = extractelement <32 x i32> undef, i32 0
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i32_4 = extractelement <32 x i32> undef, i32 4
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_5 = extractelement <32 x i32> undef, i32 5
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i32_8 = extractelement <32 x i32> undef, i32 8
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_9 = extractelement <32 x i32> undef, i32 9
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_11 = extractelement <32 x i32> undef, i32 11
-; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i32_12 = extractelement <32 x i32> undef, i32 12
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_0 = extractelement <32 x i32> undef, i32 0
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_4 = extractelement <32 x i32> undef, i32 4
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_5 = extractelement <32 x i32> undef, i32 5
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_8 = extractelement <32 x i32> undef, i32 8
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_9 = extractelement <32 x i32> undef, i32 9
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_11 = extractelement <32 x i32> undef, i32 11
+; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_12 = extractelement <32 x i32> undef, i32 12
; RV32ZVE64X-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; RV64ZVE64X-LABEL: 'extractelement_vls'
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i32_0 = extractelement <32 x i32> undef, i32 0
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i32_4 = extractelement <32 x i32> undef, i32 4
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_5 = extractelement <32 x i32> undef, i32 5
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i32_8 = extractelement <32 x i32> undef, i32 8
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_9 = extractelement <32 x i32> undef, i32 9
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_11 = extractelement <32 x i32> undef, i32 11
-; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %v32i32_12 = extractelement <32 x i32> undef, i32 12
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_0 = extractelement <32 x i32> undef, i32 0
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_4 = extractelement <32 x i32> undef, i32 4
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_5 = extractelement <32 x i32> undef, i32 5
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_8 = extractelement <32 x i32> undef, i32 8
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_9 = extractelement <32 x i32> undef, i32 9
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %v32i32_11 = extractelement <32 x i32> undef, i32 11
+; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v32i32_12 = extractelement <32 x i32> undef, i32 12
; RV64ZVE64X-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%v32i32_0 = extractelement <32 x i32> undef, i32 0
diff --git a/llvm/test/Analysis/CostModel/RISCV/rvv-shuffle.ll b/llvm/test/Analysis/CostModel/RISCV/rvv-shuffle.ll
index 5830a358a0c1a..c9e59693838c9 100644
--- a/llvm/test/Analysis/CostModel/RISCV/rvv-shuffle.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/rvv-shuffle.ll
@@ -15,10 +15,10 @@ define void @vector_broadcast() {
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %5 = shufflevector <vscale x 4 x i32> undef, <vscale x 4 x i32> undef, <vscale x 4 x i32> zeroinitializer
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %6 = shufflevector <vscale x 1 x i64> undef, <vscale x 1 x i64> undef, <vscale x 1 x i32> zeroinitializer
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %7 = shufflevector <vscale x 2 x i64> undef, <vscale x 2 x i64> undef, <vscale x 2 x i32> zeroinitializer
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %8 = shufflevector <vscale x 16 x i1> undef, <vscale x 16 x i1> undef, <vscale x 16 x i32> zeroinitializer
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %9 = shufflevector <vscale x 8 x i1> undef, <vscale x 8 x i1> undef, <vscale x 8 x i32> zeroinitializer
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %10 = shufflevector <vscale x 4 x i1> undef, <vscale x 4 x i1> undef, <vscale x 4 x i32> zeroinitializer
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %11 = shufflevector <vscale x 2 x i1> undef, <vscale x 2 x i1> undef, <vscale x 2 x i32> zeroinitializer
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %8 = shufflevector <vscale x 16 x i1> undef, <vscale x 16 x i1> undef, <vscale x 16 x i32> zeroinitializer
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %9 = shufflevector <vscale x 8 x i1> undef, <vscale x 8 x i1> undef, <vscale x 8 x i32> zeroinitializer
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %10 = shufflevector <vscale x 4 x i1> undef, <vscale x 4 x i1> undef, <vscale x 4 x i32> zeroinitializer
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %11 = shufflevector <vscale x 2 x i1> undef, <vscale x 2 x i1> undef, <vscale x 2 x i32> zeroinitializer
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; SIZE-LABEL: 'vector_broadcast'
diff --git a/llvm/test/Analysis/CostModel/RISCV/shuffle-broadcast.ll b/llvm/test/Analysis/CostModel/RISCV/shuffle-broadcast.ll
index acc340bab4cfd..76fb274634f10 100644
--- a/llvm/test/Analysis/CostModel/RISCV/shuffle-broadcast.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/shuffle-broadcast.ll
@@ -43,13 +43,13 @@ define void @broadcast_scalable() {
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %34 = shufflevector <vscale x 2 x i64> undef, <vscale x 2 x i64> undef, <vscale x 2 x i32> zeroinitializer
; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %35 = shufflevector <vscale x 4 x i64> undef, <vscale x 4 x i64> undef, <vscale x 4 x i32> zeroinitializer
; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %36 = shufflevector <vscale x 8 x i64> undef, <vscale x 8 x i64> undef, <vscale x 8 x i32> zeroinitializer
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %37 = shufflevector <vscale x 1 x i1> undef, <vscale x 1 x i1> undef, <vscale x 1 x i32> zeroinitializer
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %38 = shufflevector <vscale x 2 x i1> undef, <vscale x 2 x i1> undef, <vscale x 2 x i32> zeroinitializer
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %39 = shufflevector <vscale x 4 x i1> undef, <vscale x 4 x i1> undef, <vscale x 4 x i32> zeroinitializer
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %40 = shufflevector <vscale x 8 x i1> undef, <vscale x 8 x i1> undef, <vscale x 8 x i32> zeroinitializer
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %41 = shufflevector <vscale x 16 x i1> undef, <vscale x 16 x i1> undef, <vscale x 16 x i32> zeroinitializer
-; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %42 = shufflevector <vscale x 32 x i1> undef, <vscale x 32 x i1> undef, <vscale x 32 x i32> zeroinitializer
-; CHECK-NEXT: Cost Model: Found an estimated cost of 34 for instruction: %43 = shufflevector <vscale x 64 x i1> undef, <vscale x 64 x i1> undef, <vscale x 64 x i32> zeroinitializer
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %37 = shufflevector <vscale x 1 x i1> undef, <vscale x 1 x i1> undef, <vscale x 1 x i32> zeroinitializer
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %38 = shufflevector <vscale x 2 x i1> undef, <vscale x 2 x i1> undef, <vscale x 2 x i32> zeroinitializer
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %39 = shufflevector <vscale x 4 x i1> undef, <vscale x 4 x i1> undef, <vscale x 4 x i32> zeroinitializer
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %40 = shufflevector <vscale x 8 x i1> undef, <vscale x 8 x i1> undef, <vscale x 8 x i32> zeroinitializer
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %41 = shufflevector <vscale x 16 x i1> undef, <vscale x 16 x i1> undef, <vscale x 16 x i32> zeroinitializer
+; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %42 = shufflevector <vscale x 32 x i1> undef, <vscale x 32 x i1> undef, <vscale x 32 x i32> zeroinitializer
+; CHECK-NEXT: Cost Model: Found an estimated cost of 35 for instruction: %43 = shufflevector <vscale x 64 x i1> undef, <vscale x 64 x i1> undef, <vscale x 64 x i32> zeroinitializer
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; SIZE-LABEL: 'broadcast_scalable'
@@ -192,13 +192,13 @@ define void @broadcast_fixed() {
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %34 = shufflevector <4 x i64> undef, <4 x i64> undef, <4 x i32> zeroinitializer
; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %35 = shufflevector <8 x i64> undef, <8 x i64> undef, <8 x i32> zeroinitializer
; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %36 = shufflevector <16 x i64> undef, <16 x i64> undef, <16 x i32> zeroinitializer
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %37 = shufflevector <2 x i1> undef, <2 x i1> undef, <2 x i32> zeroinitializer
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %38 = shufflevector <4 x i1> undef, <4 x i1> undef, <4 x i32> zeroinitializer
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %39 = shufflevector <8 x i1> undef, <8 x i1> undef, <8 x i32> zeroinitializer
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %40 = shufflevector <16 x i1> undef, <16 x i1> undef, <16 x i32> zeroinitializer
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %41 = shufflevector <32 x i1> undef, <32 x i1> undef, <32 x i32> zeroinitializer
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %42 = shufflevector <64 x i1> undef, <64 x i1> undef, <64 x i32> zeroinitializer
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %43 = shufflevector <128 x i1> undef, <128 x i1> undef, <128 x i32> zeroinitializer
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %37 = shufflevector <2 x i1> undef, <2 x i1> undef, <2 x i32> zeroinitializer
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %38 = shufflevector <4 x i1> undef, <4 x i1> undef, <4 x i32> zeroinitializer
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %39 = shufflevector <8 x i1> undef, <8 x i1> undef, <8 x i32> zeroinitializer
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %40 = shufflevector <16 x i1> undef, <16 x i1> undef, <16 x i32> zeroinitializer
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %41 = shufflevector <32 x i1> undef, <32 x i1> undef, <32 x i32> zeroinitializer
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %42 = shufflevector <64 x i1> undef, <64 x i1> undef, <64 x i32> zeroinitializer
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %43 = shufflevector <128 x i1> undef, <128 x i1> undef, <128 x i32> zeroinitializer
; CHECK-NEXT: Cost Model: Found an estimated cost of 33 for instruction: %ins1 = insertelement <128 x i1> poison, i1 poison, i32 0
; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %44 = shufflevector <128 x i1> %ins1, <128 x i1> poison, <128 x i32> zeroinitializer
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %ins2 = insertelement <2 x i8> poison, i8 3, i32 0
diff --git a/llvm/test/Analysis/CostModel/RISCV/vp-intrinsics.ll b/llvm/test/Analysis/CostModel/RISCV/vp-intrinsics.ll
index 9e8f727978001..1f80147958ead 100644
--- a/llvm/test/Analysis/CostModel/RISCV/vp-intrinsics.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/vp-intrinsics.ll
@@ -984,10 +984,10 @@ define void @gather() {
; ARGBASED-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %2 = call <4 x i8> @llvm.vp.gather.v4i8.v4p0(<4 x ptr> poison, <4 x i1> poison, i32 poison)
; ARGBASED-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %3 = call <8 x i8> @llvm.vp.gather.v8i8.v8p0(<8 x ptr> poison, <8 x i1> poison, i32 poison)
; ARGBASED-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %4 = call <16 x i8> @llvm.vp.gather.v16i8.v16p0(<16 x ptr> poison, <16 x i1> poison, i32 poison)
-; ARGBASED-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %5 = call <2 x i64> @llvm.vp.gather.v2i64.v2p0(<2 x ptr> poison, <2 x i1> poison, i32 poison)
-; ARGBASED-NEXT: Cost Model: Found an estimated cost of 33 for instruction: %6 = call <4 x i64> @llvm.vp.gather.v4i64.v4p0(<4 x ptr> poison, <4 x i1> poison, i32 poison)
-; ARGBASED-NEXT: Cost Model: Found an estimated cost of 69 for instruction: %7 = call <8 x i64> @llvm.vp.gather.v8i64.v8p0(<8 x ptr> poison, <8 x i1> poison, i32 poison)
-; ARGBASED-NEXT: Cost Model: Found an estimated cost of 141 for instruction: %8 = call <16 x i64> @llvm.vp.gather.v16i64.v16p0(<16 x ptr> poison, <16 x i1> poison, i32 poison)
+; ARGBASED-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %5 = call <2 x i64> @llvm.vp.gather.v2i64.v2p0(<2 x ptr> poison, <2 x i1> poison, i32 poison)
+; ARGBASED-NEXT: Cost Model: Found an estimated cost of 41 for instruction: %6 = call <4 x i64> @llvm.vp.gather.v4i64.v4p0(<4 x ptr> poison, <4 x i1> poison, i32 poison)
+; ARGBASED-NEXT: Cost Model: Found an estimated cost of 85 for instruction: %7 = call <8 x i64> @llvm.vp.gather.v8i64.v8p0(<8 x ptr> poison, <8 x i1> poison, i32 poison)
+; ARGBASED-NEXT: Cost Model: Found an estimated cost of 173 for instruction: %8 = call <16 x i64> @llvm.vp.gather.v16i64.v16p0(<16 x ptr> poison, <16 x i1> poison, i32 poison)
; ARGBASED-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %9 = call <vscale x 2 x i8> @llvm.vp.gather.nxv2i8.nxv2p0(<vscale x 2 x ptr> poison, <vscale x 2 x i1> poison, i32 poison)
; ARGBASED-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %10 = call <vscale x 4 x i8> @llvm.vp.gather.nxv4i8.nxv4p0(<vscale x 4 x ptr> poison, <vscale x 4 x i1> poison, i32 poison)
; ARGBASED-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %11 = call <vscale x 8 x i8> @llvm.vp.gather.nxv8i8.nxv8p0(<vscale x 8 x ptr> poison, <vscale x 8 x i1> poison, i32 poison)
@@ -999,14 +999,14 @@ define void @gather() {
; ARGBASED-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; TYPEBASED-LABEL: 'gather'
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %1 = call <2 x i8> @llvm.vp.gather.v2i8.v2p0(<2 x ptr> poison, <2 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %2 = call <4 x i8> @llvm.vp.gather.v4i8.v4p0(<4 x ptr> poison, <4 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 62 for instruction: %3 = call <8 x i8> @llvm.vp.gather.v8i8.v8p0(<8 x ptr> poison, <8 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 126 for instruction: %4 = call <16 x i8> @llvm.vp.gather.v16i8.v16p0(<16 x ptr> poison, <16 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %5 = call <2 x i64> @llvm.vp.gather.v2i64.v2p0(<2 x ptr> poison, <2 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 33 for instruction: %6 = call <4 x i64> @llvm.vp.gather.v4i64.v4p0(<4 x ptr> poison, <4 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 69 for instruction: %7 = call <8 x i64> @llvm.vp.gather.v8i64.v8p0(<8 x ptr> poison, <8 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 141 for instruction: %8 = call <16 x i64> @llvm.vp.gather.v16i64.v16p0(<16 x ptr> poison, <16 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %1 = call <2 x i8> @llvm.vp.gather.v2i8.v2p0(<2 x ptr> poison, <2 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 38 for instruction: %2 = call <4 x i8> @llvm.vp.gather.v4i8.v4p0(<4 x ptr> poison, <4 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 78 for instruction: %3 = call <8 x i8> @llvm.vp.gather.v8i8.v8p0(<8 x ptr> poison, <8 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 158 for instruction: %4 = call <16 x i8> @llvm.vp.gather.v16i8.v16p0(<16 x ptr> poison, <16 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %5 = call <2 x i64> @llvm.vp.gather.v2i64.v2p0(<2 x ptr> poison, <2 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 41 for instruction: %6 = call <4 x i64> @llvm.vp.gather.v4i64.v4p0(<4 x ptr> poison, <4 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 85 for instruction: %7 = call <8 x i64> @llvm.vp.gather.v8i64.v8p0(<8 x ptr> poison, <8 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 173 for instruction: %8 = call <16 x i64> @llvm.vp.gather.v16i64.v16p0(<16 x ptr> poison, <16 x i1> poison, i32 poison)
; TYPEBASED-NEXT: Cost Model: Invalid cost for instruction: %9 = call <vscale x 2 x i8> @llvm.vp.gather.nxv2i8.nxv2p0(<vscale x 2 x ptr> poison, <vscale x 2 x i1> poison, i32 poison)
; TYPEBASED-NEXT: Cost Model: Invalid cost for instruction: %10 = call <vscale x 4 x i8> @llvm.vp.gather.nxv4i8.nxv4p0(<vscale x 4 x ptr> poison, <vscale x 4 x i1> poison, i32 poison)
; TYPEBASED-NEXT: Cost Model: Invalid cost for instruction: %11 = call <vscale x 8 x i8> @llvm.vp.gather.nxv8i8.nxv8p0(<vscale x 8 x ptr> poison, <vscale x 8 x i1> poison, i32 poison)
@@ -1042,10 +1042,10 @@ define void @scatter() {
; ARGBASED-NEXT: Cost Model: Found an estimated cost of 4 for instruction: call void @llvm.vp.scatter.v4i8.v4p0(<4 x i8> poison, <4 x ptr> poison, <4 x i1> poison, i32 poison)
; ARGBASED-NEXT: Cost Model: Found an estimated cost of 8 for instruction: call void @llvm.vp.scatter.v8i8.v8p0(<8 x i8> poison, <8 x ptr> poison, <8 x i1> poison, i32 poison)
; ARGBASED-NEXT: Cost Model: Found an estimated cost of 16 for instruction: call void @llvm.vp.scatter.v16i8.v16p0(<16 x i8> poison, <16 x ptr> poison, <16 x i1> poison, i32 poison)
-; ARGBASED-NEXT: Cost Model: Found an estimated cost of 15 for instruction: call void @llvm.vp.scatter.v2i64.v2p0(<2 x i64> poison, <2 x ptr> poison, <2 x i1> poison, i32 poison)
-; ARGBASED-NEXT: Cost Model: Found an estimated cost of 33 for instruction: call void @llvm.vp.scatter.v4i64.v4p0(<4 x i64> poison, <4 x ptr> poison, <4 x i1> poison, i32 poison)
-; ARGBASED-NEXT: Cost Model: Found an estimated cost of 69 for instruction: call void @llvm.vp.scatter.v8i64.v8p0(<8 x i64> poison, <8 x ptr> poison, <8 x i1> poison, i32 poison)
-; ARGBASED-NEXT: Cost Model: Found an estimated cost of 141 for instruction: call void @llvm.vp.scatter.v16i64.v16p0(<16 x i64> poison, <16 x ptr> poison, <16 x i1> poison, i32 poison)
+; ARGBASED-NEXT: Cost Model: Found an estimated cost of 21 for instruction: call void @llvm.vp.scatter.v2i64.v2p0(<2 x i64> poison, <2 x ptr> poison, <2 x i1> poison, i32 poison)
+; ARGBASED-NEXT: Cost Model: Found an estimated cost of 45 for instruction: call void @llvm.vp.scatter.v4i64.v4p0(<4 x i64> poison, <4 x ptr> poison, <4 x i1> poison, i32 poison)
+; ARGBASED-NEXT: Cost Model: Found an estimated cost of 93 for instruction: call void @llvm.vp.scatter.v8i64.v8p0(<8 x i64> poison, <8 x ptr> poison, <8 x i1> poison, i32 poison)
+; ARGBASED-NEXT: Cost Model: Found an estimated cost of 189 for instruction: call void @llvm.vp.scatter.v16i64.v16p0(<16 x i64> poison, <16 x ptr> poison, <16 x i1> poison, i32 poison)
; ARGBASED-NEXT: Cost Model: Found an estimated cost of 4 for instruction: call void @llvm.vp.scatter.nxv2i8.nxv2p0(<vscale x 2 x i8> poison, <vscale x 2 x ptr> poison, <vscale x 2 x i1> poison, i32 poison)
; ARGBASED-NEXT: Cost Model: Found an estimated cost of 8 for instruction: call void @llvm.vp.scatter.nxv4i8.nxv4p0(<vscale x 4 x i8> poison, <vscale x 4 x ptr> poison, <vscale x 4 x i1> poison, i32 poison)
; ARGBASED-NEXT: Cost Model: Found an estimated cost of 16 for instruction: call void @llvm.vp.scatter.nxv8i8.nxv8p0(<vscale x 8 x i8> poison, <vscale x 8 x ptr> poison, <vscale x 8 x i1> poison, i32 poison)
@@ -1057,14 +1057,14 @@ define void @scatter() {
; ARGBASED-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; TYPEBASED-LABEL: 'scatter'
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 15 for instruction: call void @llvm.vp.scatter.v2i8.v2p0(<2 x i8> poison, <2 x ptr> poison, <2 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 33 for instruction: call void @llvm.vp.scatter.v4i8.v4p0(<4 x i8> poison, <4 x ptr> poison, <4 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 69 for instruction: call void @llvm.vp.scatter.v8i8.v8p0(<8 x i8> poison, <8 x ptr> poison, <8 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 141 for instruction: call void @llvm.vp.scatter.v16i8.v16p0(<16 x i8> poison, <16 x ptr> poison, <16 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 15 for instruction: call void @llvm.vp.scatter.v2i64.v2p0(<2 x i64> poison, <2 x ptr> poison, <2 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 33 for instruction: call void @llvm.vp.scatter.v4i64.v4p0(<4 x i64> poison, <4 x ptr> poison, <4 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 69 for instruction: call void @llvm.vp.scatter.v8i64.v8p0(<8 x i64> poison, <8 x ptr> poison, <8 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 141 for instruction: call void @llvm.vp.scatter.v16i64.v16p0(<16 x i64> poison, <16 x ptr> poison, <16 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 21 for instruction: call void @llvm.vp.scatter.v2i8.v2p0(<2 x i8> poison, <2 x ptr> poison, <2 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 45 for instruction: call void @llvm.vp.scatter.v4i8.v4p0(<4 x i8> poison, <4 x ptr> poison, <4 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 93 for instruction: call void @llvm.vp.scatter.v8i8.v8p0(<8 x i8> poison, <8 x ptr> poison, <8 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 189 for instruction: call void @llvm.vp.scatter.v16i8.v16p0(<16 x i8> poison, <16 x ptr> poison, <16 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 21 for instruction: call void @llvm.vp.scatter.v2i64.v2p0(<2 x i64> poison, <2 x ptr> poison, <2 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 45 for instruction: call void @llvm.vp.scatter.v4i64.v4p0(<4 x i64> poison, <4 x ptr> poison, <4 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 93 for instruction: call void @llvm.vp.scatter.v8i64.v8p0(<8 x i64> poison, <8 x ptr> poison, <8 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 189 for instruction: call void @llvm.vp.scatter.v16i64.v16p0(<16 x i64> poison, <16 x ptr> poison, <16 x i1> poison, i32 poison)
; TYPEBASED-NEXT: Cost Model: Invalid cost for instruction: call void @llvm.vp.scatter.nxv2i8.nxv2p0(<vscale x 2 x i8> poison, <vscale x 2 x ptr> poison, <vscale x 2 x i1> poison, i32 poison)
; TYPEBASED-NEXT: Cost Model: Invalid cost for instruction: call void @llvm.vp.scatter.nxv4i8.nxv4p0(<vscale x 4 x i8> poison, <vscale x 4 x ptr> poison, <vscale x 4 x i1> poison, i32 poison)
; TYPEBASED-NEXT: Cost Model: Invalid cost for instruction: call void @llvm.vp.scatter.nxv8i8.nxv8p0(<vscale x 8 x i8> poison, <vscale x 8 x ptr> poison, <vscale x 8 x i1> poison, i32 poison)
@@ -1096,10 +1096,10 @@ define void @scatter() {
define void @strided_load() {
; ARGBASED-LABEL: 'strided_load'
-; ARGBASED-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %ti1_2 = call <2 x i1> @llvm.experimental.vp.strided.load.v2i1.p0.i64(ptr undef, i64 undef, <2 x i1> undef, i32 undef)
-; ARGBASED-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %ti1_4 = call <4 x i1> @llvm.experimental.vp.strided.load.v4i1.p0.i64(ptr undef, i64 undef, <4 x i1> undef, i32 undef)
-; ARGBASED-NEXT: Cost Model: Found an estimated cost of 33 for instruction: %ti1_8 = call <8 x i1> @llvm.experimental.vp.strided.load.v8i1.p0.i64(ptr undef, i64 undef, <8 x i1> undef, i32 undef)
-; ARGBASED-NEXT: Cost Model: Found an estimated cost of 65 for instruction: %ti1_16 = call <16 x i1> @llvm.experimental.vp.strided.load.v16i1.p0.i64(ptr undef, i64 undef, <16 x i1> undef, i32 undef)
+; ARGBASED-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %ti1_2 = call <2 x i1> @llvm.experimental.vp.strided.load.v2i1.p0.i64(ptr undef, i64 undef, <2 x i1> undef, i32 undef)
+; ARGBASED-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %ti1_4 = call <4 x i1> @llvm.experimental.vp.strided.load.v4i1.p0.i64(ptr undef, i64 undef, <4 x i1> undef, i32 undef)
+; ARGBASED-NEXT: Cost Model: Found an estimated cost of 41 for instruction: %ti1_8 = call <8 x i1> @llvm.experimental.vp.strided.load.v8i1.p0.i64(ptr undef, i64 undef, <8 x i1> undef, i32 undef)
+; ARGBASED-NEXT: Cost Model: Found an estimated cost of 81 for instruction: %ti1_16 = call <16 x i1> @llvm.experimental.vp.strided.load.v16i1.p0.i64(ptr undef, i64 undef, <16 x i1> undef, i32 undef)
; ARGBASED-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t0 = call <2 x i8> @llvm.experimental.vp.strided.load.v2i8.p0.i64(ptr undef, i64 undef, <2 x i1> undef, i32 undef)
; ARGBASED-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t2 = call <4 x i8> @llvm.experimental.vp.strided.load.v4i8.p0.i64(ptr undef, i64 undef, <4 x i1> undef, i32 undef)
; ARGBASED-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t4 = call <8 x i8> @llvm.experimental.vp.strided.load.v8i8.p0.i64(ptr undef, i64 undef, <8 x i1> undef, i32 undef)
@@ -1123,10 +1123,10 @@ define void @strided_load() {
; ARGBASED-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; TYPEBASED-LABEL: 'strided_load'
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %ti1_2 = call <2 x i1> @llvm.experimental.vp.strided.load.v2i1.p0.i64(ptr undef, i64 undef, <2 x i1> undef, i32 undef)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 32 for instruction: %ti1_4 = call <4 x i1> @llvm.experimental.vp.strided.load.v4i1.p0.i64(ptr undef, i64 undef, <4 x i1> undef, i32 undef)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 64 for instruction: %ti1_8 = call <8 x i1> @llvm.experimental.vp.strided.load.v8i1.p0.i64(ptr undef, i64 undef, <8 x i1> undef, i32 undef)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 128 for instruction: %ti1_16 = call <16 x i1> @llvm.experimental.vp.strided.load.v16i1.p0.i64(ptr undef, i64 undef, <16 x i1> undef, i32 undef)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %ti1_2 = call <2 x i1> @llvm.experimental.vp.strided.load.v2i1.p0.i64(ptr undef, i64 undef, <2 x i1> undef, i32 undef)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 40 for instruction: %ti1_4 = call <4 x i1> @llvm.experimental.vp.strided.load.v4i1.p0.i64(ptr undef, i64 undef, <4 x i1> undef, i32 undef)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 80 for instruction: %ti1_8 = call <8 x i1> @llvm.experimental.vp.strided.load.v8i1.p0.i64(ptr undef, i64 undef, <8 x i1> undef, i32 undef)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 160 for instruction: %ti1_16 = call <16 x i1> @llvm.experimental.vp.strided.load.v16i1.p0.i64(ptr undef, i64 undef, <16 x i1> undef, i32 undef)
; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t0 = call <2 x i8> @llvm.experimental.vp.strided.load.v2i8.p0.i64(ptr undef, i64 undef, <2 x i1> undef, i32 undef)
; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t2 = call <4 x i8> @llvm.experimental.vp.strided.load.v4i8.p0.i64(ptr undef, i64 undef, <4 x i1> undef, i32 undef)
; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t4 = call <8 x i8> @llvm.experimental.vp.strided.load.v8i8.p0.i64(ptr undef, i64 undef, <8 x i1> undef, i32 undef)
@@ -1227,38 +1227,38 @@ define void @strided_store() {
define void @reduce_add() {
; CHECK-LABEL: 'reduce_add'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %1 = call i8 @llvm.vp.reduce.add.v2i8(i8 undef, <2 x i8> undef, <2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %2 = call i8 @llvm.vector.reduce.add.v2i8(<2 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %3 = call i8 @llvm.vp.reduce.add.v4i8(i8 undef, <4 x i8> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %4 = call i8 @llvm.vector.reduce.add.v4i8(<4 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %5 = call i8 @llvm.vp.reduce.add.v8i8(i8 undef, <8 x i8> undef, <8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %6 = call i8 @llvm.vector.reduce.add.v8i8(<8 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %7 = call i8 @llvm.vp.reduce.add.v16i8(i8 undef, <16 x i8> undef, <16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %8 = call i8 @llvm.vector.reduce.add.v16i8(<16 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %9 = call i64 @llvm.vp.reduce.add.v2i64(i64 undef, <2 x i64> undef, <2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %10 = call i64 @llvm.vector.reduce.add.v2i64(<2 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %11 = call i64 @llvm.vp.reduce.add.v4i64(i64 undef, <4 x i64> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %12 = call i64 @llvm.vector.reduce.add.v4i64(<4 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %13 = call i64 @llvm.vp.reduce.add.v8i64(i64 undef, <8 x i64> undef, <8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %14 = call i64 @llvm.vector.reduce.add.v8i64(<8 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %15 = call i64 @llvm.vp.reduce.add.v16i64(i64 undef, <16 x i64> undef, <16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %16 = call i64 @llvm.vector.reduce.add.v16i64(<16 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %17 = call i8 @llvm.vp.reduce.add.nxv8i8(i8 undef, <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: %18 = call i8 @llvm.vector.reduce.add.nxv2i8(<vscale x 2 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %19 = call i8 @llvm.vp.reduce.add.nxv4i8(i8 undef, <vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %20 = call i8 @llvm.vector.reduce.add.nxv4i8(<vscale x 4 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %21 = call i8 @llvm.vp.reduce.add.nxv8i8(i8 undef, <vscale x 8 x i8> undef, <vscale x 8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %22 = call i8 @llvm.vector.reduce.add.nxv8i8(<vscale x 8 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %23 = call i8 @llvm.vp.reduce.add.nxv16i8(i8 undef, <vscale x 16 x i8> undef, <vscale x 16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %24 = call i8 @llvm.vector.reduce.add.nxv16i8(<vscale x 16 x i8> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %25 = call i64 @llvm.vp.reduce.add.nxv2i64(i64 undef, <vscale x 2 x i64> undef, <vscale x 2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %26 = call i64 @llvm.vector.reduce.add.nxv2i64(<vscale x 2 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %27 = call i64 @llvm.vp.reduce.add.nxv4i64(i64 undef, <vscale x 4 x i64> undef, <vscale x 4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %28 = call i64 @llvm.vector.reduce.add.nxv4i64(<vscale x 4 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %29 = call i64 @llvm.vp.reduce.add.nxv8i64(i64 undef, <vscale x 8 x i64> undef, <vscale x 8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %30 = call i64 @llvm.vector.reduce.add.nxv8i64(<vscale x 8 x i64> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %31 = call i64 @llvm.vp.reduce.add.nxv16i64(i64 undef, <vscale x 16 x i64> undef, <vscale x 16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %32 = call i64 @llvm.vector.reduce.add.nxv16i64(<vscale x 16 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %1 = call i8 @llvm.vp.reduce.add.v2i8(i8 undef, <2 x i8> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %2 = call i8 @llvm.vector.reduce.add.v2i8(<2 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %3 = call i8 @llvm.vp.reduce.add.v4i8(i8 undef, <4 x i8> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %4 = call i8 @llvm.vector.reduce.add.v4i8(<4 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %5 = call i8 @llvm.vp.reduce.add.v8i8(i8 undef, <8 x i8> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %6 = call i8 @llvm.vector.reduce.add.v8i8(<8 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %7 = call i8 @llvm.vp.reduce.add.v16i8(i8 undef, <16 x i8> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %8 = call i8 @llvm.vector.reduce.add.v16i8(<16 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %9 = call i64 @llvm.vp.reduce.add.v2i64(i64 undef, <2 x i64> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %10 = call i64 @llvm.vector.reduce.add.v2i64(<2 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %11 = call i64 @llvm.vp.reduce.add.v4i64(i64 undef, <4 x i64> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %12 = call i64 @llvm.vector.reduce.add.v4i64(<4 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %13 = call i64 @llvm.vp.reduce.add.v8i64(i64 undef, <8 x i64> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %14 = call i64 @llvm.vector.reduce.add.v8i64(<8 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %15 = call i64 @llvm.vp.reduce.add.v16i64(i64 undef, <16 x i64> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %16 = call i64 @llvm.vector.reduce.add.v16i64(<16 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %17 = call i8 @llvm.vp.reduce.add.nxv8i8(i8 undef, <vscale x 8 x i8> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %18 = call i8 @llvm.vector.reduce.add.nxv2i8(<vscale x 2 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %19 = call i8 @llvm.vp.reduce.add.nxv4i8(i8 undef, <vscale x 4 x i8> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %20 = call i8 @llvm.vector.reduce.add.nxv4i8(<vscale x 4 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %21 = call i8 @llvm.vp.reduce.add.nxv8i8(i8 undef, <vscale x 8 x i8> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %22 = call i8 @llvm.vector.reduce.add.nxv8i8(<vscale x 8 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %23 = call i8 @llvm.vp.reduce.add.nxv16i8(i8 undef, <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: %24 = call i8 @llvm.vector.reduce.add.nxv16i8(<vscale x 16 x i8> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %25 = call i64 @llvm.vp.reduce.add.nxv2i64(i64 undef, <vscale x 2 x i64> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %26 = call i64 @llvm.vector.reduce.add.nxv2i64(<vscale x 2 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %27 = call i64 @llvm.vp.reduce.add.nxv4i64(i64 undef, <vscale x 4 x i64> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %28 = call i64 @llvm.vector.reduce.add.nxv4i64(<vscale x 4 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %29 = call i64 @llvm.vp.reduce.add.nxv8i64(i64 undef, <vscale x 8 x i64> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %30 = call i64 @llvm.vector.reduce.add.nxv8i64(<vscale x 8 x i64> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %31 = call i64 @llvm.vp.reduce.add.nxv16i64(i64 undef, <vscale x 16 x i64> undef, <vscale x 16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %32 = call i64 @llvm.vector.reduce.add.nxv16i64(<vscale x 16 x i64> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
call i8 @llvm.vp.reduce.add.v2i8(i8 undef, <2 x i8> undef, <2 x i1> undef, i32 undef)
@@ -1298,38 +1298,38 @@ define void @reduce_add() {
define void @reduce_fadd() {
; CHECK-LABEL: 'reduce_fadd'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %1 = call float @llvm.vp.reduce.fadd.v2f32(float undef, <2 x float> undef, <2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %2 = call float @llvm.vector.reduce.fadd.v2f32(float undef, <2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %3 = call float @llvm.vp.reduce.fadd.v4f32(float undef, <4 x float> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %4 = call float @llvm.vector.reduce.fadd.v4f32(float undef, <4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %5 = call float @llvm.vp.reduce.fadd.v8f32(float undef, <8 x float> undef, <8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %6 = call float @llvm.vector.reduce.fadd.v8f32(float undef, <8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %7 = call float @llvm.vp.reduce.fadd.v16f32(float undef, <16 x float> undef, <16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %8 = call float @llvm.vector.reduce.fadd.v16f32(float undef, <16 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %9 = call double @llvm.vp.reduce.fadd.v2f64(double undef, <2 x double> undef, <2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %10 = call double @llvm.vector.reduce.fadd.v2f64(double undef, <2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %11 = call double @llvm.vp.reduce.fadd.v4f64(double undef, <4 x double> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %12 = call double @llvm.vector.reduce.fadd.v4f64(double undef, <4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %13 = call double @llvm.vp.reduce.fadd.v8f64(double undef, <8 x double> undef, <8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %14 = call double @llvm.vector.reduce.fadd.v8f64(double undef, <8 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %15 = call double @llvm.vp.reduce.fadd.v16f64(double undef, <16 x double> undef, <16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %16 = call double @llvm.vector.reduce.fadd.v16f64(double undef, <16 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %17 = call float @llvm.vp.reduce.fadd.nxv2f32(float undef, <vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %18 = call float @llvm.vector.reduce.fadd.nxv2f32(float undef, <vscale x 2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %19 = call float @llvm.vp.reduce.fadd.nxv4f32(float undef, <vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %20 = call float @llvm.vector.reduce.fadd.nxv4f32(float undef, <vscale x 4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %21 = call float @llvm.vp.reduce.fadd.nxv8f32(float undef, <vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %22 = call float @llvm.vector.reduce.fadd.nxv8f32(float undef, <vscale x 8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 34 for instruction: %23 = call float @llvm.vp.reduce.fadd.nxv16f32(float undef, <vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 34 for instruction: %24 = call float @llvm.vector.reduce.fadd.nxv16f32(float undef, <vscale x 16 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %25 = call double @llvm.vp.reduce.fadd.nxv2f64(double undef, <vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %26 = call double @llvm.vector.reduce.fadd.nxv2f64(double undef, <vscale x 2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %27 = call double @llvm.vp.reduce.fadd.nxv4f64(double undef, <vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %28 = call double @llvm.vector.reduce.fadd.nxv4f64(double undef, <vscale x 4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %29 = call double @llvm.vp.reduce.fadd.nxv8f64(double undef, <vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %30 = call double @llvm.vector.reduce.fadd.nxv8f64(double undef, <vscale x 8 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 34 for instruction: %31 = call double @llvm.vp.reduce.fadd.nxv16f64(double undef, <vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 34 for instruction: %32 = call double @llvm.vector.reduce.fadd.nxv16f64(double undef, <vscale x 16 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %1 = call float @llvm.vp.reduce.fadd.v2f32(float undef, <2 x float> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %2 = call float @llvm.vector.reduce.fadd.v2f32(float undef, <2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %3 = call float @llvm.vp.reduce.fadd.v4f32(float undef, <4 x float> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %4 = call float @llvm.vector.reduce.fadd.v4f32(float undef, <4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %5 = call float @llvm.vp.reduce.fadd.v8f32(float undef, <8 x float> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %6 = call float @llvm.vector.reduce.fadd.v8f32(float undef, <8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %7 = call float @llvm.vp.reduce.fadd.v16f32(float undef, <16 x float> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %8 = call float @llvm.vector.reduce.fadd.v16f32(float undef, <16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %9 = call double @llvm.vp.reduce.fadd.v2f64(double undef, <2 x double> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %10 = call double @llvm.vector.reduce.fadd.v2f64(double undef, <2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %11 = call double @llvm.vp.reduce.fadd.v4f64(double undef, <4 x double> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %12 = call double @llvm.vector.reduce.fadd.v4f64(double undef, <4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %13 = call double @llvm.vp.reduce.fadd.v8f64(double undef, <8 x double> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %14 = call double @llvm.vector.reduce.fadd.v8f64(double undef, <8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %15 = call double @llvm.vp.reduce.fadd.v16f64(double undef, <16 x double> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %16 = call double @llvm.vector.reduce.fadd.v16f64(double undef, <16 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %17 = call float @llvm.vp.reduce.fadd.nxv2f32(float undef, <vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %18 = call float @llvm.vector.reduce.fadd.nxv2f32(float undef, <vscale x 2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %19 = call float @llvm.vp.reduce.fadd.nxv4f32(float undef, <vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %20 = call float @llvm.vector.reduce.fadd.nxv4f32(float undef, <vscale x 4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %21 = call float @llvm.vp.reduce.fadd.nxv8f32(float undef, <vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %22 = call float @llvm.vector.reduce.fadd.nxv8f32(float undef, <vscale x 8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 35 for instruction: %23 = call float @llvm.vp.reduce.fadd.nxv16f32(float undef, <vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 35 for instruction: %24 = call float @llvm.vector.reduce.fadd.nxv16f32(float undef, <vscale x 16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %25 = call double @llvm.vp.reduce.fadd.nxv2f64(double undef, <vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %26 = call double @llvm.vector.reduce.fadd.nxv2f64(double undef, <vscale x 2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %27 = call double @llvm.vp.reduce.fadd.nxv4f64(double undef, <vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %28 = call double @llvm.vector.reduce.fadd.nxv4f64(double undef, <vscale x 4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %29 = call double @llvm.vp.reduce.fadd.nxv8f64(double undef, <vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %30 = call double @llvm.vector.reduce.fadd.nxv8f64(double undef, <vscale x 8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 35 for instruction: %31 = call double @llvm.vp.reduce.fadd.nxv16f64(double undef, <vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 35 for instruction: %32 = call double @llvm.vector.reduce.fadd.nxv16f64(double undef, <vscale x 16 x double> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
call float @llvm.vp.reduce.fadd.v2f32(float undef, <2 x float> undef, <2 x i1> undef, i32 undef)
@@ -1369,20 +1369,20 @@ define void @reduce_fadd() {
define void @reduce_other() {
; CHECK-LABEL: 'reduce_other'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %1 = call i32 @llvm.vp.reduce.xor.v4i32(i32 undef, <4 x i32> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %2 = call i32 @llvm.vp.reduce.and.v4i32(i32 undef, <4 x i32> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %3 = call i32 @llvm.vp.reduce.or.v4i32(i32 undef, <4 x i32> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %4 = call i32 @llvm.vp.reduce.mul.v4i32(i32 undef, <4 x i32> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %5 = call i32 @llvm.vp.reduce.smin.v4i32(i32 undef, <4 x i32> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %6 = call i32 @llvm.vp.reduce.smax.v4i32(i32 undef, <4 x i32> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %7 = call i32 @llvm.vp.reduce.umin.v4i32(i32 undef, <4 x i32> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %8 = call i32 @llvm.vp.reduce.umax.v4i32(i32 undef, <4 x i32> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %1 = call i32 @llvm.vp.reduce.xor.v4i32(i32 undef, <4 x i32> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %2 = call i32 @llvm.vp.reduce.and.v4i32(i32 undef, <4 x i32> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %3 = call i32 @llvm.vp.reduce.or.v4i32(i32 undef, <4 x i32> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %4 = call i32 @llvm.vp.reduce.mul.v4i32(i32 undef, <4 x i32> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %5 = call i32 @llvm.vp.reduce.smin.v4i32(i32 undef, <4 x i32> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %6 = call i32 @llvm.vp.reduce.smax.v4i32(i32 undef, <4 x i32> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %7 = call i32 @llvm.vp.reduce.umin.v4i32(i32 undef, <4 x i32> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %8 = call i32 @llvm.vp.reduce.umax.v4i32(i32 undef, <4 x i32> undef, <4 x i1> undef, i32 undef)
; CHECK-NEXT: Cost Model: Invalid cost for instruction: %9 = call i32 @llvm.vp.reduce.mul.nxv4i32(i32 undef, <vscale x 4 x i32> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT: Cost Model: Invalid cost for instruction: %10 = call float @llvm.vp.reduce.fmul.nxv4f32(float undef, <vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %11 = call float @llvm.vp.reduce.fmin.nxv4f32(float undef, <vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %12 = call float @llvm.vp.reduce.fminimum.nxv4f32(float undef, <vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %13 = call float @llvm.vp.reduce.fmax.nxv4f32(float undef, <vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %14 = call float @llvm.vp.reduce.fmaximum.nxv4f32(float undef, <vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %11 = call float @llvm.vp.reduce.fmin.nxv4f32(float undef, <vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %12 = call float @llvm.vp.reduce.fminimum.nxv4f32(float undef, <vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %13 = call float @llvm.vp.reduce.fmax.nxv4f32(float undef, <vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %14 = call float @llvm.vp.reduce.fmaximum.nxv4f32(float undef, <vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
call i32 @llvm.vp.reduce.xor.v4i32(i32 undef, <4 x i32> undef, <4 x i1> undef, i32 undef)
@@ -1651,22 +1651,22 @@ define void @is.fpclass() {
; ARGBASED-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; TYPEBASED-LABEL: 'is.fpclass'
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %1 = call <2 x i1> @llvm.vp.is.fpclass.v2bf16(<2 x bfloat> poison, i32 0, <2 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %2 = call <4 x i1> @llvm.vp.is.fpclass.v4bf16(<4 x bfloat> poison, i32 0, <4 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 33 for instruction: %3 = call <8 x i1> @llvm.vp.is.fpclass.v8bf16(<8 x bfloat> poison, i32 0, <8 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 65 for instruction: %4 = call <16 x i1> @llvm.vp.is.fpclass.v16bf16(<16 x bfloat> poison, i32 0, <16 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %5 = call <2 x i1> @llvm.vp.is.fpclass.v2f16(<2 x half> poison, i32 0, <2 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %6 = call <4 x i1> @llvm.vp.is.fpclass.v4f16(<4 x half> poison, i32 0, <4 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 33 for instruction: %7 = call <8 x i1> @llvm.vp.is.fpclass.v8f16(<8 x half> poison, i32 0, <8 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 65 for instruction: %8 = call <16 x i1> @llvm.vp.is.fpclass.v16f16(<16 x half> poison, i32 0, <16 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %9 = call <2 x i1> @llvm.vp.is.fpclass.v2f32(<2 x float> poison, i32 0, <2 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %10 = call <4 x i1> @llvm.vp.is.fpclass.v4f32(<4 x float> poison, i32 0, <4 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 33 for instruction: %11 = call <8 x i1> @llvm.vp.is.fpclass.v8f32(<8 x float> poison, i32 0, <8 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 65 for instruction: %12 = call <16 x i1> @llvm.vp.is.fpclass.v16f32(<16 x float> poison, i32 0, <16 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %13 = call <2 x i1> @llvm.vp.is.fpclass.v2f64(<2 x double> poison, i32 0, <2 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %14 = call <4 x i1> @llvm.vp.is.fpclass.v4f64(<4 x double> poison, i32 0, <4 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 33 for instruction: %15 = call <8 x i1> @llvm.vp.is.fpclass.v8f64(<8 x double> poison, i32 0, <8 x i1> poison, i32 poison)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 65 for instruction: %16 = call <16 x i1> @llvm.vp.is.fpclass.v16f64(<16 x double> poison, i32 0, <16 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %1 = call <2 x i1> @llvm.vp.is.fpclass.v2bf16(<2 x bfloat> poison, i32 0, <2 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %2 = call <4 x i1> @llvm.vp.is.fpclass.v4bf16(<4 x bfloat> poison, i32 0, <4 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 41 for instruction: %3 = call <8 x i1> @llvm.vp.is.fpclass.v8bf16(<8 x bfloat> poison, i32 0, <8 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 81 for instruction: %4 = call <16 x i1> @llvm.vp.is.fpclass.v16bf16(<16 x bfloat> poison, i32 0, <16 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %5 = call <2 x i1> @llvm.vp.is.fpclass.v2f16(<2 x half> poison, i32 0, <2 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %6 = call <4 x i1> @llvm.vp.is.fpclass.v4f16(<4 x half> poison, i32 0, <4 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 41 for instruction: %7 = call <8 x i1> @llvm.vp.is.fpclass.v8f16(<8 x half> poison, i32 0, <8 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 81 for instruction: %8 = call <16 x i1> @llvm.vp.is.fpclass.v16f16(<16 x half> poison, i32 0, <16 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %9 = call <2 x i1> @llvm.vp.is.fpclass.v2f32(<2 x float> poison, i32 0, <2 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %10 = call <4 x i1> @llvm.vp.is.fpclass.v4f32(<4 x float> poison, i32 0, <4 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 41 for instruction: %11 = call <8 x i1> @llvm.vp.is.fpclass.v8f32(<8 x float> poison, i32 0, <8 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 81 for instruction: %12 = call <16 x i1> @llvm.vp.is.fpclass.v16f32(<16 x float> poison, i32 0, <16 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %13 = call <2 x i1> @llvm.vp.is.fpclass.v2f64(<2 x double> poison, i32 0, <2 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %14 = call <4 x i1> @llvm.vp.is.fpclass.v4f64(<4 x double> poison, i32 0, <4 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 41 for instruction: %15 = call <8 x i1> @llvm.vp.is.fpclass.v8f64(<8 x double> poison, i32 0, <8 x i1> poison, i32 poison)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 81 for instruction: %16 = call <16 x i1> @llvm.vp.is.fpclass.v16f64(<16 x double> poison, i32 0, <16 x i1> poison, i32 poison)
; TYPEBASED-NEXT: Cost Model: Invalid cost for instruction: %17 = call <vscale x 2 x i1> @llvm.vp.is.fpclass.nxv2bf16(<vscale x 2 x bfloat> poison, i32 0, <vscale x 2 x i1> poison, i32 poison)
; TYPEBASED-NEXT: Cost Model: Invalid cost for instruction: %18 = call <vscale x 4 x i1> @llvm.vp.is.fpclass.nxv4bf16(<vscale x 4 x bfloat> poison, i32 0, <vscale x 4 x i1> poison, i32 poison)
; TYPEBASED-NEXT: Cost Model: Invalid cost for instruction: %19 = call <vscale x 8 x i1> @llvm.vp.is.fpclass.nxv8bf16(<vscale x 8 x bfloat> poison, i32 0, <vscale x 8 x i1> poison, i32 poison)
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/conditional-scalar-assignment.ll b/llvm/test/Transforms/LoopVectorize/RISCV/conditional-scalar-assignment.ll
index 87828ae5a7480..cc7be1f4fde78 100644
--- a/llvm/test/Transforms/LoopVectorize/RISCV/conditional-scalar-assignment.ll
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/conditional-scalar-assignment.ll
@@ -87,7 +87,7 @@ define i32 @non_speculatable_find_last_reduction(ptr noalias %a, ptr noalias %b,
; CHECK-NEXT: [[SCALAR_PH:.*]]:
; CHECK-NEXT: [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; CHECK-NEXT: [[TMP1:%.*]] = shl nuw i64 [[TMP0]], 2
-; CHECK-NEXT: [[UMAX:%.*]] = call i64 @llvm.umax.i64(i64 [[TMP1]], i64 7)
+; CHECK-NEXT: [[UMAX:%.*]] = call i64 @llvm.umax.i64(i64 [[TMP1]], i64 8)
; CHECK-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[N]], [[UMAX]]
; CHECK-NEXT: br i1 [[MIN_ITERS_CHECK]], label %[[SCALAR_PH1:.*]], label %[[EXIT:.*]]
; CHECK: [[EXIT]]:
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-call-intrinsics.ll b/llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-call-intrinsics.ll
index ba1edbdd35fb3..ceeb32528fd22 100644
--- a/llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-call-intrinsics.ll
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-call-intrinsics.ll
@@ -1200,38 +1200,10 @@ define void @log10(ptr %a, ptr %b, i64 %N) {
;
; NO-VP-LABEL: define void @log10(
; NO-VP-SAME: ptr [[A:%.*]], ptr [[B:%.*]], i64 [[N:%.*]]) #[[ATTR0]] {
-; NO-VP-NEXT: [[ENTRY:.*]]:
-; NO-VP-NEXT: [[B2:%.*]] = ptrtoaddr ptr [[B]] to i64
-; NO-VP-NEXT: [[A1:%.*]] = ptrtoaddr ptr [[A]] to i64
-; NO-VP-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[N]], 16
-; NO-VP-NEXT: br i1 [[MIN_ITERS_CHECK]], label %[[SCALAR_PH:.*]], label %[[VECTOR_MEMCHECK:.*]]
-; NO-VP: [[VECTOR_MEMCHECK]]:
-; NO-VP-NEXT: [[TMP1:%.*]] = sub i64 [[A1]], [[B2]]
-; NO-VP-NEXT: [[TMP5:%.*]] = sub i64 [[TMP1]], 1
-; NO-VP-NEXT: [[DIFF_CHECK:%.*]] = icmp ult i64 [[TMP5]], 15
-; NO-VP-NEXT: br i1 [[DIFF_CHECK]], label %[[SCALAR_PH]], label %[[VECTOR_PH:.*]]
-; NO-VP: [[VECTOR_PH]]:
-; NO-VP-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[N]], 4
-; NO-VP-NEXT: [[N_VEC:%.*]] = sub i64 [[N]], [[N_MOD_VF]]
-; NO-VP-NEXT: br label %[[VECTOR_BODY:.*]]
-; NO-VP: [[VECTOR_BODY]]:
-; NO-VP-NEXT: [[IV:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
-; NO-VP-NEXT: [[GEP:%.*]] = getelementptr inbounds float, ptr [[B]], i64 [[IV]]
-; NO-VP-NEXT: [[WIDE_LOAD:%.*]] = load <4 x float>, ptr [[GEP]], align 4
-; NO-VP-NEXT: [[TMP2:%.*]] = call <4 x float> @llvm.log10.v4f32(<4 x float> [[WIDE_LOAD]])
-; NO-VP-NEXT: [[TMP3:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[IV]]
-; NO-VP-NEXT: store <4 x float> [[TMP2]], ptr [[TMP3]], align 4
-; NO-VP-NEXT: [[INDEX_NEXT]] = add nuw i64 [[IV]], 4
-; NO-VP-NEXT: [[TMP4:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; NO-VP-NEXT: br i1 [[TMP4]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP20:![0-9]+]]
-; NO-VP: [[MIDDLE_BLOCK]]:
-; NO-VP-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[N]], [[N_VEC]]
-; NO-VP-NEXT: br i1 [[CMP_N]], label %[[EXIT:.*]], label %[[SCALAR_PH]]
-; NO-VP: [[SCALAR_PH]]:
-; NO-VP-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], %[[MIDDLE_BLOCK]] ], [ 0, %[[ENTRY]] ], [ 0, %[[VECTOR_MEMCHECK]] ]
+; NO-VP-NEXT: [[SCALAR_PH:.*]]:
; NO-VP-NEXT: br label %[[LOOP:.*]]
; NO-VP: [[LOOP]]:
-; NO-VP-NEXT: [[IV1:%.*]] = phi i64 [ [[IV_NEXT:%.*]], %[[LOOP]] ], [ [[BC_RESUME_VAL]], %[[SCALAR_PH]] ]
+; NO-VP-NEXT: [[IV1:%.*]] = phi i64 [ [[IV_NEXT:%.*]], %[[LOOP]] ], [ 0, %[[SCALAR_PH]] ]
; NO-VP-NEXT: [[GEP1:%.*]] = getelementptr inbounds float, ptr [[B]], i64 [[IV1]]
; NO-VP-NEXT: [[TMP0:%.*]] = load float, ptr [[GEP1]], align 4
; NO-VP-NEXT: [[COND:%.*]] = tail call float @llvm.log10.f32(float [[TMP0]])
@@ -1239,7 +1211,7 @@ define void @log10(ptr %a, ptr %b, i64 %N) {
; NO-VP-NEXT: store float [[COND]], ptr [[GEP9]], align 4
; NO-VP-NEXT: [[IV_NEXT]] = add nuw nsw i64 [[IV1]], 1
; NO-VP-NEXT: [[EXITCOND_NOT:%.*]] = icmp eq i64 [[IV_NEXT]], [[N]]
-; NO-VP-NEXT: br i1 [[EXITCOND_NOT]], label %[[EXIT]], label %[[LOOP]], !llvm.loop [[LOOP21:![0-9]+]]
+; NO-VP-NEXT: br i1 [[EXITCOND_NOT]], label %[[EXIT:.*]], label %[[LOOP]]
; NO-VP: [[EXIT]]:
; NO-VP-NEXT: ret void
;
@@ -1305,6 +1277,4 @@ exit:
; NO-VP: [[LOOP17]] = distinct !{[[LOOP17]], [[META1]]}
; NO-VP: [[LOOP18]] = distinct !{[[LOOP18]], [[META1]], [[META2]]}
; NO-VP: [[LOOP19]] = distinct !{[[LOOP19]], [[META1]]}
-; NO-VP: [[LOOP20]] = distinct !{[[LOOP20]], [[META1]], [[META2]]}
-; NO-VP: [[LOOP21]] = distinct !{[[LOOP21]], [[META1]]}
;.
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/uniform-load-store.ll b/llvm/test/Transforms/LoopVectorize/RISCV/uniform-load-store.ll
index 256528c25a781..c7584291766a8 100644
--- a/llvm/test/Transforms/LoopVectorize/RISCV/uniform-load-store.ll
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/uniform-load-store.ll
@@ -581,18 +581,23 @@ define void @uniform_store_of_loop_varying(ptr noalias nocapture %a, ptr noalias
; FIXEDLEN-NEXT: [[ENTRY:.*:]]
; FIXEDLEN-NEXT: br label %[[VECTOR_PH:.*]]
; FIXEDLEN: [[VECTOR_PH]]:
-; FIXEDLEN-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x i64> poison, i64 [[V]], i64 0
-; FIXEDLEN-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x i64> [[BROADCAST_SPLATINSERT]], <4 x i64> poison, <4 x i32> zeroinitializer
+; FIXEDLEN-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <2 x ptr> poison, ptr [[B]], i64 0
+; FIXEDLEN-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <2 x ptr> [[BROADCAST_SPLATINSERT]], <2 x ptr> poison, <2 x i32> zeroinitializer
+; FIXEDLEN-NEXT: [[BROADCAST_SPLATINSERT1:%.*]] = insertelement <2 x i64> poison, i64 [[V]], i64 0
+; FIXEDLEN-NEXT: [[BROADCAST_SPLAT2:%.*]] = shufflevector <2 x i64> [[BROADCAST_SPLATINSERT1]], <2 x i64> poison, <2 x i32> zeroinitializer
; FIXEDLEN-NEXT: br label %[[VECTOR_BODY:.*]]
; FIXEDLEN: [[VECTOR_BODY]]:
; FIXEDLEN-NEXT: [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
-; FIXEDLEN-NEXT: [[TMP4:%.*]] = add i64 [[INDEX]], 7
-; FIXEDLEN-NEXT: store i64 [[TMP4]], ptr [[B]], align 8
+; FIXEDLEN-NEXT: [[VEC_IND:%.*]] = phi <2 x i64> [ <i64 0, i64 1>, %[[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], %[[VECTOR_BODY]] ]
+; FIXEDLEN-NEXT: [[STEP_ADD:%.*]] = add nuw <2 x i64> [[VEC_IND]], splat (i64 2)
+; FIXEDLEN-NEXT: call void @llvm.masked.scatter.v2i64.v2p0(<2 x i64> [[VEC_IND]], <2 x ptr> align 8 [[BROADCAST_SPLAT]], <2 x i1> splat (i1 true))
+; FIXEDLEN-NEXT: call void @llvm.masked.scatter.v2i64.v2p0(<2 x i64> [[STEP_ADD]], <2 x ptr> align 8 [[BROADCAST_SPLAT]], <2 x i1> splat (i1 true))
; FIXEDLEN-NEXT: [[TMP5:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[INDEX]]
-; FIXEDLEN-NEXT: [[TMP7:%.*]] = getelementptr inbounds i64, ptr [[TMP5]], i64 4
-; FIXEDLEN-NEXT: store <4 x i64> [[BROADCAST_SPLAT]], ptr [[TMP5]], align 8
-; FIXEDLEN-NEXT: store <4 x i64> [[BROADCAST_SPLAT]], ptr [[TMP7]], align 8
-; FIXEDLEN-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 8
+; FIXEDLEN-NEXT: [[TMP1:%.*]] = getelementptr inbounds i64, ptr [[TMP5]], i64 2
+; FIXEDLEN-NEXT: store <2 x i64> [[BROADCAST_SPLAT2]], ptr [[TMP5]], align 8
+; FIXEDLEN-NEXT: store <2 x i64> [[BROADCAST_SPLAT2]], ptr [[TMP1]], align 8
+; FIXEDLEN-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
+; FIXEDLEN-NEXT: [[VEC_IND_NEXT]] = add nuw nsw <2 x i64> [[STEP_ADD]], splat (i64 2)
; FIXEDLEN-NEXT: [[TMP8:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1024
; FIXEDLEN-NEXT: br i1 [[TMP8]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP12:![0-9]+]]
; FIXEDLEN: [[MIDDLE_BLOCK]]:
diff --git a/llvm/test/Transforms/SLPVectorizer/RISCV/complex-loads.ll b/llvm/test/Transforms/SLPVectorizer/RISCV/complex-loads.ll
index a431560023ba1..e3e6ce59c8a22 100644
--- a/llvm/test/Transforms/SLPVectorizer/RISCV/complex-loads.ll
+++ b/llvm/test/Transforms/SLPVectorizer/RISCV/complex-loads.ll
@@ -287,8 +287,62 @@ define i32 @test(ptr %pix1, ptr %pix2, i64 %idx.ext, i64 %idx.ext63, ptr %add.pt
; UNALIGNED_VEC_MEM-NEXT: [[ARRAYIDX5_2:%.*]] = getelementptr i8, ptr [[ADD_PTR64_1]], i64 4
; UNALIGNED_VEC_MEM-NEXT: [[ARRAYIDX5_3:%.*]] = getelementptr i8, ptr null, i64 4
; UNALIGNED_VEC_MEM-NEXT: [[ARRAYIDX5_4:%.*]] = getelementptr i8, ptr null, i64 4
+; UNALIGNED_VEC_MEM-NEXT: [[ARRAYIDX8_3:%.*]] = getelementptr i8, ptr null, i64 1
+; UNALIGNED_VEC_MEM-NEXT: [[ARRAYIDX10_3:%.*]] = getelementptr i8, ptr null, i64 1
; UNALIGNED_VEC_MEM-NEXT: [[TMP57:%.*]] = load i8, ptr null, align 1
+; UNALIGNED_VEC_MEM-NEXT: [[ARRAYIDX15_3:%.*]] = getelementptr i8, ptr null, i64 5
+; UNALIGNED_VEC_MEM-NEXT: [[ARRAYIDX20_3:%.*]] = getelementptr i8, ptr null, i64 2
+; UNALIGNED_VEC_MEM-NEXT: [[ARRAYIDX22_3:%.*]] = getelementptr i8, ptr null, i64 2
; UNALIGNED_VEC_MEM-NEXT: [[TMP61:%.*]] = load i8, ptr null, align 1
+; UNALIGNED_VEC_MEM-NEXT: [[ARRAYIDX27_3:%.*]] = getelementptr i8, ptr null, i64 6
+; UNALIGNED_VEC_MEM-NEXT: [[ARRAYIDX32_3:%.*]] = getelementptr i8, ptr null, i64 3
+; UNALIGNED_VEC_MEM-NEXT: [[ARRAYIDX34_3:%.*]] = getelementptr i8, ptr null, i64 3
+; UNALIGNED_VEC_MEM-NEXT: [[ARRAYIDX39_3:%.*]] = getelementptr i8, ptr null, i64 7
+; UNALIGNED_VEC_MEM-NEXT: [[TMP18:%.*]] = load i8, ptr [[ARRAYIDX32_3]], align 1
+; UNALIGNED_VEC_MEM-NEXT: [[TMP19:%.*]] = load i8, ptr [[ARRAYIDX20_3]], align 1
+; UNALIGNED_VEC_MEM-NEXT: [[TMP20:%.*]] = load i8, ptr [[ARRAYIDX8_3]], align 1
+; UNALIGNED_VEC_MEM-NEXT: [[TMP24:%.*]] = load i8, ptr null, align 1
+; UNALIGNED_VEC_MEM-NEXT: [[CONV_3:%.*]] = zext i8 [[TMP24]] to i32
+; UNALIGNED_VEC_MEM-NEXT: [[CONV9_3:%.*]] = zext i8 [[TMP20]] to i32
+; UNALIGNED_VEC_MEM-NEXT: [[CONV21_3:%.*]] = zext i8 [[TMP19]] to i32
+; UNALIGNED_VEC_MEM-NEXT: [[CONV33_3:%.*]] = zext i8 [[TMP18]] to i32
+; UNALIGNED_VEC_MEM-NEXT: [[TMP25:%.*]] = load i8, ptr [[ARRAYIDX34_3]], align 1
+; UNALIGNED_VEC_MEM-NEXT: [[TMP26:%.*]] = load i8, ptr [[ARRAYIDX22_3]], align 1
+; UNALIGNED_VEC_MEM-NEXT: [[TMP27:%.*]] = load i8, ptr [[ARRAYIDX10_3]], align 1
+; UNALIGNED_VEC_MEM-NEXT: [[TMP28:%.*]] = load i8, ptr null, align 1
+; UNALIGNED_VEC_MEM-NEXT: [[CONV2_3:%.*]] = zext i8 [[TMP28]] to i32
+; UNALIGNED_VEC_MEM-NEXT: [[CONV11_3:%.*]] = zext i8 [[TMP27]] to i32
+; UNALIGNED_VEC_MEM-NEXT: [[CONV23_3:%.*]] = zext i8 [[TMP26]] to i32
+; UNALIGNED_VEC_MEM-NEXT: [[CONV35_3:%.*]] = zext i8 [[TMP25]] to i32
+; UNALIGNED_VEC_MEM-NEXT: [[SUB_3:%.*]] = sub i32 [[CONV_3]], [[CONV2_3]]
+; UNALIGNED_VEC_MEM-NEXT: [[SUB12_3:%.*]] = sub i32 [[CONV9_3]], [[CONV11_3]]
+; UNALIGNED_VEC_MEM-NEXT: [[SUB24_3:%.*]] = sub i32 [[CONV21_3]], [[CONV23_3]]
+; UNALIGNED_VEC_MEM-NEXT: [[SUB36_3:%.*]] = sub i32 [[CONV33_3]], [[CONV35_3]]
+; UNALIGNED_VEC_MEM-NEXT: [[TMP29:%.*]] = load i8, ptr [[ARRAYIDX5_3]], align 1
+; UNALIGNED_VEC_MEM-NEXT: [[TMP30:%.*]] = load i8, ptr null, align 1
+; UNALIGNED_VEC_MEM-NEXT: [[CONV4_3:%.*]] = zext i8 [[TMP29]] to i32
+; UNALIGNED_VEC_MEM-NEXT: [[CONV14_3:%.*]] = zext i8 [[TMP57]] to i32
+; UNALIGNED_VEC_MEM-NEXT: [[CONV26_3:%.*]] = zext i8 [[TMP61]] to i32
+; UNALIGNED_VEC_MEM-NEXT: [[CONV38_3:%.*]] = zext i8 [[TMP30]] to i32
+; UNALIGNED_VEC_MEM-NEXT: [[TMP31:%.*]] = load i8, ptr [[ARRAYIDX39_3]], align 1
+; UNALIGNED_VEC_MEM-NEXT: [[TMP32:%.*]] = load i8, ptr [[ARRAYIDX27_3]], align 1
+; UNALIGNED_VEC_MEM-NEXT: [[TMP33:%.*]] = load i8, ptr [[ARRAYIDX15_3]], align 1
+; UNALIGNED_VEC_MEM-NEXT: [[TMP34:%.*]] = load i8, ptr [[ARRAYIDX5_4]], align 1
+; UNALIGNED_VEC_MEM-NEXT: [[CONV6_3:%.*]] = zext i8 [[TMP34]] to i32
+; UNALIGNED_VEC_MEM-NEXT: [[CONV16_3:%.*]] = zext i8 [[TMP33]] to i32
+; UNALIGNED_VEC_MEM-NEXT: [[CONV28_3:%.*]] = zext i8 [[TMP32]] to i32
+; UNALIGNED_VEC_MEM-NEXT: [[CONV40_3:%.*]] = zext i8 [[TMP31]] to i32
+; UNALIGNED_VEC_MEM-NEXT: [[SUB7_3:%.*]] = sub i32 [[CONV4_3]], [[CONV6_3]]
+; UNALIGNED_VEC_MEM-NEXT: [[SUB17_3:%.*]] = sub i32 [[CONV14_3]], [[CONV16_3]]
+; UNALIGNED_VEC_MEM-NEXT: [[SUB29_3:%.*]] = sub i32 [[CONV26_3]], [[CONV28_3]]
+; UNALIGNED_VEC_MEM-NEXT: [[SUB41_3:%.*]] = sub i32 [[CONV38_3]], [[CONV40_3]]
+; UNALIGNED_VEC_MEM-NEXT: [[SHL_3:%.*]] = shl i32 [[SUB7_3]], 16
+; UNALIGNED_VEC_MEM-NEXT: [[SHL18_3:%.*]] = shl i32 [[SUB17_3]], 16
+; UNALIGNED_VEC_MEM-NEXT: [[SHL30_3:%.*]] = shl i32 [[SUB29_3]], 16
+; UNALIGNED_VEC_MEM-NEXT: [[SHL42_3:%.*]] = shl i32 [[SUB41_3]], 16
+; UNALIGNED_VEC_MEM-NEXT: [[ADD_3:%.*]] = add i32 [[SHL_3]], [[SUB_3]]
+; UNALIGNED_VEC_MEM-NEXT: [[ADD19_3:%.*]] = add i32 [[SHL18_3]], [[SUB12_3]]
+; UNALIGNED_VEC_MEM-NEXT: [[ADD31_3:%.*]] = add i32 [[SHL30_3]], [[SUB24_3]]
; UNALIGNED_VEC_MEM-NEXT: [[TMP4:%.*]] = load <4 x i8>, ptr [[ARRAYIDX3]], align 1
; UNALIGNED_VEC_MEM-NEXT: [[TMP2:%.*]] = load <4 x i8>, ptr [[PIX1]], align 1
; UNALIGNED_VEC_MEM-NEXT: [[TMP3:%.*]] = load <4 x i8>, ptr [[PIX2]], align 1
@@ -305,35 +359,15 @@ define i32 @test(ptr %pix1, ptr %pix2, i64 %idx.ext, i64 %idx.ext63, ptr %add.pt
; UNALIGNED_VEC_MEM-NEXT: [[TMP15:%.*]] = load <4 x i8>, ptr [[ADD_PTR644]], align 1
; UNALIGNED_VEC_MEM-NEXT: [[TMP16:%.*]] = load <4 x i8>, ptr [[ARRAYIDX3_1]], align 1
; UNALIGNED_VEC_MEM-NEXT: [[TMP17:%.*]] = load <4 x i8>, ptr [[ARRAYIDX5_1]], align 1
-; UNALIGNED_VEC_MEM-NEXT: [[TMP18:%.*]] = zext <4 x i8> [[TMP14]] to <4 x i32>
-; UNALIGNED_VEC_MEM-NEXT: [[TMP19:%.*]] = zext <4 x i8> [[TMP15]] to <4 x i32>
-; UNALIGNED_VEC_MEM-NEXT: [[TMP20:%.*]] = sub <4 x i32> [[TMP18]], [[TMP19]]
-; UNALIGNED_VEC_MEM-NEXT: [[TMP21:%.*]] = zext <4 x i8> [[TMP16]] to <4 x i32>
-; UNALIGNED_VEC_MEM-NEXT: [[TMP22:%.*]] = zext <4 x i8> [[TMP17]] to <4 x i32>
+; UNALIGNED_VEC_MEM-NEXT: [[TMP21:%.*]] = zext <4 x i8> [[TMP14]] to <4 x i32>
+; UNALIGNED_VEC_MEM-NEXT: [[TMP22:%.*]] = zext <4 x i8> [[TMP15]] to <4 x i32>
; UNALIGNED_VEC_MEM-NEXT: [[TMP23:%.*]] = sub <4 x i32> [[TMP21]], [[TMP22]]
-; UNALIGNED_VEC_MEM-NEXT: [[TMP24:%.*]] = shl <4 x i32> [[TMP23]], splat (i32 16)
-; UNALIGNED_VEC_MEM-NEXT: [[TMP25:%.*]] = add <4 x i32> [[TMP24]], [[TMP20]]
-; UNALIGNED_VEC_MEM-NEXT: [[TMP26:%.*]] = load <4 x i8>, ptr null, align 1
-; UNALIGNED_VEC_MEM-NEXT: [[TMP27:%.*]] = zext <4 x i8> [[TMP26]] to <4 x i32>
-; UNALIGNED_VEC_MEM-NEXT: [[TMP28:%.*]] = load <4 x i8>, ptr null, align 1
-; UNALIGNED_VEC_MEM-NEXT: [[TMP29:%.*]] = zext <4 x i8> [[TMP28]] to <4 x i32>
-; UNALIGNED_VEC_MEM-NEXT: [[TMP30:%.*]] = sub <4 x i32> [[TMP27]], [[TMP29]]
-; UNALIGNED_VEC_MEM-NEXT: [[TMP106:%.*]] = load i8, ptr null, align 1
-; UNALIGNED_VEC_MEM-NEXT: [[TMP42:%.*]] = load i8, ptr [[ARRAYIDX5_3]], align 1
-; UNALIGNED_VEC_MEM-NEXT: [[CONV38_3:%.*]] = zext i8 [[TMP61]] to i32
-; UNALIGNED_VEC_MEM-NEXT: [[CONV40_2:%.*]] = zext i8 [[TMP42]] to i32
-; UNALIGNED_VEC_MEM-NEXT: [[TMP33:%.*]] = insertelement <2 x i8> poison, i8 [[TMP106]], i64 0
-; UNALIGNED_VEC_MEM-NEXT: [[TMP34:%.*]] = insertelement <2 x i8> [[TMP33]], i8 [[TMP57]], i64 1
-; UNALIGNED_VEC_MEM-NEXT: [[TMP35:%.*]] = load <4 x i8>, ptr [[ARRAYIDX5_4]], align 1
-; UNALIGNED_VEC_MEM-NEXT: [[TMP36:%.*]] = zext <4 x i8> [[TMP35]] to <4 x i32>
-; UNALIGNED_VEC_MEM-NEXT: [[TMP37:%.*]] = insertelement <4 x i32> poison, i32 [[CONV40_2]], i64 0
-; UNALIGNED_VEC_MEM-NEXT: [[TMP38:%.*]] = zext <2 x i8> [[TMP34]] to <2 x i32>
-; UNALIGNED_VEC_MEM-NEXT: [[TMP39:%.*]] = shufflevector <2 x i32> [[TMP38]], <2 x i32> poison, <4 x i32> <i32 0, i32 1, i32 poison, i32 poison>
-; UNALIGNED_VEC_MEM-NEXT: [[TMP40:%.*]] = shufflevector <4 x i32> [[TMP37]], <4 x i32> [[TMP39]], <4 x i32> <i32 0, i32 4, i32 5, i32 poison>
-; UNALIGNED_VEC_MEM-NEXT: [[TMP41:%.*]] = insertelement <4 x i32> [[TMP40]], i32 [[CONV38_3]], i64 3
+; UNALIGNED_VEC_MEM-NEXT: [[TMP41:%.*]] = zext <4 x i8> [[TMP16]] to <4 x i32>
+; UNALIGNED_VEC_MEM-NEXT: [[TMP36:%.*]] = zext <4 x i8> [[TMP17]] to <4 x i32>
; UNALIGNED_VEC_MEM-NEXT: [[TMP58:%.*]] = sub <4 x i32> [[TMP41]], [[TMP36]]
; UNALIGNED_VEC_MEM-NEXT: [[TMP43:%.*]] = shl <4 x i32> [[TMP58]], splat (i32 16)
-; UNALIGNED_VEC_MEM-NEXT: [[TMP44:%.*]] = add <4 x i32> [[TMP43]], [[TMP30]]
+; UNALIGNED_VEC_MEM-NEXT: [[TMP44:%.*]] = add <4 x i32> [[TMP43]], [[TMP23]]
+; UNALIGNED_VEC_MEM-NEXT: [[ADD43_3:%.*]] = add i32 [[SHL42_3]], [[SUB36_3]]
; UNALIGNED_VEC_MEM-NEXT: [[TMP45:%.*]] = load <4 x i8>, ptr [[ADD_PTR_1]], align 1
; UNALIGNED_VEC_MEM-NEXT: [[TMP46:%.*]] = load <4 x i8>, ptr [[ADD_PTR64_1]], align 1
; UNALIGNED_VEC_MEM-NEXT: [[TMP47:%.*]] = load <4 x i8>, ptr [[ARRAYIDX3_2]], align 1
@@ -349,50 +383,46 @@ define i32 @test(ptr %pix1, ptr %pix2, i64 %idx.ext, i64 %idx.ext63, ptr %add.pt
; UNALIGNED_VEC_MEM-NEXT: [[TMP75:%.*]] = extractelement <4 x i32> [[TMP13]], i64 0
; UNALIGNED_VEC_MEM-NEXT: [[TMP84:%.*]] = extractelement <4 x i32> [[TMP13]], i64 1
; UNALIGNED_VEC_MEM-NEXT: [[SUB45:%.*]] = sub i32 [[TMP75]], [[TMP84]]
-; UNALIGNED_VEC_MEM-NEXT: [[TMP59:%.*]] = extractelement <4 x i32> [[TMP25]], i64 0
-; UNALIGNED_VEC_MEM-NEXT: [[TMP60:%.*]] = extractelement <4 x i32> [[TMP25]], i64 1
-; UNALIGNED_VEC_MEM-NEXT: [[SUB45_1:%.*]] = sub i32 [[TMP59]], [[TMP60]]
; UNALIGNED_VEC_MEM-NEXT: [[TMP85:%.*]] = extractelement <4 x i32> [[TMP44]], i64 0
; UNALIGNED_VEC_MEM-NEXT: [[TMP62:%.*]] = extractelement <4 x i32> [[TMP44]], i64 1
; UNALIGNED_VEC_MEM-NEXT: [[SUB45_3:%.*]] = sub i32 [[TMP85]], [[TMP62]]
+; UNALIGNED_VEC_MEM-NEXT: [[SUB45_4:%.*]] = sub i32 [[ADD_3]], [[ADD19_3]]
; UNALIGNED_VEC_MEM-NEXT: [[TMP63:%.*]] = extractelement <4 x i32> [[TMP56]], i64 0
; UNALIGNED_VEC_MEM-NEXT: [[TMP64:%.*]] = extractelement <4 x i32> [[TMP56]], i64 1
; UNALIGNED_VEC_MEM-NEXT: [[SUB45_2:%.*]] = sub i32 [[TMP63]], [[TMP64]]
; UNALIGNED_VEC_MEM-NEXT: [[ADD44:%.*]] = add i32 [[TMP84]], [[TMP75]]
-; UNALIGNED_VEC_MEM-NEXT: [[ADD44_1:%.*]] = add i32 [[TMP60]], [[TMP59]]
; UNALIGNED_VEC_MEM-NEXT: [[ADD44_3:%.*]] = add i32 [[TMP62]], [[TMP85]]
+; UNALIGNED_VEC_MEM-NEXT: [[ADD44_4:%.*]] = add i32 [[ADD19_3]], [[ADD_3]]
; UNALIGNED_VEC_MEM-NEXT: [[ADD44_2:%.*]] = add i32 [[TMP64]], [[TMP63]]
; UNALIGNED_VEC_MEM-NEXT: [[TMP65:%.*]] = extractelement <4 x i32> [[TMP13]], i64 2
; UNALIGNED_VEC_MEM-NEXT: [[TMP86:%.*]] = extractelement <4 x i32> [[TMP13]], i64 3
; UNALIGNED_VEC_MEM-NEXT: [[SUB47:%.*]] = sub i32 [[TMP65]], [[TMP86]]
-; UNALIGNED_VEC_MEM-NEXT: [[TMP87:%.*]] = extractelement <4 x i32> [[TMP25]], i64 2
-; UNALIGNED_VEC_MEM-NEXT: [[TMP68:%.*]] = extractelement <4 x i32> [[TMP25]], i64 3
-; UNALIGNED_VEC_MEM-NEXT: [[SUB47_1:%.*]] = sub i32 [[TMP87]], [[TMP68]]
; UNALIGNED_VEC_MEM-NEXT: [[TMP88:%.*]] = extractelement <4 x i32> [[TMP44]], i64 2
; UNALIGNED_VEC_MEM-NEXT: [[TMP70:%.*]] = extractelement <4 x i32> [[TMP44]], i64 3
-; UNALIGNED_VEC_MEM-NEXT: [[SUB47_3:%.*]] = sub i32 [[TMP88]], [[TMP70]]
+; UNALIGNED_VEC_MEM-NEXT: [[SUB47_1:%.*]] = sub i32 [[TMP88]], [[TMP70]]
+; UNALIGNED_VEC_MEM-NEXT: [[SUB47_3:%.*]] = sub i32 [[ADD31_3]], [[ADD43_3]]
; UNALIGNED_VEC_MEM-NEXT: [[TMP89:%.*]] = extractelement <4 x i32> [[TMP56]], i64 2
; UNALIGNED_VEC_MEM-NEXT: [[TMP72:%.*]] = extractelement <4 x i32> [[TMP56]], i64 3
; UNALIGNED_VEC_MEM-NEXT: [[SUB47_2:%.*]] = sub i32 [[TMP89]], [[TMP72]]
; UNALIGNED_VEC_MEM-NEXT: [[ADD46:%.*]] = add i32 [[TMP86]], [[TMP65]]
-; UNALIGNED_VEC_MEM-NEXT: [[ADD46_1:%.*]] = add i32 [[TMP68]], [[TMP87]]
-; UNALIGNED_VEC_MEM-NEXT: [[ADD46_3:%.*]] = add i32 [[TMP70]], [[TMP88]]
+; UNALIGNED_VEC_MEM-NEXT: [[ADD46_1:%.*]] = add i32 [[TMP70]], [[TMP88]]
+; UNALIGNED_VEC_MEM-NEXT: [[ADD46_3:%.*]] = add i32 [[ADD43_3]], [[ADD31_3]]
; UNALIGNED_VEC_MEM-NEXT: [[ADD46_2:%.*]] = add i32 [[TMP72]], [[TMP89]]
; UNALIGNED_VEC_MEM-NEXT: [[SUB59:%.*]] = sub i32 [[SUB45]], [[SUB47]]
-; UNALIGNED_VEC_MEM-NEXT: [[SUB59_1:%.*]] = sub i32 [[SUB45_1]], [[SUB47_1]]
-; UNALIGNED_VEC_MEM-NEXT: [[SUB59_3:%.*]] = sub i32 [[SUB45_3]], [[SUB47_3]]
+; UNALIGNED_VEC_MEM-NEXT: [[SUB59_1:%.*]] = sub i32 [[SUB45_3]], [[SUB47_1]]
+; UNALIGNED_VEC_MEM-NEXT: [[SUB59_3:%.*]] = sub i32 [[SUB45_4]], [[SUB47_3]]
; UNALIGNED_VEC_MEM-NEXT: [[SUB59_2:%.*]] = sub i32 [[SUB45_2]], [[SUB47_2]]
; UNALIGNED_VEC_MEM-NEXT: [[SUB51:%.*]] = sub i32 [[ADD44]], [[ADD46]]
-; UNALIGNED_VEC_MEM-NEXT: [[SUB51_1:%.*]] = sub i32 [[ADD44_1]], [[ADD46_1]]
-; UNALIGNED_VEC_MEM-NEXT: [[SUB51_3:%.*]] = sub i32 [[ADD44_3]], [[ADD46_3]]
+; UNALIGNED_VEC_MEM-NEXT: [[SUB51_1:%.*]] = sub i32 [[ADD44_3]], [[ADD46_1]]
+; UNALIGNED_VEC_MEM-NEXT: [[SUB51_3:%.*]] = sub i32 [[ADD44_4]], [[ADD46_3]]
; UNALIGNED_VEC_MEM-NEXT: [[SUB51_2:%.*]] = sub i32 [[ADD44_2]], [[ADD46_2]]
; UNALIGNED_VEC_MEM-NEXT: [[ADD55:%.*]] = add i32 [[SUB47]], [[SUB45]]
-; UNALIGNED_VEC_MEM-NEXT: [[ADD55_1:%.*]] = add i32 [[SUB47_1]], [[SUB45_1]]
-; UNALIGNED_VEC_MEM-NEXT: [[ADD55_3:%.*]] = add i32 [[SUB47_3]], [[SUB45_3]]
+; UNALIGNED_VEC_MEM-NEXT: [[ADD55_1:%.*]] = add i32 [[SUB47_1]], [[SUB45_3]]
+; UNALIGNED_VEC_MEM-NEXT: [[ADD55_3:%.*]] = add i32 [[SUB47_3]], [[SUB45_4]]
; UNALIGNED_VEC_MEM-NEXT: [[ADD55_2:%.*]] = add i32 [[SUB47_2]], [[SUB45_2]]
; UNALIGNED_VEC_MEM-NEXT: [[ADD48:%.*]] = add i32 [[ADD46]], [[ADD44]]
-; UNALIGNED_VEC_MEM-NEXT: [[ADD48_1:%.*]] = add i32 [[ADD46_1]], [[ADD44_1]]
-; UNALIGNED_VEC_MEM-NEXT: [[ADD48_3:%.*]] = add i32 [[ADD46_3]], [[ADD44_3]]
+; UNALIGNED_VEC_MEM-NEXT: [[ADD48_1:%.*]] = add i32 [[ADD46_1]], [[ADD44_3]]
+; UNALIGNED_VEC_MEM-NEXT: [[ADD48_3:%.*]] = add i32 [[ADD46_3]], [[ADD44_4]]
; UNALIGNED_VEC_MEM-NEXT: [[ADD48_2:%.*]] = add i32 [[ADD46_2]], [[ADD44_2]]
; UNALIGNED_VEC_MEM-NEXT: [[SUB86_3:%.*]] = sub i32 [[SUB59]], [[SUB59_1]]
; UNALIGNED_VEC_MEM-NEXT: [[ADD78_3:%.*]] = add i32 [[SUB59_1]], [[SUB59]]
@@ -427,15 +457,14 @@ define i32 @test(ptr %pix1, ptr %pix2, i64 %idx.ext, i64 %idx.ext63, ptr %add.pt
; UNALIGNED_VEC_MEM-NEXT: [[ADD103:%.*]] = add i32 [[ADD94]], [[ADD78]]
; UNALIGNED_VEC_MEM-NEXT: [[ADD105:%.*]] = add i32 [[SUB102]], [[SUB86]]
; UNALIGNED_VEC_MEM-NEXT: [[TMP66:%.*]] = insertelement <16 x i32> poison, i32 [[ADD46_2]], i64 0
-; UNALIGNED_VEC_MEM-NEXT: [[TMP74:%.*]] = shufflevector <4 x i32> [[TMP27]], <4 x i32> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison>
-; UNALIGNED_VEC_MEM-NEXT: [[TMP67:%.*]] = shufflevector <16 x i32> [[TMP66]], <16 x i32> [[TMP74]], <16 x i32> <i32 0, i32 16, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison>
+; UNALIGNED_VEC_MEM-NEXT: [[TMP67:%.*]] = insertelement <16 x i32> [[TMP66]], i32 [[CONV_3]], i64 1
; UNALIGNED_VEC_MEM-NEXT: [[TMP105:%.*]] = insertelement <16 x i32> [[TMP67]], i32 [[ADD46_1]], i64 2
; UNALIGNED_VEC_MEM-NEXT: [[TMP69:%.*]] = insertelement <16 x i32> [[TMP105]], i32 [[ADD46]], i64 3
; UNALIGNED_VEC_MEM-NEXT: [[TMP78:%.*]] = shufflevector <4 x i32> [[TMP49]], <4 x i32> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison>
; UNALIGNED_VEC_MEM-NEXT: [[TMP71:%.*]] = shufflevector <16 x i32> [[TMP69]], <16 x i32> [[TMP78]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 16, i32 17, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison>
; UNALIGNED_VEC_MEM-NEXT: [[TMP107:%.*]] = insertelement <16 x i32> [[TMP71]], i32 [[SUB47_1]], i64 6
; UNALIGNED_VEC_MEM-NEXT: [[TMP73:%.*]] = insertelement <16 x i32> [[TMP107]], i32 [[SUB47]], i64 7
-; UNALIGNED_VEC_MEM-NEXT: [[TMP118:%.*]] = shufflevector <4 x i32> [[TMP18]], <4 x i32> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison>
+; UNALIGNED_VEC_MEM-NEXT: [[TMP118:%.*]] = shufflevector <4 x i32> [[TMP21]], <4 x i32> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison>
; UNALIGNED_VEC_MEM-NEXT: [[TMP76:%.*]] = shufflevector <16 x i32> [[TMP73]], <16 x i32> [[TMP118]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 16, i32 17, i32 18, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison>
; UNALIGNED_VEC_MEM-NEXT: [[TMP77:%.*]] = insertelement <16 x i32> [[TMP76]], i32 [[ADD44]], i64 11
; UNALIGNED_VEC_MEM-NEXT: [[TMP79:%.*]] = shufflevector <4 x i32> [[TMP6]], <4 x i32> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison>
diff --git a/llvm/test/Transforms/SLPVectorizer/RISCV/external.ll b/llvm/test/Transforms/SLPVectorizer/RISCV/external.ll
index be5c24384c367..1863de244204d 100644
--- a/llvm/test/Transforms/SLPVectorizer/RISCV/external.ll
+++ b/llvm/test/Transforms/SLPVectorizer/RISCV/external.ll
@@ -5,21 +5,16 @@ define void @external(ptr %dest, ptr %dest2, ptr %src) {
; DEFAULT-LABEL: define void @external(
; DEFAULT-SAME: ptr [[DEST:%.*]], ptr [[DEST2:%.*]], ptr [[SRC:%.*]]) #[[ATTR0:[0-9]+]] {
; DEFAULT-NEXT: [[ENTRY:.*:]]
+; DEFAULT-NEXT: [[INC7:%.*]] = getelementptr inbounds float, ptr [[SRC]], i64 7
+; DEFAULT-NEXT: [[TMP3:%.*]] = load float, ptr [[INC7]], align 2
; DEFAULT-NEXT: [[TMP0:%.*]] = load <8 x float>, ptr [[SRC]], align 4
-; DEFAULT-NEXT: [[L0:%.*]] = load float, ptr [[SRC]], align 4
-; DEFAULT-NEXT: store <8 x float> [[TMP0]], ptr [[DEST]], align 4
-; DEFAULT-NEXT: [[D1:%.*]] = getelementptr inbounds float, ptr [[DEST2]], i64 1
-; DEFAULT-NEXT: [[D3:%.*]] = getelementptr inbounds float, ptr [[DEST2]], i64 3
; DEFAULT-NEXT: [[D4:%.*]] = getelementptr inbounds float, ptr [[DEST2]], i64 4
; DEFAULT-NEXT: [[D5:%.*]] = getelementptr inbounds float, ptr [[DEST2]], i64 5
; DEFAULT-NEXT: [[D6:%.*]] = getelementptr inbounds float, ptr [[DEST2]], i64 6
; DEFAULT-NEXT: [[D7:%.*]] = getelementptr inbounds float, ptr [[DEST2]], i64 7
-; DEFAULT-NEXT: store float [[L0]], ptr [[DEST2]], align 4
-; DEFAULT-NEXT: [[TMP1:%.*]] = shufflevector <8 x float> [[TMP0]], <8 x float> poison, <2 x i32> <i32 1, i32 2>
-; DEFAULT-NEXT: store <2 x float> [[TMP1]], ptr [[D1]], align 2
-; DEFAULT-NEXT: [[TMP2:%.*]] = extractelement <8 x float> [[TMP0]], i64 3
-; DEFAULT-NEXT: store float [[TMP2]], ptr [[D3]], align 2
-; DEFAULT-NEXT: [[TMP3:%.*]] = extractelement <8 x float> [[TMP0]], i64 7
+; DEFAULT-NEXT: [[TMP1:%.*]] = load <4 x float>, ptr [[SRC]], align 4
+; DEFAULT-NEXT: store <8 x float> [[TMP0]], ptr [[DEST]], align 4
+; DEFAULT-NEXT: store <4 x float> [[TMP1]], ptr [[DEST2]], align 4
; DEFAULT-NEXT: store float [[TMP3]], ptr [[D7]], align 2
; DEFAULT-NEXT: ret void
;
@@ -79,22 +74,19 @@ define void @external_schedule(ptr %dest, ptr %dest2, ptr %dest3, ptr %src, ptr
; DEFAULT-LABEL: define void @external_schedule(
; DEFAULT-SAME: ptr [[DEST:%.*]], ptr [[DEST2:%.*]], ptr [[DEST3:%.*]], ptr [[SRC:%.*]], ptr [[SRC1:%.*]]) #[[ATTR0]] {
; DEFAULT-NEXT: [[ENTRY:.*:]]
+; DEFAULT-NEXT: [[INC7:%.*]] = getelementptr inbounds float, ptr [[SRC]], i64 7
; DEFAULT-NEXT: [[D4:%.*]] = getelementptr inbounds float, ptr [[DEST2]], i64 4
; DEFAULT-NEXT: [[D5:%.*]] = getelementptr inbounds float, ptr [[DEST2]], i64 5
; DEFAULT-NEXT: [[D6:%.*]] = getelementptr inbounds float, ptr [[DEST2]], i64 6
; DEFAULT-NEXT: [[D7:%.*]] = getelementptr inbounds float, ptr [[DEST2]], i64 7
; DEFAULT-NEXT: [[TMP0:%.*]] = load <4 x float>, ptr [[SRC1]], align 4
; DEFAULT-NEXT: store i8 0, ptr [[DEST3]], align 4
+; DEFAULT-NEXT: [[TMP7:%.*]] = load float, ptr [[INC7]], align 2
; DEFAULT-NEXT: [[TMP1:%.*]] = load <8 x float>, ptr [[SRC]], align 4
-; DEFAULT-NEXT: [[L0:%.*]] = load float, ptr [[SRC]], align 4
+; DEFAULT-NEXT: [[TMP5:%.*]] = load <4 x float>, ptr [[SRC]], align 4
; DEFAULT-NEXT: store <8 x float> [[TMP1]], ptr [[DEST]], align 4
-; DEFAULT-NEXT: [[TMP2:%.*]] = shufflevector <8 x float> [[TMP1]], <8 x float> poison, <4 x i32> <i32 poison, i32 1, i32 2, i32 3>
-; DEFAULT-NEXT: [[TMP3:%.*]] = insertelement <4 x float> poison, float [[L0]], i64 0
-; DEFAULT-NEXT: [[TMP4:%.*]] = shufflevector <4 x float> [[TMP3]], <4 x float> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 poison, i32 poison, i32 poison, i32 poison>
-; DEFAULT-NEXT: [[TMP5:%.*]] = shufflevector <8 x float> [[TMP1]], <8 x float> [[TMP4]], <4 x i32> <i32 8, i32 1, i32 2, i32 3>
; DEFAULT-NEXT: [[TMP6:%.*]] = fadd <4 x float> [[TMP5]], [[TMP0]]
; DEFAULT-NEXT: store <4 x float> [[TMP6]], ptr [[DEST2]], align 4
-; DEFAULT-NEXT: [[TMP7:%.*]] = extractelement <8 x float> [[TMP1]], i64 7
; DEFAULT-NEXT: store float [[TMP7]], ptr [[D7]], align 2
; DEFAULT-NEXT: ret void
;
diff --git a/llvm/test/Transforms/SLPVectorizer/RISCV/extract-cost.ll b/llvm/test/Transforms/SLPVectorizer/RISCV/extract-cost.ll
index afba2b7089441..d4aba97f65e10 100644
--- a/llvm/test/Transforms/SLPVectorizer/RISCV/extract-cost.ll
+++ b/llvm/test/Transforms/SLPVectorizer/RISCV/extract-cost.ll
@@ -6,11 +6,13 @@
define double @f(ptr %p, double %c) {
; CHECK-LABEL: define double @f(
; CHECK-SAME: ptr [[P:%.*]], double [[C:%.*]]) #[[ATTR0:[0-9]+]] {
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x double>, ptr [[P]], align 8
-; CHECK-NEXT: [[TMP2:%.*]] = fmul <2 x double> splat (double 2.000000e+00), [[TMP1]]
-; CHECK-NEXT: [[TMP3:%.*]] = fadd <2 x double> splat (double 3.000000e+00), [[TMP2]]
-; CHECK-NEXT: [[TMP4:%.*]] = extractelement <2 x double> [[TMP3]], i64 0
-; CHECK-NEXT: [[TMP5:%.*]] = extractelement <2 x double> [[TMP3]], i64 1
+; CHECK-NEXT: [[X0:%.*]] = load double, ptr [[P]], align 8
+; CHECK-NEXT: [[Q:%.*]] = getelementptr double, ptr [[P]], i64 1
+; CHECK-NEXT: [[X1:%.*]] = load double, ptr [[Q]], align 8
+; CHECK-NEXT: [[M0:%.*]] = fmul double 2.000000e+00, [[X1]]
+; CHECK-NEXT: [[TMP5:%.*]] = fadd double 3.000000e+00, [[M0]]
+; CHECK-NEXT: [[M1:%.*]] = fmul double 2.000000e+00, [[X0]]
+; CHECK-NEXT: [[TMP4:%.*]] = fadd double 3.000000e+00, [[M1]]
; CHECK-NEXT: [[RES:%.*]] = call double @llvm.fmuladd.f64(double [[C]], double [[TMP5]], double [[TMP4]])
; CHECK-NEXT: ret double [[RES]]
;
diff --git a/llvm/test/Transforms/SLPVectorizer/RISCV/horizontal-list.ll b/llvm/test/Transforms/SLPVectorizer/RISCV/horizontal-list.ll
index 457f2600b539f..405637945af7c 100644
--- a/llvm/test/Transforms/SLPVectorizer/RISCV/horizontal-list.ll
+++ b/llvm/test/Transforms/SLPVectorizer/RISCV/horizontal-list.ll
@@ -8,7 +8,7 @@
; YAML-NEXT: Function: test
; YAML-NEXT: Args:
; YAML-NEXT: - String: 'Vectorized horizontal reduction with cost '
-; YAML-NEXT: - Cost: '-35'
+; YAML-NEXT: - Cost: '-34'
; YAML-NEXT: - String: ' and with tree size '
; YAML-NEXT: - TreeSize: '1'
; YAML-NEXT: ...
diff --git a/llvm/test/Transforms/SLPVectorizer/RISCV/load-store.ll b/llvm/test/Transforms/SLPVectorizer/RISCV/load-store.ll
index b27c20c62f8a5..dc022108dfc1e 100644
--- a/llvm/test/Transforms/SLPVectorizer/RISCV/load-store.ll
+++ b/llvm/test/Transforms/SLPVectorizer/RISCV/load-store.ll
@@ -269,20 +269,28 @@ entry:
define void @shared-chain-ordering(ptr %dest, ptr %p, i64 %offset) {
; CHECK-LABEL: @shared-chain-ordering(
; CHECK-NEXT: entry:
+; CHECK-NEXT: [[INC:%.*]] = getelementptr inbounds i16, ptr [[P:%.*]], i64 1
; CHECK-NEXT: [[INCS0:%.*]] = getelementptr inbounds i16, ptr [[DEST:%.*]], i64 [[OFFSET:%.*]]
-; CHECK-NEXT: [[TMP0:%.*]] = load <4 x i16>, ptr [[P:%.*]], align 4
+; CHECK-NEXT: [[TMP0:%.*]] = load <4 x i16>, ptr [[P]], align 4
+; CHECK-NEXT: [[E1:%.*]] = load i16, ptr [[INC]], align 2
; CHECK-NEXT: store <4 x i16> [[TMP0]], ptr [[INCS0]], align 4
-; CHECK-NEXT: [[TMP1:%.*]] = shufflevector <4 x i16> [[TMP0]], <4 x i16> poison, <2 x i32> <i32 0, i32 1>
-; CHECK-NEXT: store <2 x i16> [[TMP1]], ptr [[DEST]], align 4
+; CHECK-NEXT: [[TMP1:%.*]] = extractelement <4 x i16> [[TMP0]], i64 0
+; CHECK-NEXT: store i16 [[TMP1]], ptr [[DEST]], align 4
+; CHECK-NEXT: [[INCS:%.*]] = getelementptr inbounds i16, ptr [[DEST]], i64 1
+; CHECK-NEXT: store i16 [[E1]], ptr [[INCS]], align 2
; CHECK-NEXT: ret void
;
; DEFAULT-LABEL: @shared-chain-ordering(
; DEFAULT-NEXT: entry:
+; DEFAULT-NEXT: [[INC:%.*]] = getelementptr inbounds i16, ptr [[P:%.*]], i64 1
; DEFAULT-NEXT: [[INCS0:%.*]] = getelementptr inbounds i16, ptr [[DEST:%.*]], i64 [[OFFSET:%.*]]
-; DEFAULT-NEXT: [[TMP0:%.*]] = load <4 x i16>, ptr [[P:%.*]], align 4
+; DEFAULT-NEXT: [[TMP0:%.*]] = load <4 x i16>, ptr [[P]], align 4
+; DEFAULT-NEXT: [[E1:%.*]] = load i16, ptr [[INC]], align 2
; DEFAULT-NEXT: store <4 x i16> [[TMP0]], ptr [[INCS0]], align 4
-; DEFAULT-NEXT: [[TMP1:%.*]] = shufflevector <4 x i16> [[TMP0]], <4 x i16> poison, <2 x i32> <i32 0, i32 1>
-; DEFAULT-NEXT: store <2 x i16> [[TMP1]], ptr [[DEST]], align 4
+; DEFAULT-NEXT: [[TMP1:%.*]] = extractelement <4 x i16> [[TMP0]], i64 0
+; DEFAULT-NEXT: store i16 [[TMP1]], ptr [[DEST]], align 4
+; DEFAULT-NEXT: [[INCS:%.*]] = getelementptr inbounds i16, ptr [[DEST]], i64 1
+; DEFAULT-NEXT: store i16 [[E1]], ptr [[INCS]], align 2
; DEFAULT-NEXT: ret void
;
entry:
diff --git a/llvm/test/Transforms/SLPVectorizer/RISCV/reduced-value-repeated-and-vectorized.ll b/llvm/test/Transforms/SLPVectorizer/RISCV/reduced-value-repeated-and-vectorized.ll
index dcf93273f2054..5b9610f14c18a 100644
--- a/llvm/test/Transforms/SLPVectorizer/RISCV/reduced-value-repeated-and-vectorized.ll
+++ b/llvm/test/Transforms/SLPVectorizer/RISCV/reduced-value-repeated-and-vectorized.ll
@@ -5,8 +5,9 @@ define void @test() {
; CHECK-LABEL: define void @test(
; CHECK-SAME: ) #[[ATTR0:[0-9]+]] {
; CHECK-NEXT: [[ENTRY:.*:]]
-; CHECK-NEXT: [[TMP1:%.*]] = call <3 x i16> @llvm.experimental.vp.strided.load.v3i16.p0.i64(ptr align 2 null, i64 6, <3 x i1> splat (i1 true), i32 3)
-; CHECK-NEXT: [[TMP0:%.*]] = shufflevector <3 x i16> [[TMP1]], <3 x i16> poison, <4 x i32> <i32 0, i32 0, i32 1, i32 2>
+; CHECK-NEXT: [[TMP4:%.*]] = call <7 x i16> @llvm.masked.load.v7i16.p0(ptr align 2 null, <7 x i1> <i1 true, i1 false, i1 false, i1 true, i1 false, i1 false, i1 true>, <7 x i16> poison)
+; CHECK-NEXT: [[TMP1:%.*]] = shufflevector <7 x i16> [[TMP4]], <7 x i16> poison, <3 x i32> <i32 0, i32 3, i32 6>
+; CHECK-NEXT: [[TMP0:%.*]] = shufflevector <7 x i16> [[TMP4]], <7 x i16> poison, <4 x i32> <i32 0, i32 0, i32 3, i32 6>
; CHECK-NEXT: [[TMP2:%.*]] = xor <4 x i16> [[TMP0]], zeroinitializer
; CHECK-NEXT: [[TMP8:%.*]] = load i16, ptr getelementptr (i8, ptr null, i64 18), align 2
; CHECK-NEXT: [[TMP9:%.*]] = xor i16 [[TMP8]], 0
diff --git a/llvm/test/Transforms/SLPVectorizer/RISCV/reductions.ll b/llvm/test/Transforms/SLPVectorizer/RISCV/reductions.ll
index 2cbb99284a088..6c1c3141cf2cc 100644
--- a/llvm/test/Transforms/SLPVectorizer/RISCV/reductions.ll
+++ b/llvm/test/Transforms/SLPVectorizer/RISCV/reductions.ll
@@ -35,11 +35,45 @@ entry:
}
define i64 @red_ld_4xi64(ptr %ptr) {
-; CHECK-LABEL: @red_ld_4xi64(
-; CHECK-NEXT: entry:
-; CHECK-NEXT: [[TMP0:%.*]] = load <4 x i64>, ptr [[PTR:%.*]], align 8
-; CHECK-NEXT: [[TMP1:%.*]] = call i64 @llvm.vector.reduce.add.v4i64(<4 x i64> [[TMP0]])
-; CHECK-NEXT: ret i64 [[TMP1]]
+; ZVFHMIN-LABEL: @red_ld_4xi64(
+; ZVFHMIN-NEXT: entry:
+; ZVFHMIN-NEXT: [[LD0:%.*]] = load i64, ptr [[PTR:%.*]], align 8
+; ZVFHMIN-NEXT: [[GEP:%.*]] = getelementptr inbounds i64, ptr [[PTR]], i64 1
+; ZVFHMIN-NEXT: [[LD1:%.*]] = load i64, ptr [[GEP]], align 8
+; ZVFHMIN-NEXT: [[ADD_1:%.*]] = add nuw nsw i64 [[LD0]], [[LD1]]
+; ZVFHMIN-NEXT: [[GEP_1:%.*]] = getelementptr inbounds i64, ptr [[PTR]], i64 2
+; ZVFHMIN-NEXT: [[LD2:%.*]] = load i64, ptr [[GEP_1]], align 8
+; ZVFHMIN-NEXT: [[ADD_2:%.*]] = add nuw nsw i64 [[ADD_1]], [[LD2]]
+; ZVFHMIN-NEXT: [[GEP_2:%.*]] = getelementptr inbounds i64, ptr [[PTR]], i64 3
+; ZVFHMIN-NEXT: [[LD3:%.*]] = load i64, ptr [[GEP_2]], align 8
+; ZVFHMIN-NEXT: [[ADD_3:%.*]] = add nuw nsw i64 [[ADD_2]], [[LD3]]
+; ZVFHMIN-NEXT: ret i64 [[ADD_3]]
+;
+; ZVFHDEFAULT-LABEL: @red_ld_4xi64(
+; ZVFHDEFAULT-NEXT: entry:
+; ZVFHDEFAULT-NEXT: [[LD0:%.*]] = load i64, ptr [[PTR:%.*]], align 8
+; ZVFHDEFAULT-NEXT: [[GEP:%.*]] = getelementptr inbounds i64, ptr [[PTR]], i64 1
+; ZVFHDEFAULT-NEXT: [[LD1:%.*]] = load i64, ptr [[GEP]], align 8
+; ZVFHDEFAULT-NEXT: [[ADD_1:%.*]] = add nuw nsw i64 [[LD0]], [[LD1]]
+; ZVFHDEFAULT-NEXT: [[GEP_1:%.*]] = getelementptr inbounds i64, ptr [[PTR]], i64 2
+; ZVFHDEFAULT-NEXT: [[LD2:%.*]] = load i64, ptr [[GEP_1]], align 8
+; ZVFHDEFAULT-NEXT: [[ADD_2:%.*]] = add nuw nsw i64 [[ADD_1]], [[LD2]]
+; ZVFHDEFAULT-NEXT: [[GEP_2:%.*]] = getelementptr inbounds i64, ptr [[PTR]], i64 3
+; ZVFHDEFAULT-NEXT: [[LD3:%.*]] = load i64, ptr [[GEP_2]], align 8
+; ZVFHDEFAULT-NEXT: [[ADD_3:%.*]] = add nuw nsw i64 [[ADD_2]], [[LD3]]
+; ZVFHDEFAULT-NEXT: ret i64 [[ADD_3]]
+;
+; ZVFH256-LABEL: @red_ld_4xi64(
+; ZVFH256-NEXT: entry:
+; ZVFH256-NEXT: [[TMP0:%.*]] = load <4 x i64>, ptr [[PTR:%.*]], align 8
+; ZVFH256-NEXT: [[TMP1:%.*]] = call i64 @llvm.vector.reduce.add.v4i64(<4 x i64> [[TMP0]])
+; ZVFH256-NEXT: ret i64 [[TMP1]]
+;
+; ZVFH512-LABEL: @red_ld_4xi64(
+; ZVFH512-NEXT: entry:
+; ZVFH512-NEXT: [[TMP0:%.*]] = load <4 x i64>, ptr [[PTR:%.*]], align 8
+; ZVFH512-NEXT: [[TMP1:%.*]] = call i64 @llvm.vector.reduce.add.v4i64(<4 x i64> [[TMP0]])
+; ZVFH512-NEXT: ret i64 [[TMP1]]
;
entry:
%ld0 = load i64, ptr %ptr
@@ -147,11 +181,73 @@ entry:
define i64 @red_strided_ld_16xi64(ptr %ptr) {
-; CHECK-LABEL: @red_strided_ld_16xi64(
-; CHECK-NEXT: entry:
-; CHECK-NEXT: [[TMP0:%.*]] = call <16 x i64> @llvm.experimental.vp.strided.load.v16i64.p0.i64(ptr align 8 [[PTR:%.*]], i64 16, <16 x i1> splat (i1 true), i32 16)
-; CHECK-NEXT: [[TMP1:%.*]] = call i64 @llvm.vector.reduce.add.v16i64(<16 x i64> [[TMP0]])
-; CHECK-NEXT: ret i64 [[TMP1]]
+; ZVFHMIN-LABEL: @red_strided_ld_16xi64(
+; ZVFHMIN-NEXT: entry:
+; ZVFHMIN-NEXT: [[TMP0:%.*]] = call <16 x i64> @llvm.experimental.vp.strided.load.v16i64.p0.i64(ptr align 8 [[PTR:%.*]], i64 16, <16 x i1> splat (i1 true), i32 16)
+; ZVFHMIN-NEXT: [[TMP1:%.*]] = call i64 @llvm.vector.reduce.add.v16i64(<16 x i64> [[TMP0]])
+; ZVFHMIN-NEXT: ret i64 [[TMP1]]
+;
+; ZVFHDEFAULT-LABEL: @red_strided_ld_16xi64(
+; ZVFHDEFAULT-NEXT: entry:
+; ZVFHDEFAULT-NEXT: [[TMP0:%.*]] = call <16 x i64> @llvm.experimental.vp.strided.load.v16i64.p0.i64(ptr align 8 [[PTR:%.*]], i64 16, <16 x i1> splat (i1 true), i32 16)
+; ZVFHDEFAULT-NEXT: [[TMP1:%.*]] = call i64 @llvm.vector.reduce.add.v16i64(<16 x i64> [[TMP0]])
+; ZVFHDEFAULT-NEXT: ret i64 [[TMP1]]
+;
+; ZVFH256-LABEL: @red_strided_ld_16xi64(
+; ZVFH256-NEXT: entry:
+; ZVFH256-NEXT: [[TMP0:%.*]] = call <16 x i64> @llvm.experimental.vp.strided.load.v16i64.p0.i64(ptr align 8 [[PTR:%.*]], i64 16, <16 x i1> splat (i1 true), i32 16)
+; ZVFH256-NEXT: [[TMP1:%.*]] = call i64 @llvm.vector.reduce.add.v16i64(<16 x i64> [[TMP0]])
+; ZVFH256-NEXT: ret i64 [[TMP1]]
+;
+; ZVFH512-LABEL: @red_strided_ld_16xi64(
+; ZVFH512-NEXT: entry:
+; ZVFH512-NEXT: [[LD0:%.*]] = load i64, ptr [[PTR:%.*]], align 8
+; ZVFH512-NEXT: [[GEP:%.*]] = getelementptr inbounds i64, ptr [[PTR]], i64 2
+; ZVFH512-NEXT: [[LD1:%.*]] = load i64, ptr [[GEP]], align 8
+; ZVFH512-NEXT: [[ADD_1:%.*]] = add nuw nsw i64 [[LD0]], [[LD1]]
+; ZVFH512-NEXT: [[GEP_1:%.*]] = getelementptr inbounds i64, ptr [[PTR]], i64 4
+; ZVFH512-NEXT: [[LD2:%.*]] = load i64, ptr [[GEP_1]], align 8
+; ZVFH512-NEXT: [[ADD_2:%.*]] = add nuw nsw i64 [[ADD_1]], [[LD2]]
+; ZVFH512-NEXT: [[GEP_2:%.*]] = getelementptr inbounds i64, ptr [[PTR]], i64 6
+; ZVFH512-NEXT: [[LD3:%.*]] = load i64, ptr [[GEP_2]], align 8
+; ZVFH512-NEXT: [[ADD_3:%.*]] = add nuw nsw i64 [[ADD_2]], [[LD3]]
+; ZVFH512-NEXT: [[GEP_3:%.*]] = getelementptr inbounds i64, ptr [[PTR]], i64 8
+; ZVFH512-NEXT: [[LD4:%.*]] = load i64, ptr [[GEP_3]], align 8
+; ZVFH512-NEXT: [[ADD_4:%.*]] = add nuw nsw i64 [[ADD_3]], [[LD4]]
+; ZVFH512-NEXT: [[GEP_4:%.*]] = getelementptr inbounds i64, ptr [[PTR]], i64 10
+; ZVFH512-NEXT: [[LD5:%.*]] = load i64, ptr [[GEP_4]], align 8
+; ZVFH512-NEXT: [[ADD_5:%.*]] = add nuw nsw i64 [[ADD_4]], [[LD5]]
+; ZVFH512-NEXT: [[GEP_5:%.*]] = getelementptr inbounds i64, ptr [[PTR]], i64 12
+; ZVFH512-NEXT: [[LD6:%.*]] = load i64, ptr [[GEP_5]], align 8
+; ZVFH512-NEXT: [[ADD_6:%.*]] = add nuw nsw i64 [[ADD_5]], [[LD6]]
+; ZVFH512-NEXT: [[GEP_6:%.*]] = getelementptr inbounds i64, ptr [[PTR]], i64 14
+; ZVFH512-NEXT: [[LD7:%.*]] = load i64, ptr [[GEP_6]], align 8
+; ZVFH512-NEXT: [[ADD_7:%.*]] = add nuw nsw i64 [[ADD_6]], [[LD7]]
+; ZVFH512-NEXT: [[GEP_7:%.*]] = getelementptr inbounds i64, ptr [[PTR]], i64 16
+; ZVFH512-NEXT: [[LD8:%.*]] = load i64, ptr [[GEP_7]], align 8
+; ZVFH512-NEXT: [[ADD_8:%.*]] = add nuw nsw i64 [[ADD_7]], [[LD8]]
+; ZVFH512-NEXT: [[GEP_8:%.*]] = getelementptr inbounds i64, ptr [[PTR]], i64 18
+; ZVFH512-NEXT: [[LD9:%.*]] = load i64, ptr [[GEP_8]], align 8
+; ZVFH512-NEXT: [[ADD_9:%.*]] = add nuw nsw i64 [[ADD_8]], [[LD9]]
+; ZVFH512-NEXT: [[GEP_9:%.*]] = getelementptr inbounds i64, ptr [[PTR]], i64 20
+; ZVFH512-NEXT: [[LD10:%.*]] = load i64, ptr [[GEP_9]], align 8
+; ZVFH512-NEXT: [[ADD_10:%.*]] = add nuw nsw i64 [[ADD_9]], [[LD10]]
+; ZVFH512-NEXT: [[GEP_10:%.*]] = getelementptr inbounds i64, ptr [[PTR]], i64 22
+; ZVFH512-NEXT: [[LD11:%.*]] = load i64, ptr [[GEP_10]], align 8
+; ZVFH512-NEXT: [[ADD_11:%.*]] = add nuw nsw i64 [[ADD_10]], [[LD11]]
+; ZVFH512-NEXT: [[GEP_11:%.*]] = getelementptr inbounds i64, ptr [[PTR]], i64 24
+; ZVFH512-NEXT: [[LD12:%.*]] = load i64, ptr [[GEP_11]], align 8
+; ZVFH512-NEXT: [[ADD_12:%.*]] = add nuw nsw i64 [[ADD_11]], [[LD12]]
+; ZVFH512-NEXT: [[GEP_12:%.*]] = getelementptr inbounds i64, ptr [[PTR]], i64 26
+; ZVFH512-NEXT: [[LD13:%.*]] = load i64, ptr [[GEP_12]], align 8
+; ZVFH512-NEXT: [[ADD_13:%.*]] = add nuw nsw i64 [[ADD_12]], [[LD13]]
+; ZVFH512-NEXT: [[GEP_13:%.*]] = getelementptr inbounds i64, ptr [[PTR]], i64 28
+; ZVFH512-NEXT: [[LD14:%.*]] = load i64, ptr [[GEP_13]], align 8
+; ZVFH512-NEXT: [[ADD_14:%.*]] = add nuw nsw i64 [[ADD_13]], [[LD14]]
+; ZVFH512-NEXT: [[GEP_14:%.*]] = getelementptr inbounds i64, ptr [[PTR]], i64 30
+; ZVFH512-NEXT: [[LD15:%.*]] = load i64, ptr [[GEP_14]], align 8
+; ZVFH512-NEXT: [[ADD_15:%.*]] = add nuw nsw i64 [[ADD_14]], [[LD15]]
+; ZVFH512-NEXT: ret i64 [[ADD_15]]
;
entry:
%ld0 = load i64, ptr %ptr
@@ -867,10 +963,20 @@ entry:
define i64 @red_zext_ld_4xi64(ptr %ptr) {
; CHECK-LABEL: @red_zext_ld_4xi64(
; CHECK-NEXT: entry:
-; CHECK-NEXT: [[TMP0:%.*]] = load <4 x i8>, ptr [[PTR:%.*]], align 1
-; CHECK-NEXT: [[TMP1:%.*]] = zext <4 x i8> [[TMP0]] to <4 x i16>
-; CHECK-NEXT: [[TMP2:%.*]] = call i16 @llvm.vector.reduce.add.v4i16(<4 x i16> [[TMP1]])
-; CHECK-NEXT: [[ADD_3:%.*]] = zext i16 [[TMP2]] to i64
+; CHECK-NEXT: [[LD0:%.*]] = load i8, ptr [[PTR:%.*]], align 1
+; CHECK-NEXT: [[ZEXT:%.*]] = zext i8 [[LD0]] to i64
+; CHECK-NEXT: [[GEP:%.*]] = getelementptr inbounds i8, ptr [[PTR]], i64 1
+; CHECK-NEXT: [[LD1:%.*]] = load i8, ptr [[GEP]], align 1
+; CHECK-NEXT: [[ZEXT_1:%.*]] = zext i8 [[LD1]] to i64
+; CHECK-NEXT: [[ADD_1:%.*]] = add nuw nsw i64 [[ZEXT]], [[ZEXT_1]]
+; CHECK-NEXT: [[GEP_1:%.*]] = getelementptr inbounds i8, ptr [[PTR]], i64 2
+; CHECK-NEXT: [[LD2:%.*]] = load i8, ptr [[GEP_1]], align 1
+; CHECK-NEXT: [[ZEXT_2:%.*]] = zext i8 [[LD2]] to i64
+; CHECK-NEXT: [[ADD_2:%.*]] = add nuw nsw i64 [[ADD_1]], [[ZEXT_2]]
+; CHECK-NEXT: [[GEP_2:%.*]] = getelementptr inbounds i8, ptr [[PTR]], i64 3
+; CHECK-NEXT: [[LD3:%.*]] = load i8, ptr [[GEP_2]], align 1
+; CHECK-NEXT: [[ZEXT_3:%.*]] = zext i8 [[LD3]] to i64
+; CHECK-NEXT: [[ADD_3:%.*]] = add nuw nsw i64 [[ADD_2]], [[ZEXT_3]]
; CHECK-NEXT: ret i64 [[ADD_3]]
;
entry:
diff --git a/llvm/test/Transforms/SLPVectorizer/RISCV/reordered-interleaved-loads.ll b/llvm/test/Transforms/SLPVectorizer/RISCV/reordered-interleaved-loads.ll
index 7d761244d3f30..ffb004c59b8af 100644
--- a/llvm/test/Transforms/SLPVectorizer/RISCV/reordered-interleaved-loads.ll
+++ b/llvm/test/Transforms/SLPVectorizer/RISCV/reordered-interleaved-loads.ll
@@ -9,54 +9,73 @@ define i1 @test(i32 %conv15.12, i16 %0, ptr %1, i16 %2, i16 %3, i16 %4, i16 %5,
; CHECK-SAME: i32 [[CONV15_12:%.*]], i16 [[TMP0:%.*]], ptr [[TMP1:%.*]], i16 [[TMP2:%.*]], i16 [[TMP3:%.*]], i16 [[TMP4:%.*]], i16 [[TMP5:%.*]], i32 [[CONV15_1_3:%.*]], i16 [[TMP6:%.*]], i32 [[CONV15_1_4:%.*]]) #[[ATTR0:[0-9]+]] {
; CHECK-NEXT: [[ENTRY:.*:]]
; CHECK-NEXT: [[TMP7:%.*]] = load i16, ptr [[TMP1]], align 2
-; CHECK-NEXT: [[TMP8:%.*]] = load i16, ptr @h, align 2
-; CHECK-NEXT: [[TMP9:%.*]] = insertelement <4 x i16> poison, i16 [[TMP2]], i64 0
-; CHECK-NEXT: [[TMP10:%.*]] = insertelement <4 x i16> [[TMP9]], i16 [[TMP0]], i64 2
-; CHECK-NEXT: [[TMP11:%.*]] = insertelement <4 x i16> [[TMP10]], i16 [[TMP8]], i64 1
-; CHECK-NEXT: [[TMP12:%.*]] = insertelement <4 x i16> [[TMP11]], i16 [[TMP7]], i64 3
-; CHECK-NEXT: [[TMP13:%.*]] = sext <4 x i16> [[TMP12]] to <4 x i32>
-; CHECK-NEXT: [[TMP14:%.*]] = call <4 x i32> @llvm.smax.v4i32(<4 x i32> [[TMP13]], <4 x i32> zeroinitializer)
-; CHECK-NEXT: [[TMP15:%.*]] = icmp eq <4 x i16> [[TMP12]], zeroinitializer
-; CHECK-NEXT: [[TMP16:%.*]] = select <4 x i1> [[TMP15]], <4 x i32> splat (i32 8), <4 x i32> [[TMP13]]
-; CHECK-NEXT: [[TMP17:%.*]] = insertelement <4 x i32> [[TMP13]], i32 [[CONV15_12]], i64 3
-; CHECK-NEXT: [[TMP18:%.*]] = xor <4 x i32> [[TMP16]], [[TMP17]]
-; CHECK-NEXT: [[TMP19:%.*]] = icmp sgt <4 x i32> [[TMP14]], [[TMP18]]
-; CHECK-NEXT: [[TMP29:%.*]] = extractelement <4 x i1> [[TMP19]], i64 3
+; CHECK-NEXT: [[CONV15_121:%.*]] = sext i16 [[TMP7]] to i32
+; CHECK-NEXT: [[COND_13:%.*]] = tail call i32 @llvm.smax.i32(i32 [[CONV15_121]], i32 0)
+; CHECK-NEXT: [[TOBOOL_NOT_14:%.*]] = icmp eq i16 [[TMP7]], 0
+; CHECK-NEXT: [[COND27_15:%.*]] = select i1 [[TOBOOL_NOT_14]], i32 8, i32 [[CONV15_121]]
+; CHECK-NEXT: [[XOR_16:%.*]] = xor i32 [[COND27_15]], [[CONV15_12]]
+; CHECK-NEXT: [[TMP29:%.*]] = icmp sgt i32 [[COND_13]], [[XOR_16]]
; CHECK-NEXT: [[CONV30_18:%.*]] = zext i1 [[TMP29]] to i16
; CHECK-NEXT: store i16 [[CONV30_18]], ptr @a, align 2
-; CHECK-NEXT: [[TMP30:%.*]] = extractelement <4 x i1> [[TMP19]], i64 2
+; CHECK-NEXT: [[CONV15_213:%.*]] = sext i16 [[TMP0]] to i32
+; CHECK-NEXT: [[COND_214:%.*]] = tail call i32 @llvm.smax.i32(i32 [[CONV15_213]], i32 0)
+; CHECK-NEXT: [[TOBOOL_NOT_215:%.*]] = icmp eq i16 [[TMP0]], 0
+; CHECK-NEXT: [[COND27_216:%.*]] = select i1 [[TOBOOL_NOT_215]], i32 8, i32 [[CONV15_213]]
+; CHECK-NEXT: [[XOR_217:%.*]] = xor i32 [[COND27_216]], [[CONV15_213]]
+; CHECK-NEXT: [[TMP30:%.*]] = icmp sgt i32 [[COND_214]], [[XOR_217]]
; CHECK-NEXT: [[CONV30_219:%.*]] = zext i1 [[TMP30]] to i16
; CHECK-NEXT: store i16 [[CONV30_219]], ptr @a, align 2
-; CHECK-NEXT: [[TMP31:%.*]] = extractelement <4 x i1> [[TMP19]], i64 1
+; CHECK-NEXT: [[TMP8:%.*]] = load i16, ptr @h, align 2
+; CHECK-NEXT: [[CONV15_324:%.*]] = sext i16 [[TMP8]] to i32
+; CHECK-NEXT: [[COND_325:%.*]] = tail call i32 @llvm.smax.i32(i32 [[CONV15_324]], i32 0)
+; CHECK-NEXT: [[TOBOOL_NOT_326:%.*]] = icmp eq i16 [[TMP8]], 0
+; CHECK-NEXT: [[COND27_327:%.*]] = select i1 [[TOBOOL_NOT_326]], i32 8, i32 [[CONV15_324]]
+; CHECK-NEXT: [[XOR_328:%.*]] = xor i32 [[COND27_327]], [[CONV15_324]]
+; CHECK-NEXT: [[TMP31:%.*]] = icmp sgt i32 [[COND_325]], [[XOR_328]]
; CHECK-NEXT: [[CONV30_330:%.*]] = zext i1 [[TMP31]] to i16
; CHECK-NEXT: store i16 [[CONV30_330]], ptr @a, align 2
-; CHECK-NEXT: [[TMP32:%.*]] = extractelement <4 x i1> [[TMP19]], i64 0
+; CHECK-NEXT: [[CONV15_4:%.*]] = sext i16 [[TMP2]] to i32
+; CHECK-NEXT: [[COND_4:%.*]] = tail call i32 @llvm.smax.i32(i32 [[CONV15_4]], i32 0)
+; CHECK-NEXT: [[TOBOOL_NOT_4:%.*]] = icmp eq i16 [[TMP2]], 0
+; CHECK-NEXT: [[COND27_4:%.*]] = select i1 [[TOBOOL_NOT_4]], i32 8, i32 [[CONV15_4]]
+; CHECK-NEXT: [[XOR_4:%.*]] = xor i32 [[COND27_4]], [[CONV15_4]]
+; CHECK-NEXT: [[TMP32:%.*]] = icmp sgt i32 [[COND_4]], [[XOR_4]]
; CHECK-NEXT: [[CONV30_4:%.*]] = zext i1 [[TMP32]] to i16
; CHECK-NEXT: store i16 [[CONV30_4]], ptr @a, align 2
-; CHECK-NEXT: [[TMP24:%.*]] = insertelement <4 x i16> poison, i16 [[TMP3]], i64 0
-; CHECK-NEXT: [[TMP25:%.*]] = insertelement <4 x i16> [[TMP24]], i16 [[TMP4]], i64 1
-; CHECK-NEXT: [[TMP26:%.*]] = insertelement <4 x i16> [[TMP25]], i16 [[TMP5]], i64 2
-; CHECK-NEXT: [[TMP27:%.*]] = insertelement <4 x i16> [[TMP26]], i16 [[TMP6]], i64 3
-; CHECK-NEXT: [[TMP28:%.*]] = sext <4 x i16> [[TMP27]] to <4 x i32>
-; CHECK-NEXT: [[TMP38:%.*]] = load <16 x i16>, ptr getelementptr inbounds nuw (i8, ptr @h, i64 6), align 2
-; CHECK-NEXT: [[TMP39:%.*]] = shufflevector <16 x i16> [[TMP38]], <16 x i16> poison, <4 x i32> <i32 0, i32 3, i32 6, i32 9>
-; CHECK-NEXT: [[TMP40:%.*]] = call <4 x i32> @llvm.smax.v4i32(<4 x i32> [[TMP28]], <4 x i32> zeroinitializer)
-; CHECK-NEXT: [[TMP41:%.*]] = icmp eq <4 x i16> [[TMP39]], zeroinitializer
-; CHECK-NEXT: [[TMP42:%.*]] = select <4 x i1> [[TMP41]], <4 x i32> splat (i32 8), <4 x i32> [[TMP28]]
-; CHECK-NEXT: [[TMP43:%.*]] = insertelement <4 x i32> [[TMP28]], i32 [[CONV15_1_3]], i64 2
-; CHECK-NEXT: [[TMP44:%.*]] = insertelement <4 x i32> [[TMP43]], i32 [[CONV15_1_4]], i64 3
-; CHECK-NEXT: [[TMP45:%.*]] = xor <4 x i32> [[TMP42]], [[TMP44]]
-; CHECK-NEXT: [[TMP37:%.*]] = icmp sgt <4 x i32> [[TMP40]], [[TMP45]]
-; CHECK-NEXT: [[TMP33:%.*]] = extractelement <4 x i1> [[TMP37]], i64 0
+; CHECK-NEXT: [[TMP9:%.*]] = load i16, ptr getelementptr inbounds nuw (i8, ptr @h, i64 6), align 2
+; CHECK-NEXT: [[CONV15_1_1:%.*]] = sext i16 [[TMP3]] to i32
+; CHECK-NEXT: [[COND_1_1:%.*]] = tail call i32 @llvm.smax.i32(i32 [[CONV15_1_1]], i32 0)
+; CHECK-NEXT: [[TOBOOL_NOT_1_1:%.*]] = icmp eq i16 [[TMP9]], 0
+; CHECK-NEXT: [[COND27_1_1:%.*]] = select i1 [[TOBOOL_NOT_1_1]], i32 8, i32 [[CONV15_1_1]]
+; CHECK-NEXT: [[XOR_1_1:%.*]] = xor i32 [[COND27_1_1]], [[CONV15_1_1]]
+; CHECK-NEXT: [[TMP33:%.*]] = icmp sgt i32 [[COND_1_1]], [[XOR_1_1]]
; CHECK-NEXT: [[CONV30_1_1:%.*]] = zext i1 [[TMP33]] to i16
; CHECK-NEXT: store i16 [[CONV30_1_1]], ptr @a, align 2
-; CHECK-NEXT: [[TMP34:%.*]] = extractelement <4 x i1> [[TMP37]], i64 1
+; CHECK-NEXT: [[TMP10:%.*]] = load i16, ptr getelementptr inbounds nuw (i8, ptr @h, i64 12), align 4
+; CHECK-NEXT: [[CONV15_1_2:%.*]] = sext i16 [[TMP4]] to i32
+; CHECK-NEXT: [[COND_1_2:%.*]] = tail call i32 @llvm.smax.i32(i32 [[CONV15_1_2]], i32 0)
+; CHECK-NEXT: [[TOBOOL_NOT_1_2:%.*]] = icmp eq i16 [[TMP10]], 0
+; CHECK-NEXT: [[COND27_1_2:%.*]] = select i1 [[TOBOOL_NOT_1_2]], i32 8, i32 [[CONV15_1_2]]
+; CHECK-NEXT: [[XOR_1_2:%.*]] = xor i32 [[COND27_1_2]], [[CONV15_1_2]]
+; CHECK-NEXT: [[TMP34:%.*]] = icmp sgt i32 [[COND_1_2]], [[XOR_1_2]]
; CHECK-NEXT: [[CONV30_1_2:%.*]] = zext i1 [[TMP34]] to i16
; CHECK-NEXT: store i16 [[CONV30_1_2]], ptr @a, align 2
-; CHECK-NEXT: [[TMP35:%.*]] = extractelement <4 x i1> [[TMP37]], i64 2
+; CHECK-NEXT: [[TMP11:%.*]] = load i16, ptr getelementptr inbounds nuw (i8, ptr @h, i64 18), align 2
+; CHECK-NEXT: [[CONV15_1_32:%.*]] = sext i16 [[TMP5]] to i32
+; CHECK-NEXT: [[COND_1_3:%.*]] = tail call i32 @llvm.smax.i32(i32 [[CONV15_1_32]], i32 0)
+; CHECK-NEXT: [[TOBOOL_NOT_1_3:%.*]] = icmp eq i16 [[TMP11]], 0
+; CHECK-NEXT: [[COND27_1_3:%.*]] = select i1 [[TOBOOL_NOT_1_3]], i32 8, i32 [[CONV15_1_32]]
+; CHECK-NEXT: [[XOR_1_3:%.*]] = xor i32 [[COND27_1_3]], [[CONV15_1_3]]
+; CHECK-NEXT: [[TMP35:%.*]] = icmp sgt i32 [[COND_1_3]], [[XOR_1_3]]
; CHECK-NEXT: [[CONV30_1_3:%.*]] = zext i1 [[TMP35]] to i16
; CHECK-NEXT: store i16 [[CONV30_1_3]], ptr @a, align 2
-; CHECK-NEXT: [[TMP36:%.*]] = extractelement <4 x i1> [[TMP37]], i64 3
+; CHECK-NEXT: [[TMP12:%.*]] = load i16, ptr getelementptr inbounds nuw (i8, ptr @h, i64 24), align 8
+; CHECK-NEXT: [[CONV15_1_43:%.*]] = sext i16 [[TMP6]] to i32
+; CHECK-NEXT: [[COND_1_4:%.*]] = tail call i32 @llvm.smax.i32(i32 [[CONV15_1_43]], i32 0)
+; CHECK-NEXT: [[TOBOOL_NOT_1_4:%.*]] = icmp eq i16 [[TMP12]], 0
+; CHECK-NEXT: [[COND27_1_4:%.*]] = select i1 [[TOBOOL_NOT_1_4]], i32 8, i32 [[CONV15_1_43]]
+; CHECK-NEXT: [[XOR_1_4:%.*]] = xor i32 [[COND27_1_4]], [[CONV15_1_4]]
+; CHECK-NEXT: [[TMP36:%.*]] = icmp sgt i32 [[COND_1_4]], [[XOR_1_4]]
; CHECK-NEXT: ret i1 [[TMP36]]
;
entry:
diff --git a/llvm/test/Transforms/SLPVectorizer/RISCV/runtime-strided-stores.ll b/llvm/test/Transforms/SLPVectorizer/RISCV/runtime-strided-stores.ll
index 6eb4ddc12ef50..9c759761b14f8 100644
--- a/llvm/test/Transforms/SLPVectorizer/RISCV/runtime-strided-stores.ll
+++ b/llvm/test/Transforms/SLPVectorizer/RISCV/runtime-strided-stores.ll
@@ -6,7 +6,21 @@ define void @runtime_stride(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-LABEL: define void @runtime_stride(
; CHECK-SAME: ptr [[PL:%.*]], ptr [[PS:%.*]], i64 [[STRIDE:%.*]]) #[[ATTR0:[0-9]+]] {
; CHECK-NEXT: [[GEP_L0:%.*]] = getelementptr i8, ptr [[PL]], i64 0
+; CHECK-NEXT: [[GEP_L1:%.*]] = getelementptr i8, ptr [[PL]], i64 1
+; CHECK-NEXT: [[GEP_L2:%.*]] = getelementptr i8, ptr [[PL]], i64 2
+; CHECK-NEXT: [[GEP_L3:%.*]] = getelementptr i8, ptr [[PL]], i64 3
+; CHECK-NEXT: [[GEP_L4:%.*]] = getelementptr i8, ptr [[PL]], i64 4
+; CHECK-NEXT: [[GEP_L5:%.*]] = getelementptr i8, ptr [[PL]], i64 5
+; CHECK-NEXT: [[GEP_L6:%.*]] = getelementptr i8, ptr [[PL]], i64 6
+; CHECK-NEXT: [[GEP_L7:%.*]] = getelementptr i8, ptr [[PL]], i64 7
+; CHECK-NEXT: [[TMP9:%.*]] = load i8, ptr [[GEP_L7]], align 1
; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i8>, ptr [[GEP_L0]], align 1
+; CHECK-NEXT: [[TMP8:%.*]] = load i8, ptr [[GEP_L6]], align 1
+; CHECK-NEXT: [[TMP7:%.*]] = load i8, ptr [[GEP_L5]], align 1
+; CHECK-NEXT: [[TMP6:%.*]] = load i8, ptr [[GEP_L4]], align 1
+; CHECK-NEXT: [[TMP5:%.*]] = load i8, ptr [[GEP_L3]], align 1
+; CHECK-NEXT: [[TMP4:%.*]] = load i8, ptr [[GEP_L2]], align 1
+; CHECK-NEXT: [[TMP3:%.*]] = load i8, ptr [[GEP_L1]], align 1
; CHECK-NEXT: [[GEP_S0:%.*]] = getelementptr i8, ptr [[PS]], i64 0
; CHECK-NEXT: [[GEP_S1:%.*]] = getelementptr i8, ptr [[GEP_S0]], i64 [[STRIDE]]
; CHECK-NEXT: [[GEP_S2:%.*]] = getelementptr i8, ptr [[GEP_S1]], i64 [[STRIDE]]
@@ -17,19 +31,12 @@ define void @runtime_stride(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-NEXT: [[GEP_S7:%.*]] = getelementptr i8, ptr [[GEP_S6]], i64 [[STRIDE]]
; CHECK-NEXT: [[TMP2:%.*]] = extractelement <8 x i8> [[TMP1]], i64 0
; CHECK-NEXT: store i8 [[TMP2]], ptr [[GEP_S0]], align 1
-; CHECK-NEXT: [[TMP3:%.*]] = extractelement <8 x i8> [[TMP1]], i64 1
; CHECK-NEXT: store i8 [[TMP3]], ptr [[GEP_S1]], align 1
-; CHECK-NEXT: [[TMP4:%.*]] = extractelement <8 x i8> [[TMP1]], i64 2
; CHECK-NEXT: store i8 [[TMP4]], ptr [[GEP_S2]], align 1
-; CHECK-NEXT: [[TMP5:%.*]] = extractelement <8 x i8> [[TMP1]], i64 3
; CHECK-NEXT: store i8 [[TMP5]], ptr [[GEP_S3]], align 1
-; CHECK-NEXT: [[TMP6:%.*]] = extractelement <8 x i8> [[TMP1]], i64 4
; CHECK-NEXT: store i8 [[TMP6]], ptr [[GEP_S4]], align 1
-; CHECK-NEXT: [[TMP7:%.*]] = extractelement <8 x i8> [[TMP1]], i64 5
; CHECK-NEXT: store i8 [[TMP7]], ptr [[GEP_S5]], align 1
-; CHECK-NEXT: [[TMP8:%.*]] = extractelement <8 x i8> [[TMP1]], i64 6
; CHECK-NEXT: store i8 [[TMP8]], ptr [[GEP_S6]], align 1
-; CHECK-NEXT: [[TMP9:%.*]] = extractelement <8 x i8> [[TMP1]], i64 7
; CHECK-NEXT: store i8 [[TMP9]], ptr [[GEP_S7]], align 1
; CHECK-NEXT: ret void
;
@@ -76,7 +83,21 @@ define void @runtime_stride_complex_scev(ptr %pl, ptr %ps, i64 %stride, i64 %str
; CHECK-LABEL: define void @runtime_stride_complex_scev(
; CHECK-SAME: ptr [[PL:%.*]], ptr [[PS:%.*]], i64 [[STRIDE:%.*]], i64 [[STRIDE1:%.*]]) #[[ATTR0]] {
; CHECK-NEXT: [[GEP_L0:%.*]] = getelementptr i8, ptr [[PL]], i64 0
+; CHECK-NEXT: [[GEP_L1:%.*]] = getelementptr i8, ptr [[PL]], i64 1
+; CHECK-NEXT: [[GEP_L2:%.*]] = getelementptr i8, ptr [[PL]], i64 2
+; CHECK-NEXT: [[GEP_L3:%.*]] = getelementptr i8, ptr [[PL]], i64 3
+; CHECK-NEXT: [[GEP_L4:%.*]] = getelementptr i8, ptr [[PL]], i64 4
+; CHECK-NEXT: [[GEP_L5:%.*]] = getelementptr i8, ptr [[PL]], i64 5
+; CHECK-NEXT: [[GEP_L6:%.*]] = getelementptr i8, ptr [[PL]], i64 6
+; CHECK-NEXT: [[GEP_L7:%.*]] = getelementptr i8, ptr [[PL]], i64 7
+; CHECK-NEXT: [[TMP9:%.*]] = load i8, ptr [[GEP_L7]], align 1
; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i8>, ptr [[GEP_L0]], align 1
+; CHECK-NEXT: [[TMP8:%.*]] = load i8, ptr [[GEP_L6]], align 1
+; CHECK-NEXT: [[TMP7:%.*]] = load i8, ptr [[GEP_L5]], align 1
+; CHECK-NEXT: [[TMP6:%.*]] = load i8, ptr [[GEP_L4]], align 1
+; CHECK-NEXT: [[TMP5:%.*]] = load i8, ptr [[GEP_L3]], align 1
+; CHECK-NEXT: [[TMP4:%.*]] = load i8, ptr [[GEP_L2]], align 1
+; CHECK-NEXT: [[TMP3:%.*]] = load i8, ptr [[GEP_L1]], align 1
; CHECK-NEXT: [[GEP_S0:%.*]] = getelementptr i8, ptr [[PS]], i64 0
; CHECK-NEXT: [[GEP_S1:%.*]] = getelementptr i8, ptr [[GEP_S0]], i64 [[STRIDE]]
; CHECK-NEXT: [[GEP_S2:%.*]] = getelementptr i8, ptr [[GEP_S1]], i64 [[STRIDE]]
@@ -95,19 +116,12 @@ define void @runtime_stride_complex_scev(ptr %pl, ptr %ps, i64 %stride, i64 %str
; CHECK-NEXT: [[GEP_SS7:%.*]] = getelementptr i8, ptr [[GEP_S7]], i64 [[STRIDE1]]
; CHECK-NEXT: [[TMP2:%.*]] = extractelement <8 x i8> [[TMP1]], i64 0
; CHECK-NEXT: store i8 [[TMP2]], ptr [[GEP_SS0]], align 1
-; CHECK-NEXT: [[TMP3:%.*]] = extractelement <8 x i8> [[TMP1]], i64 1
; CHECK-NEXT: store i8 [[TMP3]], ptr [[GEP_SS1]], align 1
-; CHECK-NEXT: [[TMP4:%.*]] = extractelement <8 x i8> [[TMP1]], i64 2
; CHECK-NEXT: store i8 [[TMP4]], ptr [[GEP_SS2]], align 1
-; CHECK-NEXT: [[TMP5:%.*]] = extractelement <8 x i8> [[TMP1]], i64 3
; CHECK-NEXT: store i8 [[TMP5]], ptr [[GEP_SS3]], align 1
-; CHECK-NEXT: [[TMP6:%.*]] = extractelement <8 x i8> [[TMP1]], i64 4
; CHECK-NEXT: store i8 [[TMP6]], ptr [[GEP_SS4]], align 1
-; CHECK-NEXT: [[TMP7:%.*]] = extractelement <8 x i8> [[TMP1]], i64 5
; CHECK-NEXT: store i8 [[TMP7]], ptr [[GEP_SS5]], align 1
-; CHECK-NEXT: [[TMP8:%.*]] = extractelement <8 x i8> [[TMP1]], i64 6
; CHECK-NEXT: store i8 [[TMP8]], ptr [[GEP_SS6]], align 1
-; CHECK-NEXT: [[TMP9:%.*]] = extractelement <8 x i8> [[TMP1]], i64 7
; CHECK-NEXT: store i8 [[TMP9]], ptr [[GEP_SS7]], align 1
; CHECK-NEXT: ret void
;
@@ -164,7 +178,21 @@ define void @two_runtime_strides(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-LABEL: define void @two_runtime_strides(
; CHECK-SAME: ptr [[PL:%.*]], ptr [[PS:%.*]], i64 [[STRIDE:%.*]]) #[[ATTR0]] {
; CHECK-NEXT: [[GEP_L0:%.*]] = getelementptr i8, ptr [[PL]], i64 0
+; CHECK-NEXT: [[GEP_L1:%.*]] = getelementptr i8, ptr [[PL]], i64 1
+; CHECK-NEXT: [[GEP_L2:%.*]] = getelementptr i8, ptr [[PL]], i64 2
+; CHECK-NEXT: [[GEP_L3:%.*]] = getelementptr i8, ptr [[PL]], i64 3
+; CHECK-NEXT: [[GEP_L4:%.*]] = getelementptr i8, ptr [[PL]], i64 4
+; CHECK-NEXT: [[GEP_L5:%.*]] = getelementptr i8, ptr [[PL]], i64 5
+; CHECK-NEXT: [[GEP_L6:%.*]] = getelementptr i8, ptr [[PL]], i64 6
+; CHECK-NEXT: [[GEP_L7:%.*]] = getelementptr i8, ptr [[PL]], i64 7
+; CHECK-NEXT: [[TMP9:%.*]] = load i8, ptr [[GEP_L7]], align 1
; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i8>, ptr [[GEP_L0]], align 1
+; CHECK-NEXT: [[TMP8:%.*]] = load i8, ptr [[GEP_L6]], align 1
+; CHECK-NEXT: [[TMP7:%.*]] = load i8, ptr [[GEP_L5]], align 1
+; CHECK-NEXT: [[TMP6:%.*]] = load i8, ptr [[GEP_L4]], align 1
+; CHECK-NEXT: [[TMP5:%.*]] = load i8, ptr [[GEP_L3]], align 1
+; CHECK-NEXT: [[TMP4:%.*]] = load i8, ptr [[GEP_L2]], align 1
+; CHECK-NEXT: [[TMP3:%.*]] = load i8, ptr [[GEP_L1]], align 1
; CHECK-NEXT: [[PS1:%.*]] = mul i64 [[STRIDE]], 16
; CHECK-NEXT: [[GEP_S0:%.*]] = getelementptr i8, ptr [[PS]], i64 0
; CHECK-NEXT: [[GEP_S1:%.*]] = getelementptr i8, ptr [[GEP_S0]], i64 [[STRIDE]]
@@ -176,19 +204,12 @@ define void @two_runtime_strides(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-NEXT: [[GEP_S7:%.*]] = getelementptr i8, ptr [[GEP_S6]], i64 [[STRIDE]]
; CHECK-NEXT: [[TMP2:%.*]] = extractelement <8 x i8> [[TMP1]], i64 0
; CHECK-NEXT: store i8 [[TMP2]], ptr [[GEP_S0]], align 1
-; CHECK-NEXT: [[TMP3:%.*]] = extractelement <8 x i8> [[TMP1]], i64 1
; CHECK-NEXT: store i8 [[TMP3]], ptr [[GEP_S1]], align 1
-; CHECK-NEXT: [[TMP4:%.*]] = extractelement <8 x i8> [[TMP1]], i64 2
; CHECK-NEXT: store i8 [[TMP4]], ptr [[GEP_S2]], align 1
-; CHECK-NEXT: [[TMP5:%.*]] = extractelement <8 x i8> [[TMP1]], i64 3
; CHECK-NEXT: store i8 [[TMP5]], ptr [[GEP_S3]], align 1
-; CHECK-NEXT: [[TMP6:%.*]] = extractelement <8 x i8> [[TMP1]], i64 4
; CHECK-NEXT: store i8 [[TMP6]], ptr [[GEP_S4]], align 1
-; CHECK-NEXT: [[TMP7:%.*]] = extractelement <8 x i8> [[TMP1]], i64 5
; CHECK-NEXT: store i8 [[TMP7]], ptr [[GEP_S5]], align 1
-; CHECK-NEXT: [[TMP8:%.*]] = extractelement <8 x i8> [[TMP1]], i64 6
; CHECK-NEXT: store i8 [[TMP8]], ptr [[GEP_S6]], align 1
-; CHECK-NEXT: [[TMP9:%.*]] = extractelement <8 x i8> [[TMP1]], i64 7
; CHECK-NEXT: store i8 [[TMP9]], ptr [[GEP_S7]], align 1
; CHECK-NEXT: ret void
;
@@ -236,7 +257,21 @@ define void @runtime_strides_constant_gap(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-LABEL: define void @runtime_strides_constant_gap(
; CHECK-SAME: ptr [[PL:%.*]], ptr [[PS:%.*]], i64 [[STRIDE:%.*]]) #[[ATTR0]] {
; CHECK-NEXT: [[GEP_L0:%.*]] = getelementptr i8, ptr [[PL]], i64 0
+; CHECK-NEXT: [[GEP_L1:%.*]] = getelementptr i8, ptr [[PL]], i64 1
+; CHECK-NEXT: [[GEP_L2:%.*]] = getelementptr i8, ptr [[PL]], i64 2
+; CHECK-NEXT: [[GEP_L3:%.*]] = getelementptr i8, ptr [[PL]], i64 3
+; CHECK-NEXT: [[GEP_L4:%.*]] = getelementptr i8, ptr [[PL]], i64 4
+; CHECK-NEXT: [[GEP_L5:%.*]] = getelementptr i8, ptr [[PL]], i64 5
+; CHECK-NEXT: [[GEP_L6:%.*]] = getelementptr i8, ptr [[PL]], i64 6
+; CHECK-NEXT: [[GEP_L7:%.*]] = getelementptr i8, ptr [[PL]], i64 7
+; CHECK-NEXT: [[TMP9:%.*]] = load i8, ptr [[GEP_L7]], align 1
; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i8>, ptr [[GEP_L0]], align 1
+; CHECK-NEXT: [[TMP8:%.*]] = load i8, ptr [[GEP_L6]], align 1
+; CHECK-NEXT: [[TMP7:%.*]] = load i8, ptr [[GEP_L5]], align 1
+; CHECK-NEXT: [[TMP6:%.*]] = load i8, ptr [[GEP_L4]], align 1
+; CHECK-NEXT: [[TMP5:%.*]] = load i8, ptr [[GEP_L3]], align 1
+; CHECK-NEXT: [[TMP4:%.*]] = load i8, ptr [[GEP_L2]], align 1
+; CHECK-NEXT: [[TMP3:%.*]] = load i8, ptr [[GEP_L1]], align 1
; CHECK-NEXT: [[GEP_S0:%.*]] = getelementptr i8, ptr [[PS]], i64 0
; CHECK-NEXT: [[GEP_S1:%.*]] = getelementptr i8, ptr [[GEP_S0]], i64 [[STRIDE]]
; CHECK-NEXT: [[GEP_S2:%.*]] = getelementptr i8, ptr [[GEP_S1]], i64 [[STRIDE]]
@@ -247,19 +282,12 @@ define void @runtime_strides_constant_gap(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-NEXT: [[GEP_S7:%.*]] = getelementptr i8, ptr [[GEP_S6]], i64 [[STRIDE]]
; CHECK-NEXT: [[TMP2:%.*]] = extractelement <8 x i8> [[TMP1]], i64 0
; CHECK-NEXT: store i8 [[TMP2]], ptr [[GEP_S0]], align 1
-; CHECK-NEXT: [[TMP3:%.*]] = extractelement <8 x i8> [[TMP1]], i64 1
; CHECK-NEXT: store i8 [[TMP3]], ptr [[GEP_S1]], align 1
-; CHECK-NEXT: [[TMP4:%.*]] = extractelement <8 x i8> [[TMP1]], i64 2
; CHECK-NEXT: store i8 [[TMP4]], ptr [[GEP_S2]], align 1
-; CHECK-NEXT: [[TMP5:%.*]] = extractelement <8 x i8> [[TMP1]], i64 3
; CHECK-NEXT: store i8 [[TMP5]], ptr [[GEP_S3]], align 1
-; CHECK-NEXT: [[TMP6:%.*]] = extractelement <8 x i8> [[TMP1]], i64 4
; CHECK-NEXT: store i8 [[TMP6]], ptr [[GEP_S4]], align 1
-; CHECK-NEXT: [[TMP7:%.*]] = extractelement <8 x i8> [[TMP1]], i64 5
; CHECK-NEXT: store i8 [[TMP7]], ptr [[GEP_S5]], align 1
-; CHECK-NEXT: [[TMP8:%.*]] = extractelement <8 x i8> [[TMP1]], i64 6
; CHECK-NEXT: store i8 [[TMP8]], ptr [[GEP_S6]], align 1
-; CHECK-NEXT: [[TMP9:%.*]] = extractelement <8 x i8> [[TMP1]], i64 7
; CHECK-NEXT: store i8 [[TMP9]], ptr [[GEP_S7]], align 1
; CHECK-NEXT: ret void
;
@@ -306,7 +334,21 @@ define void @overlapping_strides(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-LABEL: define void @overlapping_strides(
; CHECK-SAME: ptr [[PL:%.*]], ptr [[PS:%.*]], i64 [[STRIDE:%.*]]) #[[ATTR0]] {
; CHECK-NEXT: [[GEP_L0:%.*]] = getelementptr i8, ptr [[PL]], i64 0
+; CHECK-NEXT: [[GEP_L1:%.*]] = getelementptr i8, ptr [[PL]], i64 1
+; CHECK-NEXT: [[GEP_L2:%.*]] = getelementptr i8, ptr [[PL]], i64 2
+; CHECK-NEXT: [[GEP_L3:%.*]] = getelementptr i8, ptr [[PL]], i64 3
+; CHECK-NEXT: [[GEP_L4:%.*]] = getelementptr i8, ptr [[PL]], i64 4
+; CHECK-NEXT: [[GEP_L5:%.*]] = getelementptr i8, ptr [[PL]], i64 5
+; CHECK-NEXT: [[GEP_L6:%.*]] = getelementptr i8, ptr [[PL]], i64 6
+; CHECK-NEXT: [[GEP_L7:%.*]] = getelementptr i8, ptr [[PL]], i64 7
+; CHECK-NEXT: [[TMP9:%.*]] = load i8, ptr [[GEP_L7]], align 1
; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i8>, ptr [[GEP_L0]], align 1
+; CHECK-NEXT: [[TMP8:%.*]] = load i8, ptr [[GEP_L6]], align 1
+; CHECK-NEXT: [[TMP7:%.*]] = load i8, ptr [[GEP_L5]], align 1
+; CHECK-NEXT: [[TMP6:%.*]] = load i8, ptr [[GEP_L4]], align 1
+; CHECK-NEXT: [[TMP5:%.*]] = load i8, ptr [[GEP_L3]], align 1
+; CHECK-NEXT: [[TMP4:%.*]] = load i8, ptr [[GEP_L2]], align 1
+; CHECK-NEXT: [[TMP3:%.*]] = load i8, ptr [[GEP_L1]], align 1
; CHECK-NEXT: [[STRIDE2:%.*]] = mul i64 [[STRIDE]], 1
; CHECK-NEXT: [[STRIDE1:%.*]] = mul i64 [[STRIDE]], 2
; CHECK-NEXT: [[GEP_S0:%.*]] = getelementptr i8, ptr [[PS]], i64 0
@@ -319,19 +361,12 @@ define void @overlapping_strides(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-NEXT: [[GEP_S7:%.*]] = getelementptr i8, ptr [[GEP_S6]], i64 [[STRIDE2]]
; CHECK-NEXT: [[TMP2:%.*]] = extractelement <8 x i8> [[TMP1]], i64 0
; CHECK-NEXT: store i8 [[TMP2]], ptr [[GEP_S0]], align 1
-; CHECK-NEXT: [[TMP3:%.*]] = extractelement <8 x i8> [[TMP1]], i64 1
; CHECK-NEXT: store i8 [[TMP3]], ptr [[GEP_S1]], align 1
-; CHECK-NEXT: [[TMP4:%.*]] = extractelement <8 x i8> [[TMP1]], i64 2
; CHECK-NEXT: store i8 [[TMP4]], ptr [[GEP_S2]], align 1
-; CHECK-NEXT: [[TMP5:%.*]] = extractelement <8 x i8> [[TMP1]], i64 3
; CHECK-NEXT: store i8 [[TMP5]], ptr [[GEP_S3]], align 1
-; CHECK-NEXT: [[TMP6:%.*]] = extractelement <8 x i8> [[TMP1]], i64 4
; CHECK-NEXT: store i8 [[TMP6]], ptr [[GEP_S4]], align 1
-; CHECK-NEXT: [[TMP7:%.*]] = extractelement <8 x i8> [[TMP1]], i64 5
; CHECK-NEXT: store i8 [[TMP7]], ptr [[GEP_S5]], align 1
-; CHECK-NEXT: [[TMP8:%.*]] = extractelement <8 x i8> [[TMP1]], i64 6
; CHECK-NEXT: store i8 [[TMP8]], ptr [[GEP_S6]], align 1
-; CHECK-NEXT: [[TMP9:%.*]] = extractelement <8 x i8> [[TMP1]], i64 7
; CHECK-NEXT: store i8 [[TMP9]], ptr [[GEP_S7]], align 1
; CHECK-NEXT: ret void
;
@@ -380,9 +415,15 @@ define void @runtime_stride_unit_stride(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-LABEL: define void @runtime_stride_unit_stride(
; CHECK-SAME: ptr [[PL:%.*]], ptr [[PS:%.*]], i64 [[STRIDE:%.*]]) #[[ATTR0]] {
; CHECK-NEXT: [[GEP_L0:%.*]] = getelementptr i8, ptr [[PL]], i64 0
+; CHECK-NEXT: [[GEP_L1:%.*]] = getelementptr i8, ptr [[PL]], i64 1
+; CHECK-NEXT: [[GEP_L2:%.*]] = getelementptr i8, ptr [[PL]], i64 2
; CHECK-NEXT: [[GEP_L3:%.*]] = getelementptr i8, ptr [[PL]], i64 3
+; CHECK-NEXT: [[GEP_L7:%.*]] = getelementptr i8, ptr [[PL]], i64 7
+; CHECK-NEXT: [[LOAD7:%.*]] = load i8, ptr [[GEP_L7]], align 1
; CHECK-NEXT: [[TMP3:%.*]] = call <8 x i8> @llvm.masked.load.v8i8.p0(ptr align 1 [[GEP_L0]], <8 x i1> <i1 true, i1 true, i1 true, i1 false, i1 false, i1 false, i1 false, i1 true>, <8 x i8> poison)
; CHECK-NEXT: [[TMP2:%.*]] = shufflevector <8 x i8> [[TMP3]], <8 x i8> poison, <4 x i32> <i32 0, i32 1, i32 2, i32 7>
+; CHECK-NEXT: [[LOAD2:%.*]] = load i8, ptr [[GEP_L2]], align 1
+; CHECK-NEXT: [[TMP5:%.*]] = load i8, ptr [[GEP_L1]], align 1
; CHECK-NEXT: [[GEP_S0:%.*]] = getelementptr i8, ptr [[PS]], i64 0
; CHECK-NEXT: [[GEP_S1:%.*]] = getelementptr i8, ptr [[GEP_S0]], i64 [[STRIDE]]
; CHECK-NEXT: [[GEP_S2:%.*]] = getelementptr i8, ptr [[GEP_S1]], i64 [[STRIDE]]
@@ -394,12 +435,9 @@ define void @runtime_stride_unit_stride(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i8>, ptr [[GEP_L3]], align 1
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <4 x i8> [[TMP2]], i64 0
; CHECK-NEXT: store i8 [[TMP4]], ptr [[GEP_S0]], align 1
-; CHECK-NEXT: [[TMP5:%.*]] = extractelement <4 x i8> [[TMP2]], i64 1
; CHECK-NEXT: store i8 [[TMP5]], ptr [[GEP_S1]], align 1
-; CHECK-NEXT: [[LOAD2:%.*]] = extractelement <4 x i8> [[TMP2]], i64 2
; CHECK-NEXT: store i8 [[LOAD2]], ptr [[GEP_S2]], align 1
; CHECK-NEXT: store <4 x i8> [[TMP1]], ptr [[GEP_S3]], align 1
-; CHECK-NEXT: [[LOAD7:%.*]] = extractelement <4 x i8> [[TMP2]], i64 3
; CHECK-NEXT: store i8 [[LOAD7]], ptr [[GEP_S7]], align 1
; CHECK-NEXT: ret void
;
@@ -447,7 +485,13 @@ define void @unit_stride_runtime_stride(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-SAME: ptr [[PL:%.*]], ptr [[PS:%.*]], i64 [[STRIDE:%.*]]) #[[ATTR0]] {
; CHECK-NEXT: [[GEP_L0:%.*]] = getelementptr i8, ptr [[PL]], i64 0
; CHECK-NEXT: [[GEP_L4:%.*]] = getelementptr i8, ptr [[PL]], i64 4
+; CHECK-NEXT: [[GEP_L5:%.*]] = getelementptr i8, ptr [[PL]], i64 5
+; CHECK-NEXT: [[GEP_L6:%.*]] = getelementptr i8, ptr [[PL]], i64 6
+; CHECK-NEXT: [[GEP_L7:%.*]] = getelementptr i8, ptr [[PL]], i64 7
+; CHECK-NEXT: [[TMP6:%.*]] = load i8, ptr [[GEP_L7]], align 1
; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i8>, ptr [[GEP_L4]], align 1
+; CHECK-NEXT: [[TMP5:%.*]] = load i8, ptr [[GEP_L6]], align 1
+; CHECK-NEXT: [[TMP4:%.*]] = load i8, ptr [[GEP_L5]], align 1
; CHECK-NEXT: [[GEP_S0:%.*]] = getelementptr i8, ptr [[PS]], i64 0
; CHECK-NEXT: [[GEP_S1:%.*]] = getelementptr i8, ptr [[GEP_S0]], i64 1
; CHECK-NEXT: [[GEP_S2:%.*]] = getelementptr i8, ptr [[GEP_S1]], i64 1
@@ -460,11 +504,8 @@ define void @unit_stride_runtime_stride(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-NEXT: store <4 x i8> [[TMP1]], ptr [[GEP_S0]], align 1
; CHECK-NEXT: [[TMP3:%.*]] = extractelement <4 x i8> [[TMP2]], i64 0
; CHECK-NEXT: store i8 [[TMP3]], ptr [[GEP_S4]], align 1
-; CHECK-NEXT: [[TMP4:%.*]] = extractelement <4 x i8> [[TMP2]], i64 1
; CHECK-NEXT: store i8 [[TMP4]], ptr [[GEP_S5]], align 1
-; CHECK-NEXT: [[TMP5:%.*]] = extractelement <4 x i8> [[TMP2]], i64 2
; CHECK-NEXT: store i8 [[TMP5]], ptr [[GEP_S6]], align 1
-; CHECK-NEXT: [[TMP6:%.*]] = extractelement <4 x i8> [[TMP2]], i64 3
; CHECK-NEXT: store i8 [[TMP6]], ptr [[GEP_S7]], align 1
; CHECK-NEXT: ret void
;
@@ -511,8 +552,14 @@ define void @runtime_stride_constant_stride(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-LABEL: define void @runtime_stride_constant_stride(
; CHECK-SAME: ptr [[PL:%.*]], ptr [[PS:%.*]], i64 [[STRIDE:%.*]]) #[[ATTR0]] {
; CHECK-NEXT: [[GEP_L0:%.*]] = getelementptr i8, ptr [[PL]], i64 0
+; CHECK-NEXT: [[GEP_L1:%.*]] = getelementptr i8, ptr [[PL]], i64 1
+; CHECK-NEXT: [[GEP_L2:%.*]] = getelementptr i8, ptr [[PL]], i64 2
+; CHECK-NEXT: [[GEP_L3:%.*]] = getelementptr i8, ptr [[PL]], i64 3
; CHECK-NEXT: [[GEP_L4:%.*]] = getelementptr i8, ptr [[PL]], i64 4
+; CHECK-NEXT: [[TMP6:%.*]] = load i8, ptr [[GEP_L3]], align 1
; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i8>, ptr [[GEP_L0]], align 1
+; CHECK-NEXT: [[TMP5:%.*]] = load i8, ptr [[GEP_L2]], align 1
+; CHECK-NEXT: [[TMP4:%.*]] = load i8, ptr [[GEP_L1]], align 1
; CHECK-NEXT: [[GEP_S0:%.*]] = getelementptr i8, ptr [[PS]], i64 0
; CHECK-NEXT: [[GEP_S1:%.*]] = getelementptr i8, ptr [[GEP_S0]], i64 [[STRIDE]]
; CHECK-NEXT: [[GEP_S2:%.*]] = getelementptr i8, ptr [[GEP_S1]], i64 [[STRIDE]]
@@ -521,11 +568,8 @@ define void @runtime_stride_constant_stride(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-NEXT: [[TMP3:%.*]] = load <4 x i8>, ptr [[GEP_L4]], align 1
; CHECK-NEXT: [[TMP7:%.*]] = extractelement <4 x i8> [[TMP1]], i64 0
; CHECK-NEXT: store i8 [[TMP7]], ptr [[GEP_S0]], align 1
-; CHECK-NEXT: [[TMP4:%.*]] = extractelement <4 x i8> [[TMP1]], i64 1
; CHECK-NEXT: store i8 [[TMP4]], ptr [[GEP_S1]], align 1
-; CHECK-NEXT: [[TMP5:%.*]] = extractelement <4 x i8> [[TMP1]], i64 2
; CHECK-NEXT: store i8 [[TMP5]], ptr [[GEP_S2]], align 1
-; CHECK-NEXT: [[TMP6:%.*]] = extractelement <4 x i8> [[TMP1]], i64 3
; CHECK-NEXT: store i8 [[TMP6]], ptr [[GEP_S3]], align 1
; CHECK-NEXT: call void @llvm.experimental.vp.strided.store.v4i8.p0.i64(<4 x i8> [[TMP3]], ptr align 1 [[GEP_S4]], i64 2, <4 x i1> splat (i1 true), i32 4)
; CHECK-NEXT: ret void
@@ -574,7 +618,13 @@ define void @constant_stride_runtime_stride(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-SAME: ptr [[PL:%.*]], ptr [[PS:%.*]], i64 [[STRIDE:%.*]]) #[[ATTR0]] {
; CHECK-NEXT: [[GEP_L0:%.*]] = getelementptr i8, ptr [[PL]], i64 0
; CHECK-NEXT: [[GEP_L4:%.*]] = getelementptr i8, ptr [[PL]], i64 4
+; CHECK-NEXT: [[GEP_L5:%.*]] = getelementptr i8, ptr [[PL]], i64 5
+; CHECK-NEXT: [[GEP_L6:%.*]] = getelementptr i8, ptr [[PL]], i64 6
+; CHECK-NEXT: [[GEP_L7:%.*]] = getelementptr i8, ptr [[PL]], i64 7
+; CHECK-NEXT: [[TMP6:%.*]] = load i8, ptr [[GEP_L7]], align 1
; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i8>, ptr [[GEP_L4]], align 1
+; CHECK-NEXT: [[TMP5:%.*]] = load i8, ptr [[GEP_L6]], align 1
+; CHECK-NEXT: [[TMP4:%.*]] = load i8, ptr [[GEP_L5]], align 1
; CHECK-NEXT: [[GEP_S0:%.*]] = getelementptr i8, ptr [[PS]], i64 0
; CHECK-NEXT: [[GEP_S1:%.*]] = getelementptr i8, ptr [[GEP_S0]], i64 2
; CHECK-NEXT: [[GEP_S2:%.*]] = getelementptr i8, ptr [[GEP_S1]], i64 2
@@ -587,11 +637,8 @@ define void @constant_stride_runtime_stride(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-NEXT: call void @llvm.experimental.vp.strided.store.v4i8.p0.i64(<4 x i8> [[TMP1]], ptr align 1 [[GEP_S0]], i64 2, <4 x i1> splat (i1 true), i32 4)
; CHECK-NEXT: [[TMP3:%.*]] = extractelement <4 x i8> [[TMP2]], i64 0
; CHECK-NEXT: store i8 [[TMP3]], ptr [[GEP_S4]], align 1
-; CHECK-NEXT: [[TMP4:%.*]] = extractelement <4 x i8> [[TMP2]], i64 1
; CHECK-NEXT: store i8 [[TMP4]], ptr [[GEP_S5]], align 1
-; CHECK-NEXT: [[TMP5:%.*]] = extractelement <4 x i8> [[TMP2]], i64 2
; CHECK-NEXT: store i8 [[TMP5]], ptr [[GEP_S6]], align 1
-; CHECK-NEXT: [[TMP6:%.*]] = extractelement <4 x i8> [[TMP2]], i64 3
; CHECK-NEXT: store i8 [[TMP6]], ptr [[GEP_S7]], align 1
; CHECK-NEXT: ret void
;
@@ -638,8 +685,10 @@ define void @overlap(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-LABEL: define void @overlap(
; CHECK-SAME: ptr [[PL:%.*]], ptr [[PS:%.*]], i64 [[STRIDE:%.*]]) #[[ATTR0]] {
; CHECK-NEXT: [[GEP_L0:%.*]] = getelementptr i8, ptr [[PL]], i64 0
+; CHECK-NEXT: [[GEP_L1:%.*]] = getelementptr i8, ptr [[PL]], i64 1
; CHECK-NEXT: [[GEP_L2:%.*]] = getelementptr i8, ptr [[PL]], i64 2
; CHECK-NEXT: [[GEP_L3:%.*]] = getelementptr i8, ptr [[PL]], i64 3
+; CHECK-NEXT: [[TMP4:%.*]] = load i8, ptr [[GEP_L1]], align 1
; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i8>, ptr [[GEP_L0]], align 1
; CHECK-NEXT: [[LOAD2:%.*]] = load i8, ptr [[GEP_L2]], align 1
; CHECK-NEXT: [[GEP_S0:%.*]] = getelementptr i8, ptr [[PS]], i64 0
@@ -649,7 +698,6 @@ define void @overlap(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i8>, ptr [[GEP_L3]], align 1
; CHECK-NEXT: [[TMP3:%.*]] = extractelement <2 x i8> [[TMP2]], i64 0
; CHECK-NEXT: store i8 [[TMP3]], ptr [[GEP_S0]], align 1
-; CHECK-NEXT: [[TMP4:%.*]] = extractelement <2 x i8> [[TMP2]], i64 1
; CHECK-NEXT: store i8 [[TMP4]], ptr [[GEP_S1]], align 1
; CHECK-NEXT: store i8 [[LOAD2]], ptr [[GEP_S2]], align 1
; CHECK-NEXT: store <4 x i8> [[TMP1]], ptr [[GEP_S3]], align 1
@@ -694,8 +742,10 @@ define void @overlap_constant_stride(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-LABEL: define void @overlap_constant_stride(
; CHECK-SAME: ptr [[PL:%.*]], ptr [[PS:%.*]], i64 [[STRIDE:%.*]]) #[[ATTR0]] {
; CHECK-NEXT: [[GEP_L0:%.*]] = getelementptr i8, ptr [[PL]], i64 0
+; CHECK-NEXT: [[GEP_L1:%.*]] = getelementptr i8, ptr [[PL]], i64 1
; CHECK-NEXT: [[GEP_L2:%.*]] = getelementptr i8, ptr [[PL]], i64 2
; CHECK-NEXT: [[GEP_L3:%.*]] = getelementptr i8, ptr [[PL]], i64 3
+; CHECK-NEXT: [[TMP4:%.*]] = load i8, ptr [[GEP_L1]], align 1
; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i8>, ptr [[GEP_L0]], align 1
; CHECK-NEXT: [[LOAD2:%.*]] = load i8, ptr [[GEP_L2]], align 1
; CHECK-NEXT: [[GEP_S0:%.*]] = getelementptr i8, ptr [[PS]], i64 0
@@ -705,7 +755,6 @@ define void @overlap_constant_stride(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i8>, ptr [[GEP_L3]], align 1
; CHECK-NEXT: [[TMP3:%.*]] = extractelement <2 x i8> [[TMP2]], i64 0
; CHECK-NEXT: store i8 [[TMP3]], ptr [[GEP_S0]], align 1
-; CHECK-NEXT: [[TMP4:%.*]] = extractelement <2 x i8> [[TMP2]], i64 1
; CHECK-NEXT: store i8 [[TMP4]], ptr [[GEP_S1]], align 1
; CHECK-NEXT: store i8 [[LOAD2]], ptr [[GEP_S2]], align 1
; CHECK-NEXT: call void @llvm.experimental.vp.strided.store.v4i8.p0.i64(<4 x i8> [[TMP1]], ptr align 1 [[GEP_S3]], i64 2, <4 x i1> splat (i1 true), i32 4)
@@ -750,7 +799,21 @@ define void @runtime_stride_reorder(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-LABEL: define void @runtime_stride_reorder(
; CHECK-SAME: ptr [[PL:%.*]], ptr [[PS:%.*]], i64 [[STRIDE:%.*]]) #[[ATTR0]] {
; CHECK-NEXT: [[GEP_L0:%.*]] = getelementptr i8, ptr [[PL]], i64 0
+; CHECK-NEXT: [[GEP_L1:%.*]] = getelementptr i8, ptr [[PL]], i64 2
+; CHECK-NEXT: [[GEP_L2:%.*]] = getelementptr i8, ptr [[PL]], i64 1
+; CHECK-NEXT: [[GEP_L3:%.*]] = getelementptr i8, ptr [[PL]], i64 4
+; CHECK-NEXT: [[GEP_L4:%.*]] = getelementptr i8, ptr [[PL]], i64 7
+; CHECK-NEXT: [[GEP_L5:%.*]] = getelementptr i8, ptr [[PL]], i64 5
+; CHECK-NEXT: [[GEP_L6:%.*]] = getelementptr i8, ptr [[PL]], i64 6
+; CHECK-NEXT: [[GEP_L7:%.*]] = getelementptr i8, ptr [[PL]], i64 3
+; CHECK-NEXT: [[TMP9:%.*]] = load i8, ptr [[GEP_L7]], align 1
; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i8>, ptr [[GEP_L0]], align 1
+; CHECK-NEXT: [[TMP8:%.*]] = load i8, ptr [[GEP_L6]], align 1
+; CHECK-NEXT: [[TMP7:%.*]] = load i8, ptr [[GEP_L5]], align 1
+; CHECK-NEXT: [[TMP6:%.*]] = load i8, ptr [[GEP_L4]], align 1
+; CHECK-NEXT: [[TMP5:%.*]] = load i8, ptr [[GEP_L3]], align 1
+; CHECK-NEXT: [[TMP4:%.*]] = load i8, ptr [[GEP_L2]], align 1
+; CHECK-NEXT: [[TMP3:%.*]] = load i8, ptr [[GEP_L1]], align 1
; CHECK-NEXT: [[GEP_S0:%.*]] = getelementptr i8, ptr [[PS]], i64 0
; CHECK-NEXT: [[GEP_S1:%.*]] = getelementptr i8, ptr [[GEP_S0]], i64 [[STRIDE]]
; CHECK-NEXT: [[GEP_S2:%.*]] = getelementptr i8, ptr [[GEP_S1]], i64 [[STRIDE]]
@@ -761,19 +824,12 @@ define void @runtime_stride_reorder(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-NEXT: [[GEP_S7:%.*]] = getelementptr i8, ptr [[GEP_S6]], i64 [[STRIDE]]
; CHECK-NEXT: [[TMP2:%.*]] = extractelement <8 x i8> [[TMP1]], i64 0
; CHECK-NEXT: store i8 [[TMP2]], ptr [[GEP_S0]], align 1
-; CHECK-NEXT: [[TMP3:%.*]] = extractelement <8 x i8> [[TMP1]], i64 2
; CHECK-NEXT: store i8 [[TMP3]], ptr [[GEP_S1]], align 1
-; CHECK-NEXT: [[TMP4:%.*]] = extractelement <8 x i8> [[TMP1]], i64 1
; CHECK-NEXT: store i8 [[TMP4]], ptr [[GEP_S2]], align 1
-; CHECK-NEXT: [[TMP5:%.*]] = extractelement <8 x i8> [[TMP1]], i64 4
; CHECK-NEXT: store i8 [[TMP5]], ptr [[GEP_S3]], align 1
-; CHECK-NEXT: [[TMP6:%.*]] = extractelement <8 x i8> [[TMP1]], i64 7
; CHECK-NEXT: store i8 [[TMP6]], ptr [[GEP_S4]], align 1
-; CHECK-NEXT: [[TMP7:%.*]] = extractelement <8 x i8> [[TMP1]], i64 5
; CHECK-NEXT: store i8 [[TMP7]], ptr [[GEP_S5]], align 1
-; CHECK-NEXT: [[TMP8:%.*]] = extractelement <8 x i8> [[TMP1]], i64 6
; CHECK-NEXT: store i8 [[TMP8]], ptr [[GEP_S6]], align 1
-; CHECK-NEXT: [[TMP9:%.*]] = extractelement <8 x i8> [[TMP1]], i64 3
; CHECK-NEXT: store i8 [[TMP9]], ptr [[GEP_S7]], align 1
; CHECK-NEXT: ret void
;
@@ -820,7 +876,21 @@ define void @runtime_stride_unschedulable(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-LABEL: define void @runtime_stride_unschedulable(
; CHECK-SAME: ptr [[PL:%.*]], ptr [[PS:%.*]], i64 [[STRIDE:%.*]]) #[[ATTR0]] {
; CHECK-NEXT: [[GEP_L0:%.*]] = getelementptr i8, ptr [[PL]], i64 0
+; CHECK-NEXT: [[GEP_L1:%.*]] = getelementptr i8, ptr [[PL]], i64 1
+; CHECK-NEXT: [[GEP_L2:%.*]] = getelementptr i8, ptr [[PL]], i64 2
+; CHECK-NEXT: [[GEP_L3:%.*]] = getelementptr i8, ptr [[PL]], i64 3
+; CHECK-NEXT: [[GEP_L4:%.*]] = getelementptr i8, ptr [[PL]], i64 4
+; CHECK-NEXT: [[GEP_L5:%.*]] = getelementptr i8, ptr [[PL]], i64 5
+; CHECK-NEXT: [[GEP_L6:%.*]] = getelementptr i8, ptr [[PL]], i64 6
+; CHECK-NEXT: [[GEP_L7:%.*]] = getelementptr i8, ptr [[PL]], i64 7
+; CHECK-NEXT: [[TMP8:%.*]] = load i8, ptr [[GEP_L7]], align 1
; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i8>, ptr [[GEP_L0]], align 1
+; CHECK-NEXT: [[TMP7:%.*]] = load i8, ptr [[GEP_L6]], align 1
+; CHECK-NEXT: [[TMP11:%.*]] = load i8, ptr [[GEP_L5]], align 1
+; CHECK-NEXT: [[TMP6:%.*]] = load i8, ptr [[GEP_L4]], align 1
+; CHECK-NEXT: [[TMP10:%.*]] = load i8, ptr [[GEP_L3]], align 1
+; CHECK-NEXT: [[TMP9:%.*]] = load i8, ptr [[GEP_L2]], align 1
+; CHECK-NEXT: [[TMP5:%.*]] = load i8, ptr [[GEP_L1]], align 1
; CHECK-NEXT: [[GEP_S0:%.*]] = getelementptr i8, ptr [[PS]], i64 0
; CHECK-NEXT: [[GEP_S1:%.*]] = getelementptr i8, ptr [[GEP_S0]], i64 [[STRIDE]]
; CHECK-NEXT: [[GEP_S2:%.*]] = getelementptr i8, ptr [[GEP_S1]], i64 [[STRIDE]]
@@ -831,19 +901,12 @@ define void @runtime_stride_unschedulable(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-NEXT: [[GEP_S7:%.*]] = getelementptr i8, ptr [[GEP_S6]], i64 [[STRIDE]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <8 x i8> [[TMP1]], i64 0
; CHECK-NEXT: store i8 [[TMP4]], ptr [[GEP_S0]], align 1
-; CHECK-NEXT: [[TMP5:%.*]] = extractelement <8 x i8> [[TMP1]], i64 1
; CHECK-NEXT: store i8 [[TMP5]], ptr [[GEP_S6]], align 1
-; CHECK-NEXT: [[TMP9:%.*]] = extractelement <8 x i8> [[TMP1]], i64 2
; CHECK-NEXT: store i8 [[TMP9]], ptr [[GEP_S2]], align 1
-; CHECK-NEXT: [[TMP10:%.*]] = extractelement <8 x i8> [[TMP1]], i64 3
; CHECK-NEXT: store i8 [[TMP10]], ptr [[GEP_S3]], align 1
-; CHECK-NEXT: [[TMP6:%.*]] = extractelement <8 x i8> [[TMP1]], i64 4
; CHECK-NEXT: store i8 [[TMP6]], ptr [[GEP_S4]], align 1
-; CHECK-NEXT: [[TMP11:%.*]] = extractelement <8 x i8> [[TMP1]], i64 5
; CHECK-NEXT: store i8 [[TMP11]], ptr [[GEP_S5]], align 1
-; CHECK-NEXT: [[TMP7:%.*]] = extractelement <8 x i8> [[TMP1]], i64 6
; CHECK-NEXT: store i8 [[TMP7]], ptr [[GEP_S1]], align 1
-; CHECK-NEXT: [[TMP8:%.*]] = extractelement <8 x i8> [[TMP1]], i64 7
; CHECK-NEXT: store i8 [[TMP8]], ptr [[GEP_S7]], align 1
; CHECK-NEXT: ret void
;
@@ -919,10 +982,16 @@ define void @runtime_stride_diff_types(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-SAME: ptr [[PL:%.*]], ptr [[PS:%.*]], i64 [[STRIDE:%.*]]) #[[ATTR0]] {
; CHECK-NEXT: [[GEP_L0:%.*]] = getelementptr i8, ptr [[PL]], i64 0
; CHECK-NEXT: [[GEP_L1:%.*]] = getelementptr i8, ptr [[PL]], i64 1
+; CHECK-NEXT: [[GEP_L2:%.*]] = getelementptr i8, ptr [[PL]], i64 2
+; CHECK-NEXT: [[GEP_L3:%.*]] = getelementptr i8, ptr [[PL]], i64 3
+; CHECK-NEXT: [[GEP_L4:%.*]] = getelementptr i8, ptr [[PL]], i64 4
; CHECK-NEXT: [[GEP_L6:%.*]] = getelementptr i8, ptr [[PL]], i64 6
; CHECK-NEXT: [[GEP_L7:%.*]] = getelementptr i8, ptr [[PL]], i64 7
+; CHECK-NEXT: [[TMP9:%.*]] = load i8, ptr [[GEP_L4]], align 1
; CHECK-NEXT: [[TMP1:%.*]] = call <5 x i8> @llvm.masked.load.v5i8.p0(ptr align 1 [[GEP_L0]], <5 x i1> <i1 true, i1 false, i1 true, i1 true, i1 true>, <5 x i8> poison)
; CHECK-NEXT: [[TMP2:%.*]] = shufflevector <5 x i8> [[TMP1]], <5 x i8> poison, <4 x i32> <i32 0, i32 2, i32 3, i32 4>
+; CHECK-NEXT: [[TMP8:%.*]] = load i8, ptr [[GEP_L3]], align 1
+; CHECK-NEXT: [[TMP7:%.*]] = load i8, ptr [[GEP_L2]], align 1
; CHECK-NEXT: [[LOAD6:%.*]] = load i8, ptr [[GEP_L6]], align 1
; CHECK-NEXT: [[LOAD7:%.*]] = load i8, ptr [[GEP_L7]], align 1
; CHECK-NEXT: [[TMP3:%.*]] = call <2 x i8> @llvm.experimental.vp.strided.load.v2i8.p0.i64(ptr align 1 [[GEP_L1]], i64 4, <2 x i1> splat (i1 true), i32 2)
@@ -936,9 +1005,6 @@ define void @runtime_stride_diff_types(ptr %pl, ptr %ps, i64 %stride) {
; CHECK-NEXT: [[GEP_S7:%.*]] = getelementptr i8, ptr [[GEP_S6]], i64 [[STRIDE]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <4 x i8> [[TMP2]], i64 0
; CHECK-NEXT: store i8 [[TMP4]], ptr [[GEP_S0]], align 1
-; CHECK-NEXT: [[TMP7:%.*]] = extractelement <4 x i8> [[TMP2]], i64 1
-; CHECK-NEXT: [[TMP8:%.*]] = extractelement <4 x i8> [[TMP2]], i64 2
-; CHECK-NEXT: [[TMP9:%.*]] = extractelement <4 x i8> [[TMP2]], i64 3
; CHECK-NEXT: [[TMP5:%.*]] = extractelement <2 x i8> [[TMP3]], i64 0
; CHECK-NEXT: [[TMP6:%.*]] = zext i8 [[TMP5]] to i16
; CHECK-NEXT: store i16 [[TMP6]], ptr [[GEP_S1]], align 2
diff --git a/llvm/test/Transforms/SLPVectorizer/RISCV/strided-loads-with-external-use-ptr.ll b/llvm/test/Transforms/SLPVectorizer/RISCV/strided-loads-with-external-use-ptr.ll
index 7e20cb383384d..b116bdebdd8c5 100644
--- a/llvm/test/Transforms/SLPVectorizer/RISCV/strided-loads-with-external-use-ptr.ll
+++ b/llvm/test/Transforms/SLPVectorizer/RISCV/strided-loads-with-external-use-ptr.ll
@@ -17,9 +17,9 @@ define i16 @test() {
; CHECK-NEXT: [[PPREV_0_I]] = getelementptr [[S]], ptr [[PPREV_062_I]], i64 -1
; CHECK-NEXT: [[TMP1:%.*]] = call <3 x i16> @llvm.masked.load.v3i16.p0(ptr align 2 [[PPREV_0_I]], <3 x i1> <i1 true, i1 false, i1 true>, <3 x i16> poison)
; CHECK-NEXT: [[TMP2:%.*]] = shufflevector <3 x i16> [[TMP1]], <3 x i16> poison, <2 x i32> <i32 0, i32 2>
-; CHECK-NEXT: [[TMP3:%.*]] = extractelement <2 x i16> [[TMP2]], i64 0
-; CHECK-NEXT: [[TMP4:%.*]] = extractelement <2 x i16> [[TMP2]], i64 1
-; CHECK-NEXT: [[CMP_I178:%.*]] = icmp ult i16 [[TMP4]], [[TMP3]]
+; CHECK-NEXT: [[TMP3:%.*]] = load i16, ptr [[INCDEC_PTR_I]], align 2
+; CHECK-NEXT: [[TMP4:%.*]] = extractelement <2 x i16> [[TMP2]], i64 0
+; CHECK-NEXT: [[CMP_I178:%.*]] = icmp ult i16 [[TMP3]], [[TMP4]]
; CHECK-NEXT: br label [[WHILE_BODY_I]]
;
entry:
diff --git a/llvm/test/Transforms/SLPVectorizer/RISCV/vec3-base.ll b/llvm/test/Transforms/SLPVectorizer/RISCV/vec3-base.ll
index 9a284df074cb4..d9a2a495cd55d 100644
--- a/llvm/test/Transforms/SLPVectorizer/RISCV/vec3-base.ll
+++ b/llvm/test/Transforms/SLPVectorizer/RISCV/vec3-base.ll
@@ -508,18 +508,23 @@ define i32 @dot_product_i32(ptr %a, ptr %b) {
;
; POW2-ONLY-LABEL: @dot_product_i32(
; POW2-ONLY-NEXT: [[GEP_A_0:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i32 0
+; POW2-ONLY-NEXT: [[L_A_0:%.*]] = load i32, ptr [[GEP_A_0]], align 4
+; POW2-ONLY-NEXT: [[GEP_A_1:%.*]] = getelementptr inbounds i32, ptr [[A]], i32 1
+; POW2-ONLY-NEXT: [[L_A_1:%.*]] = load i32, ptr [[GEP_A_1]], align 4
; POW2-ONLY-NEXT: [[GEP_A_2:%.*]] = getelementptr inbounds i32, ptr [[A]], i32 2
; POW2-ONLY-NEXT: [[L_A_2:%.*]] = load i32, ptr [[GEP_A_2]], align 4
; POW2-ONLY-NEXT: [[GEP_B_0:%.*]] = getelementptr inbounds i32, ptr [[B:%.*]], i32 0
+; POW2-ONLY-NEXT: [[L_B_0:%.*]] = load i32, ptr [[GEP_B_0]], align 4
+; POW2-ONLY-NEXT: [[GEP_B_1:%.*]] = getelementptr inbounds i32, ptr [[B]], i32 1
+; POW2-ONLY-NEXT: [[L_B_1:%.*]] = load i32, ptr [[GEP_B_1]], align 4
; POW2-ONLY-NEXT: [[GEP_B_2:%.*]] = getelementptr inbounds i32, ptr [[B]], i32 2
; POW2-ONLY-NEXT: [[L_B_2:%.*]] = load i32, ptr [[GEP_B_2]], align 4
-; POW2-ONLY-NEXT: [[TMP1:%.*]] = load <2 x i32>, ptr [[GEP_A_0]], align 4
-; POW2-ONLY-NEXT: [[TMP2:%.*]] = load <2 x i32>, ptr [[GEP_B_0]], align 4
-; POW2-ONLY-NEXT: [[TMP3:%.*]] = mul nsw <2 x i32> [[TMP1]], [[TMP2]]
-; POW2-ONLY-NEXT: [[MUL_2:%.*]] = mul nsw i32 [[L_A_2]], [[L_B_2]]
-; POW2-ONLY-NEXT: [[ADD_0:%.*]] = call i32 @llvm.vector.reduce.add.v2i32(<2 x i32> [[TMP3]])
+; POW2-ONLY-NEXT: [[ADD_0:%.*]] = mul nsw i32 [[L_A_0]], [[L_B_0]]
+; POW2-ONLY-NEXT: [[MUL_2:%.*]] = mul nsw i32 [[L_A_1]], [[L_B_1]]
+; POW2-ONLY-NEXT: [[MUL_3:%.*]] = mul nsw i32 [[L_A_2]], [[L_B_2]]
; POW2-ONLY-NEXT: [[ADD_1:%.*]] = add i32 [[ADD_0]], [[MUL_2]]
-; POW2-ONLY-NEXT: ret i32 [[ADD_1]]
+; POW2-ONLY-NEXT: [[ADD_2:%.*]] = add i32 [[ADD_1]], [[MUL_3]]
+; POW2-ONLY-NEXT: ret i32 [[ADD_2]]
;
%gep.a.0 = getelementptr inbounds i32, ptr %a, i32 0
%l.a.0 = load i32, ptr %gep.a.0, align 4
@@ -558,18 +563,23 @@ define i32 @dot_product_i32_reorder(ptr %a, ptr %b) {
;
; POW2-ONLY-LABEL: @dot_product_i32_reorder(
; POW2-ONLY-NEXT: [[GEP_A_0:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i32 0
+; POW2-ONLY-NEXT: [[L_A_0:%.*]] = load i32, ptr [[GEP_A_0]], align 4
+; POW2-ONLY-NEXT: [[GEP_A_1:%.*]] = getelementptr inbounds i32, ptr [[A]], i32 1
+; POW2-ONLY-NEXT: [[L_A_1:%.*]] = load i32, ptr [[GEP_A_1]], align 4
; POW2-ONLY-NEXT: [[GEP_A_2:%.*]] = getelementptr inbounds i32, ptr [[A]], i32 2
; POW2-ONLY-NEXT: [[L_A_2:%.*]] = load i32, ptr [[GEP_A_2]], align 4
; POW2-ONLY-NEXT: [[GEP_B_0:%.*]] = getelementptr inbounds i32, ptr [[B:%.*]], i32 0
+; POW2-ONLY-NEXT: [[L_B_0:%.*]] = load i32, ptr [[GEP_B_0]], align 4
+; POW2-ONLY-NEXT: [[GEP_B_1:%.*]] = getelementptr inbounds i32, ptr [[B]], i32 1
+; POW2-ONLY-NEXT: [[L_B_1:%.*]] = load i32, ptr [[GEP_B_1]], align 4
; POW2-ONLY-NEXT: [[GEP_B_2:%.*]] = getelementptr inbounds i32, ptr [[B]], i32 2
; POW2-ONLY-NEXT: [[L_B_2:%.*]] = load i32, ptr [[GEP_B_2]], align 4
-; POW2-ONLY-NEXT: [[TMP1:%.*]] = load <2 x i32>, ptr [[GEP_A_0]], align 4
-; POW2-ONLY-NEXT: [[TMP2:%.*]] = load <2 x i32>, ptr [[GEP_B_0]], align 4
-; POW2-ONLY-NEXT: [[TMP3:%.*]] = mul nsw <2 x i32> [[TMP1]], [[TMP2]]
-; POW2-ONLY-NEXT: [[MUL_2:%.*]] = mul nsw i32 [[L_A_2]], [[L_B_2]]
-; POW2-ONLY-NEXT: [[ADD_0:%.*]] = call i32 @llvm.vector.reduce.add.v2i32(<2 x i32> [[TMP3]])
+; POW2-ONLY-NEXT: [[MUL_2:%.*]] = mul nsw i32 [[L_A_0]], [[L_B_0]]
+; POW2-ONLY-NEXT: [[ADD_0:%.*]] = mul nsw i32 [[L_A_1]], [[L_B_1]]
+; POW2-ONLY-NEXT: [[MUL_3:%.*]] = mul nsw i32 [[L_A_2]], [[L_B_2]]
; POW2-ONLY-NEXT: [[ADD_1:%.*]] = add i32 [[ADD_0]], [[MUL_2]]
-; POW2-ONLY-NEXT: ret i32 [[ADD_1]]
+; POW2-ONLY-NEXT: [[ADD_2:%.*]] = add i32 [[ADD_1]], [[MUL_3]]
+; POW2-ONLY-NEXT: ret i32 [[ADD_2]]
;
%gep.a.0 = getelementptr inbounds i32, ptr %a, i32 0
%l.a.0 = load i32, ptr %gep.a.0, align 4
diff --git a/llvm/test/Transforms/VectorCombine/RISCV/fold-equivalent-reduction-cmp.ll b/llvm/test/Transforms/VectorCombine/RISCV/fold-equivalent-reduction-cmp.ll
index 9de5b9c07ba9d..2b219f8d4f2e8 100644
--- a/llvm/test/Transforms/VectorCombine/RISCV/fold-equivalent-reduction-cmp.ll
+++ b/llvm/test/Transforms/VectorCombine/RISCV/fold-equivalent-reduction-cmp.ll
@@ -232,8 +232,8 @@ define i1 @or_eq_0_i8(<16 x i8> %x) {
define i1 @or_eq_0_i64(<2 x i64> %x) {
; CHECK-LABEL: define i1 @or_eq_0_i64(
; CHECK-SAME: <2 x i64> [[X:%.*]]) #[[ATTR0]] {
-; CHECK-NEXT: [[RED:%.*]] = call i64 @llvm.vector.reduce.or.v2i64(<2 x i64> [[X]])
-; CHECK-NEXT: [[CMP:%.*]] = icmp eq i64 [[RED]], 0
+; CHECK-NEXT: [[TMP1:%.*]] = icmp eq <2 x i64> [[X]], zeroinitializer
+; CHECK-NEXT: [[CMP:%.*]] = call i1 @llvm.vector.reduce.and.v2i1(<2 x i1> [[TMP1]])
; CHECK-NEXT: ret i1 [[CMP]]
;
%red = call i64 @llvm.vector.reduce.or.v2i64(<2 x i64> %x)
More information about the llvm-commits
mailing list