[AArch64] Add -aarch64-insert-extract-base-cost

[peterwaller-arm]

Summary

The new flag -aarch64-insert-extract-base-cost can be used to
set the value of AArch64Subtarget::getVectorInsertExtractBaseCost(),
for the purposes of experimentation.

[peterwaller-arm]

• Add tests:
  • Switch to utils/update_analyze_test_checks.py
  • Increasing
  • Decreasing
  • Underflow
  • FP

[dmgreen]

I think it would be clearer if the option just override the existing value, as opposed to being an adjustment from it. Otherwise it might start to mean different things as VectorInsertExtractBaseCost changes.

[peterwaller-arm]

• Make the flag absolute, per review comment (-aarch64-insert-extract-cost-adjustment => -aarch64-insert-extract-base-cost)

Thanks for the comment, updated.

[peterwaller-arm]

In the previous patch I missed that the choice of CPU can override the default
value, and by chance this continued to work. Add a test and fix this.

[dmgreen]

Thanks. SGTM

llvm/lib/Target/AArch64/AArch64Subtarget.cpp

60–62

You can drop the {} brackets from single statement blocks.

[peterwaller-arm]

• Drop unneeded braces.
• Drop unneeded default argument.

[peterwaller-arm]

llvm/lib/Target/AArch64/AArch64Subtarget.cpp

60–62

Oops, thanks.

llvm/lib/Target/AArch64/AArch64Subtarget.h

#include "AArch64GenSubtargetInfo.inc"
   /// Return true if the CPU supports any kind of instruction fusion.
   bool hasFusion() const {
     return hasArithmeticBccFusion() || hasArithmeticCbzFusion() ||
            hasFuseAES() || hasFuseArithmeticLogic() || hasFuseCCSelect() ||
            hasFuseAdrpAdd() || hasFuseLiterals();
   }
 
   unsigned getMaxInterleaveFactor() const { return MaxInterleaveFactor; }
-  unsigned getVectorInsertExtractBaseCost() const {
-    return VectorInsertExtractBaseCost;
-  }
+  unsigned getVectorInsertExtractBaseCost() const;
   unsigned getCacheLineSize() const override { return CacheLineSize; }
   unsigned getPrefetchDistance() const override { return PrefetchDistance; }
   unsigned getMinPrefetchStride(unsigned NumMemAccesses,
                                 unsigned NumStridedMemAccesses,
                                 unsigned NumPrefetches,
                                 bool HasCall) const override {
     return MinPrefetchStride;
   }

llvm/lib/Target/AArch64/AArch64Subtarget.cpp

 static cl::opt<bool>
     UseNonLazyBind("aarch64-enable-nonlazybind",
                    cl::desc("Call nonlazybind functions via direct GOT load"),
                    cl::init(false), cl::Hidden);
 
 static cl::opt<bool> UseAA("aarch64-use-aa", cl::init(true),
                            cl::desc("Enable the use of AA during codegen."));
 
+static cl::opt<unsigned> OverrideVectorInsertExtractBaseCost(
+    "aarch64-insert-extract-base-cost",
+    cl::desc("Base cost of vector insert/extract element"), cl::Hidden);
+
+unsigned AArch64Subtarget::getVectorInsertExtractBaseCost() const {
+  if (OverrideVectorInsertExtractBaseCost.getNumOccurrences() > 0)
+    return OverrideVectorInsertExtractBaseCost;
+  return VectorInsertExtractBaseCost;
+}
+
 AArch64Subtarget &AArch64Subtarget::initializeSubtargetDependencies(
     StringRef FS, StringRef CPUString, StringRef TuneCPUString) {
   // Determine default and user-specified characteristics
 
   if (CPUString.empty())
     CPUString = "generic";
 
   if (TuneCPUString.empty())

llvm/test/Analysis/CostModel/AArch64/sve-insert-extract.ll

-; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
-; RUN: opt -passes='print<cost-model>' 2>&1 -disable-output -S < %s | FileCheck %s
+; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
+
+; RUN: opt -passes='print<cost-model>' 2>&1 -disable-output -S < %s | FileCheck --check-prefix=CHECK-DEFAULT %s
+; RUN: opt -aarch64-insert-extract-base-cost=0 -passes='print<cost-model>' 2>&1 -disable-output -S < %s | FileCheck --check-prefix=CHECK-LOW %s
+; RUN: opt -aarch64-insert-extract-base-cost=100000 -passes='print<cost-model>' 2>&1 -disable-output -S < %s | FileCheck --check-prefix=CHECK-HIGH %s
 
 target triple = "aarch64-unknown-linux-gnu"
 
 
 define void @ins_el0() #0 {
-; CHECK: Cost Model: Found an estimated cost of 0 for instruction:   %v0 = insertelement <vscale x 16 x i8> zeroinitializer, i8 0, i64 0
-; CHECK: Cost Model: Found an estimated cost of 0 for instruction:   %v1 = insertelement <vscale x 8 x i16> zeroinitializer, i16 0, i64 0
-; CHECK: Cost Model: Found an estimated cost of 0 for instruction:   %v2 = insertelement <vscale x 4 x i32> zeroinitializer, i32 0, i64 0
-; CHECK: Cost Model: Found an estimated cost of 0 for instruction:   %v3 = insertelement <vscale x 2 x i64> zeroinitializer, i64 0, i64 0
+; CHECK-DEFAULT-LABEL: 'ins_el0'
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v0 = insertelement <vscale x 16 x i8> zeroinitializer, i8 0, i64 0
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v1 = insertelement <vscale x 8 x i16> zeroinitializer, i16 0, i64 0
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v2 = insertelement <vscale x 4 x i32> zeroinitializer, i32 0, i64 0
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v3 = insertelement <vscale x 2 x i64> zeroinitializer, i64 0, i64 0
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v4 = insertelement <vscale x 4 x float> zeroinitializer, float 0.000000e+00, i64 0
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v5 = insertelement <vscale x 2 x double> zeroinitializer, double 0.000000e+00, i64 0
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
+;
+; CHECK-LOW-LABEL: 'ins_el0'
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v0 = insertelement <vscale x 16 x i8> zeroinitializer, i8 0, i64 0
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v1 = insertelement <vscale x 8 x i16> zeroinitializer, i16 0, i64 0
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v2 = insertelement <vscale x 4 x i32> zeroinitializer, i32 0, i64 0
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v3 = insertelement <vscale x 2 x i64> zeroinitializer, i64 0, i64 0
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v4 = insertelement <vscale x 4 x float> zeroinitializer, float 0.000000e+00, i64 0
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v5 = insertelement <vscale x 2 x double> zeroinitializer, double 0.000000e+00, i64 0
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
+;
+; CHECK-HIGH-LABEL: 'ins_el0'
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v0 = insertelement <vscale x 16 x i8> zeroinitializer, i8 0, i64 0
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v1 = insertelement <vscale x 8 x i16> zeroinitializer, i16 0, i64 0
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v2 = insertelement <vscale x 4 x i32> zeroinitializer, i32 0, i64 0
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v3 = insertelement <vscale x 2 x i64> zeroinitializer, i64 0, i64 0
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v4 = insertelement <vscale x 4 x float> zeroinitializer, float 0.000000e+00, i64 0
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v5 = insertelement <vscale x 2 x double> zeroinitializer, double 0.000000e+00, i64 0
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
+;
   %v0 = insertelement <vscale x 16 x i8> zeroinitializer, i8 0, i64 0
   %v1 = insertelement <vscale x 8 x i16> zeroinitializer, i16 0, i64 0
   %v2 = insertelement <vscale x 4 x i32> zeroinitializer, i32 0, i64 0
   %v3 = insertelement <vscale x 2 x i64> zeroinitializer, i64 0, i64 0
+  %v4 = insertelement <vscale x 4 x float> zeroinitializer, float 0., i64 0
+  %v5 = insertelement <vscale x 2 x double> zeroinitializer, double 0., i64 0
   ret void
 }
 
 define void @ins_el1() #0 {
-; CHECK: Cost Model: Found an estimated cost of 3 for instruction:   %v0 = insertelement <vscale x 16 x i8> zeroinitializer, i8 0, i64 1
-; CHECK: Cost Model: Found an estimated cost of 3 for instruction:   %v1 = insertelement <vscale x 8 x i16> zeroinitializer, i16 0, i64 1
-; CHECK: Cost Model: Found an estimated cost of 3 for instruction:   %v2 = insertelement <vscale x 4 x i32> zeroinitializer, i32 0, i64 1
-; CHECK: Cost Model: Found an estimated cost of 3 for instruction:   %v3 = insertelement <vscale x 2 x i64> zeroinitializer, i64 0, i64 1
+; CHECK-DEFAULT-LABEL: 'ins_el1'
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v0 = insertelement <vscale x 16 x i8> zeroinitializer, i8 0, i64 1
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v1 = insertelement <vscale x 8 x i16> zeroinitializer, i16 0, i64 1
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2 = insertelement <vscale x 4 x i32> zeroinitializer, i32 0, i64 1
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v3 = insertelement <vscale x 2 x i64> zeroinitializer, i64 0, i64 1
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v4 = insertelement <vscale x 4 x float> zeroinitializer, float 0.000000e+00, i64 1
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v5 = insertelement <vscale x 2 x double> zeroinitializer, double 0.000000e+00, i64 1
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
+;
+; CHECK-LOW-LABEL: 'ins_el1'
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v0 = insertelement <vscale x 16 x i8> zeroinitializer, i8 0, i64 1
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v1 = insertelement <vscale x 8 x i16> zeroinitializer, i16 0, i64 1
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v2 = insertelement <vscale x 4 x i32> zeroinitializer, i32 0, i64 1
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v3 = insertelement <vscale x 2 x i64> zeroinitializer, i64 0, i64 1
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v4 = insertelement <vscale x 4 x float> zeroinitializer, float 0.000000e+00, i64 1
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v5 = insertelement <vscale x 2 x double> zeroinitializer, double 0.000000e+00, i64 1
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
+;
+; CHECK-HIGH-LABEL: 'ins_el1'
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 100000 for instruction: %v0 = insertelement <vscale x 16 x i8> zeroinitializer, i8 0, i64 1
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 100000 for instruction: %v1 = insertelement <vscale x 8 x i16> zeroinitializer, i16 0, i64 1
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 100000 for instruction: %v2 = insertelement <vscale x 4 x i32> zeroinitializer, i32 0, i64 1
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 100000 for instruction: %v3 = insertelement <vscale x 2 x i64> zeroinitializer, i64 0, i64 1
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 100000 for instruction: %v4 = insertelement <vscale x 4 x float> zeroinitializer, float 0.000000e+00, i64 1
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 100000 for instruction: %v5 = insertelement <vscale x 2 x double> zeroinitializer, double 0.000000e+00, i64 1
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
+;
   %v0 = insertelement <vscale x 16 x i8> zeroinitializer, i8 0, i64 1
   %v1 = insertelement <vscale x 8 x i16> zeroinitializer, i16 0, i64 1
   %v2 = insertelement <vscale x 4 x i32> zeroinitializer, i32 0, i64 1
   %v3 = insertelement <vscale x 2 x i64> zeroinitializer, i64 0, i64 1
+  %v4 = insertelement <vscale x 4 x float> zeroinitializer, float 0., i64 1
+  %v5 = insertelement <vscale x 2 x double> zeroinitializer, double 0., i64 1
   ret void
 }
 
 
 define void @ext_el0() #0 {
-; CHECK: Cost Model: Found an estimated cost of 0 for instruction:   %v0 = extractelement <vscale x 16 x i8> zeroinitializer, i64 0
-; CHECK: Cost Model: Found an estimated cost of 0 for instruction:   %v1 = extractelement <vscale x 8 x i16> zeroinitializer, i64 0
-; CHECK: Cost Model: Found an estimated cost of 0 for instruction:   %v2 = extractelement <vscale x 4 x i32> zeroinitializer, i64 0
-; CHECK: Cost Model: Found an estimated cost of 0 for instruction:   %v3 = extractelement <vscale x 2 x i64> zeroinitializer, i64 0
+; CHECK-DEFAULT-LABEL: 'ext_el0'
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v0 = extractelement <vscale x 16 x i8> zeroinitializer, i64 0
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v1 = extractelement <vscale x 8 x i16> zeroinitializer, i64 0
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v2 = extractelement <vscale x 4 x i32> zeroinitializer, i64 0
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v3 = extractelement <vscale x 2 x i64> zeroinitializer, i64 0
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v4 = extractelement <vscale x 4 x float> zeroinitializer, i64 0
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v5 = extractelement <vscale x 2 x double> zeroinitializer, i64 0
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
+;
+; CHECK-LOW-LABEL: 'ext_el0'
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v0 = extractelement <vscale x 16 x i8> zeroinitializer, i64 0
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v1 = extractelement <vscale x 8 x i16> zeroinitializer, i64 0
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v2 = extractelement <vscale x 4 x i32> zeroinitializer, i64 0
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v3 = extractelement <vscale x 2 x i64> zeroinitializer, i64 0
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v4 = extractelement <vscale x 4 x float> zeroinitializer, i64 0
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v5 = extractelement <vscale x 2 x double> zeroinitializer, i64 0
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
+;
+; CHECK-HIGH-LABEL: 'ext_el0'
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v0 = extractelement <vscale x 16 x i8> zeroinitializer, i64 0
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v1 = extractelement <vscale x 8 x i16> zeroinitializer, i64 0
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v2 = extractelement <vscale x 4 x i32> zeroinitializer, i64 0
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v3 = extractelement <vscale x 2 x i64> zeroinitializer, i64 0
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v4 = extractelement <vscale x 4 x float> zeroinitializer, i64 0
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v5 = extractelement <vscale x 2 x double> zeroinitializer, i64 0
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
+;
   %v0 = extractelement <vscale x 16 x i8> zeroinitializer, i64 0
   %v1 = extractelement <vscale x 8 x i16> zeroinitializer, i64 0
   %v2 = extractelement <vscale x 4 x i32> zeroinitializer, i64 0
   %v3 = extractelement <vscale x 2 x i64> zeroinitializer, i64 0
+  %v4 = extractelement <vscale x 4 x float> zeroinitializer, i64 0
+  %v5 = extractelement <vscale x 2 x double> zeroinitializer, i64 0
   ret void
 }
 
 define void @ext_el1() #0 {
-; CHECK: Cost Model: Found an estimated cost of 3 for instruction:   %v0 = extractelement <vscale x 16 x i8> zeroinitializer, i64 1
-; CHECK: Cost Model: Found an estimated cost of 3 for instruction:   %v1 = extractelement <vscale x 8 x i16> zeroinitializer, i64 1
-; CHECK: Cost Model: Found an estimated cost of 3 for instruction:   %v2 = extractelement <vscale x 4 x i32> zeroinitializer, i64 1
-; CHECK: Cost Model: Found an estimated cost of 3 for instruction:   %v3 = extractelement <vscale x 2 x i64> zeroinitializer, i64 1
+; CHECK-DEFAULT-LABEL: 'ext_el1'
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v0 = extractelement <vscale x 16 x i8> zeroinitializer, i64 1
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v1 = extractelement <vscale x 8 x i16> zeroinitializer, i64 1
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v2 = extractelement <vscale x 4 x i32> zeroinitializer, i64 1
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v3 = extractelement <vscale x 2 x i64> zeroinitializer, i64 1
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v4 = extractelement <vscale x 4 x float> zeroinitializer, i64 1
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v5 = extractelement <vscale x 2 x double> zeroinitializer, i64 1
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
+;
+; CHECK-LOW-LABEL: 'ext_el1'
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v0 = extractelement <vscale x 16 x i8> zeroinitializer, i64 1
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v1 = extractelement <vscale x 8 x i16> zeroinitializer, i64 1
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v2 = extractelement <vscale x 4 x i32> zeroinitializer, i64 1
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v3 = extractelement <vscale x 2 x i64> zeroinitializer, i64 1
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v4 = extractelement <vscale x 4 x float> zeroinitializer, i64 1
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v5 = extractelement <vscale x 2 x double> zeroinitializer, i64 1
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
+;
+; CHECK-HIGH-LABEL: 'ext_el1'
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 100000 for instruction: %v0 = extractelement <vscale x 16 x i8> zeroinitializer, i64 1
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 100000 for instruction: %v1 = extractelement <vscale x 8 x i16> zeroinitializer, i64 1
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 100000 for instruction: %v2 = extractelement <vscale x 4 x i32> zeroinitializer, i64 1
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 100000 for instruction: %v3 = extractelement <vscale x 2 x i64> zeroinitializer, i64 1
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 100000 for instruction: %v4 = extractelement <vscale x 4 x float> zeroinitializer, i64 1
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 100000 for instruction: %v5 = extractelement <vscale x 2 x double> zeroinitializer, i64 1
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
+;
+  %v0 = extractelement <vscale x 16 x i8> zeroinitializer, i64 1
+  %v1 = extractelement <vscale x 8 x i16> zeroinitializer, i64 1
+  %v2 = extractelement <vscale x 4 x i32> zeroinitializer, i64 1
+  %v3 = extractelement <vscale x 2 x i64> zeroinitializer, i64 1
+  %v4 = extractelement <vscale x 4 x float> zeroinitializer, i64 1
+  %v5 = extractelement <vscale x 2 x double> zeroinitializer, i64 1
+  ret void
+}
+
+
+; Test the behaviour in the presence of a CPU-specific override in AArch64Subtarget (via attribute set).
+define void @test_override_cpu_given() #1 {
+; CHECK-DEFAULT-LABEL: 'test_override_cpu_given'
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v0 = extractelement <vscale x 16 x i8> zeroinitializer, i64 1
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v1 = extractelement <vscale x 8 x i16> zeroinitializer, i64 1
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v2 = extractelement <vscale x 4 x i32> zeroinitializer, i64 1
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v3 = extractelement <vscale x 2 x i64> zeroinitializer, i64 1
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4 = extractelement <vscale x 4 x float> zeroinitializer, i64 1
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v5 = extractelement <vscale x 2 x double> zeroinitializer, i64 1
+; CHECK-DEFAULT-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
+;
+; CHECK-LOW-LABEL: 'test_override_cpu_given'
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v0 = extractelement <vscale x 16 x i8> zeroinitializer, i64 1
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v1 = extractelement <vscale x 8 x i16> zeroinitializer, i64 1
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v2 = extractelement <vscale x 4 x i32> zeroinitializer, i64 1
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v3 = extractelement <vscale x 2 x i64> zeroinitializer, i64 1
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v4 = extractelement <vscale x 4 x float> zeroinitializer, i64 1
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v5 = extractelement <vscale x 2 x double> zeroinitializer, i64 1
+; CHECK-LOW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
+;
+; CHECK-HIGH-LABEL: 'test_override_cpu_given'
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 100000 for instruction: %v0 = extractelement <vscale x 16 x i8> zeroinitializer, i64 1
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 100000 for instruction: %v1 = extractelement <vscale x 8 x i16> zeroinitializer, i64 1
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 100000 for instruction: %v2 = extractelement <vscale x 4 x i32> zeroinitializer, i64 1
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 100000 for instruction: %v3 = extractelement <vscale x 2 x i64> zeroinitializer, i64 1
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 100000 for instruction: %v4 = extractelement <vscale x 4 x float> zeroinitializer, i64 1
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 100000 for instruction: %v5 = extractelement <vscale x 2 x double> zeroinitializer, i64 1
+; CHECK-HIGH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
+;
   %v0 = extractelement <vscale x 16 x i8> zeroinitializer, i64 1
   %v1 = extractelement <vscale x 8 x i16> zeroinitializer, i64 1
   %v2 = extractelement <vscale x 4 x i32> zeroinitializer, i64 1
   %v3 = extractelement <vscale x 2 x i64> zeroinitializer, i64 1
+  %v4 = extractelement <vscale x 4 x float> zeroinitializer, i64 1
+  %v5 = extractelement <vscale x 2 x double> zeroinitializer, i64 1
   ret void
 }
 
 
+
+
+
 attributes #0 = { "target-features"="+sve" vscale_range(1, 16) }
+attributes #1 = { "target-features"="+sve" vscale_range(1, 16) "target-cpu"="kryo" }