Reland: [turbofan] Various fixes to allow unboxed doubles as arguments in registers and on the stac…

test/cctest/compiler/test-run-native-calls.cc

+// Copyright 2014 the V8 project authors. All rights reserved.
+// Use of this source code is governed by a BSD-style license that can be
+// found in the LICENSE file.
+
+#include "src/assembler.h"
+#include "src/codegen.h"
+#include "src/compiler/linkage.h"
+#include "src/compiler/machine-type.h"
+#include "src/compiler/raw-machine-assembler.h"
+
+#include "test/cctest/cctest.h"
+#include "test/cctest/compiler/codegen-tester.h"
+#include "test/cctest/compiler/graph-builder-tester.h"
+#include "test/cctest/compiler/value-helper.h"
+
+using namespace v8::base;
+using namespace v8::internal;
+using namespace v8::internal::compiler;
+
+typedef RawMachineAssembler::Label MLabel;
+
+#if !V8_TARGET_ARCH_ARM64
+// TODO(titzer): fix native stack parameters on arm64
+#define NATIVE_STACK_PARAMS_OK
+#endif
+
+namespace {
+// Picks a representative set of registers from the allocatable set.
+// If there are less than 100 possible pairs, do them all, otherwise try
+// to select a representative set.
+class RegisterPairs {
+ public:
+  RegisterPairs()
+      : max_(std::min(100, Register::kMaxNumAllocatableRegisters *
+                               Register::kMaxNumAllocatableRegisters)),
+        counter_(0) {}
+
+  bool More() { return counter_ < max_; }
+
+  void Next(int* r0, int* r1, bool same_is_ok) {
+    do {
+      // Find the next pair.
+      if (exhaustive()) {
+        *r0 = counter_ % Register::kMaxNumAllocatableRegisters;
+        *r1 = counter_ / Register::kMaxNumAllocatableRegisters;
+      } else {
+        // Try each register at least once for both r0 and r1.
+        int index = counter_ / 2;
+        if (counter_ & 1) {
+          *r0 = index % Register::kMaxNumAllocatableRegisters;
+          *r1 = index / Register::kMaxNumAllocatableRegisters;
+        } else {
+          *r1 = index % Register::kMaxNumAllocatableRegisters;
+          *r0 = index / Register::kMaxNumAllocatableRegisters;
+        }
+      }
+      counter_++;
+      if (same_is_ok) break;
+      if (*r0 == *r1) {
+        if (counter_ >= max_) {
+          // For the last hurrah, reg#0 with reg#n-1
+          *r0 = 0;
+          *r1 = Register::kMaxNumAllocatableRegisters - 1;
+          break;
+        }
+      }
+    } while (true);
+
+    DCHECK(*r0 >= 0 && *r0 < Register::kMaxNumAllocatableRegisters);
+    DCHECK(*r1 >= 0 && *r1 < Register::kMaxNumAllocatableRegisters);
+    printf("pair = %d, %d\n", *r0, *r1);
+  }
+
+ private:
+  int max_;
+  int counter_;
+  bool exhaustive() {
+    return max_ == (Register::kMaxNumAllocatableRegisters *
+                    Register::kMaxNumAllocatableRegisters);
+  }
+};
+
+
+// Helper for allocating either an GP or FP reg, or the next stack slot.
+struct Allocator {
+  Allocator(int* gp, int gpc, int* fp, int fpc)
+      : gp_count(gpc),
+        gp_offset(0),
+        gp_regs(gp),
+        fp_count(fpc),
+        fp_offset(0),
+        fp_regs(fp),
+        stack_offset(0) {}
+
+  int gp_count;
+  int gp_offset;
+  int* gp_regs;
+
+  int fp_count;
+  int fp_offset;
+  int* fp_regs;
+
+  int stack_offset;
+
+  LinkageLocation Next(MachineType type) {
+    if (IsFloatingPoint(type)) {
+      // Allocate a floating point register/stack location.
+      if (fp_offset < fp_count) {
+        return LinkageLocation::ForRegister(fp_regs[fp_offset++]);
+      } else {
+        int offset = -1 - stack_offset;
+        stack_offset += Words(type);
+        return LinkageLocation::ForCallerFrameSlot(offset);
+      }
+    } else {
+      // Allocate a general purpose register/stack location.
+      if (gp_offset < gp_count) {
+        return LinkageLocation::ForRegister(gp_regs[gp_offset++]);
+      } else {
+        int offset = -1 - stack_offset;
+        stack_offset += Words(type);
+        return LinkageLocation::ForCallerFrameSlot(offset);
+      }
+    }
+  }
+  bool IsFloatingPoint(MachineType type) {
+    return RepresentationOf(type) == kRepFloat32 ||
+           RepresentationOf(type) == kRepFloat64;
+  }
+  int Words(MachineType type) {
+    int size = ElementSizeOf(type);
+    return size <= kPointerSize ? 1 : size / kPointerSize;
+  }
+  void Reset() {
+    fp_offset = 0;
+    gp_offset = 0;
+    stack_offset = 0;
+  }
+};
+
+
+class RegisterConfig {
+ public:
+  RegisterConfig(Allocator& p, Allocator& r) : params(p), rets(r) {}
+
+  CallDescriptor* Create(Zone* zone, MachineSignature* msig) {
+    rets.Reset();
+    params.Reset();
+
+    LocationSignature::Builder locations(zone, msig->return_count(),
+                                         msig->parameter_count());
+    // Add return location(s).
+    const int return_count = static_cast<int>(locations.return_count_);
+    for (int i = 0; i < return_count; i++) {
+      locations.AddReturn(rets.Next(msig->GetReturn(i)));
+    }
+
+    // Add register and/or stack parameter(s).
+    const int parameter_count = static_cast<int>(msig->parameter_count());
+    for (int i = 0; i < parameter_count; i++) {
+      locations.AddParam(params.Next(msig->GetParam(i)));
+    }
+
+    const RegList kCalleeSaveRegisters = 0;
+    const RegList kCalleeSaveFPRegisters = 0;
+
+    MachineType target_type = compiler::kMachAnyTagged;
+    LinkageLocation target_loc = LinkageLocation::ForAnyRegister();
+    int stack_param_count = params.stack_offset;
+    return new (zone) CallDescriptor(       // --
+        CallDescriptor::kCallCodeObject,    // kind
+        target_type,                        // target MachineType
+        target_loc,                         // target location
+        msig,                               // machine_sig
+        locations.Build(),                  // location_sig
+        stack_param_count,                  // stack_parameter_count
+        compiler::Operator::kNoProperties,  // properties
+        kCalleeSaveRegisters,               // callee-saved registers
+        kCalleeSaveFPRegisters,             // callee-saved fp regs
+        CallDescriptor::kNoFlags,           // flags
+        "c-call");
+  }
+
+ private:
+  Allocator& params;
+  Allocator& rets;
+};
+
+const int kMaxParamCount = 64;
+
+MachineType kIntTypes[kMaxParamCount + 1] = {
+    kMachInt32, kMachInt32, kMachInt32, kMachInt32, kMachInt32, kMachInt32,
+    kMachInt32, kMachInt32, kMachInt32, kMachInt32, kMachInt32, kMachInt32,
+    kMachInt32, kMachInt32, kMachInt32, kMachInt32, kMachInt32, kMachInt32,
+    kMachInt32, kMachInt32, kMachInt32, kMachInt32, kMachInt32, kMachInt32,
+    kMachInt32, kMachInt32, kMachInt32, kMachInt32, kMachInt32, kMachInt32,
+    kMachInt32, kMachInt32, kMachInt32, kMachInt32, kMachInt32, kMachInt32,
+    kMachInt32, kMachInt32, kMachInt32, kMachInt32, kMachInt32, kMachInt32,
+    kMachInt32, kMachInt32, kMachInt32, kMachInt32, kMachInt32, kMachInt32,
+    kMachInt32, kMachInt32, kMachInt32, kMachInt32, kMachInt32, kMachInt32,
+    kMachInt32, kMachInt32, kMachInt32, kMachInt32, kMachInt32, kMachInt32,
+    kMachInt32, kMachInt32, kMachInt32, kMachInt32, kMachInt32};
+
+
+// For making uniform int32 signatures shorter.
+class Int32Signature : public MachineSignature {
+ public:
+  explicit Int32Signature(int param_count)
+      : MachineSignature(1, param_count, kIntTypes) {
+    CHECK(param_count <= kMaxParamCount);
+  }
+};
+
+
+Handle<Code> CompileGraph(const char* name, CallDescriptor* desc, Graph* graph,
+                          Schedule* schedule = nullptr) {
+  Isolate* isolate = CcTest::InitIsolateOnce();
+  Handle<Code> code =
+      Pipeline::GenerateCodeForTesting(isolate, desc, graph, schedule);
+  CHECK(!code.is_null());
+#ifdef ENABLE_DISASSEMBLER
+  if (FLAG_print_opt_code) {
+    OFStream os(stdout);
+    code->Disassemble(name, os);
+  }
+#endif
+  return code;
+}
+
+
+Handle<Code> WrapWithCFunction(Handle<Code> inner, CallDescriptor* desc) {
+  Zone zone;
+  MachineSignature* msig =
+      const_cast<MachineSignature*>(desc->GetMachineSignature());
+  int param_count = static_cast<int>(msig->parameter_count());
+  GraphAndBuilders caller(&zone);
+  {
+    GraphAndBuilders& b = caller;
+    Node* start = b.graph()->NewNode(b.common()->Start(param_count + 3));
+    b.graph()->SetStart(start);
+    Unique<HeapObject> unique = Unique<HeapObject>::CreateUninitialized(inner);
+    Node* target = b.graph()->NewNode(b.common()->HeapConstant(unique));
+
+    // Add arguments to the call.
+    Node** args = zone.NewArray<Node*>(param_count + 3);
+    int index = 0;
+    args[index++] = target;
+    for (int i = 0; i < param_count; i++) {
+      args[index] = b.graph()->NewNode(b.common()->Parameter(i), start);
+      index++;
+    }
+    args[index++] = start;  // effect.
+    args[index++] = start;  // control.
+
+    // Build the call and return nodes.
+    Node* call =
+        b.graph()->NewNode(b.common()->Call(desc), param_count + 3, args);
+    Node* ret = b.graph()->NewNode(b.common()->Return(), call, call, start);
+    b.graph()->SetEnd(ret);
+  }
+
+  CallDescriptor* cdesc = Linkage::GetSimplifiedCDescriptor(&zone, msig);
+
+  return CompileGraph("wrapper", cdesc, caller.graph());
+}
+
+
+}  // namespace
+
+
+static void TestInt32Sub(CallDescriptor* desc) {
+  Isolate* isolate = CcTest::InitIsolateOnce();
+  HandleScope scope(isolate);
+  Zone zone;
+  GraphAndBuilders inner(&zone);
+  {
+    // Build the add function.
+    GraphAndBuilders& b = inner;
+    Node* start = b.graph()->NewNode(b.common()->Start(5));
+    b.graph()->SetStart(start);
+    Node* p0 = b.graph()->NewNode(b.common()->Parameter(0), start);
+    Node* p1 = b.graph()->NewNode(b.common()->Parameter(1), start);
+    Node* add = b.graph()->NewNode(b.machine()->Int32Sub(), p0, p1);
+    Node* ret = b.graph()->NewNode(b.common()->Return(), add, start, start);
+    b.graph()->SetEnd(ret);
+  }
+
+  Handle<Code> inner_code = CompileGraph("Int32Sub", desc, inner.graph());
+  Handle<Code> wrapper = WrapWithCFunction(inner_code, desc);
+  MachineSignature* msig =
+      const_cast<MachineSignature*>(desc->GetMachineSignature());
+  CodeRunner<int32_t> runnable(isolate, wrapper,
+                               CSignature::FromMachine(&zone, msig));
+
+  FOR_INT32_INPUTS(i) {
+    FOR_INT32_INPUTS(j) {
+      int32_t expected = static_cast<int32_t>(static_cast<uint32_t>(*i) -
+                                              static_cast<uint32_t>(*j));
+      int32_t result = runnable.Call(*i, *j);
+      CHECK_EQ(expected, result);
+    }
+  }
+}
+
+
+#ifdef NATIVE_STACK_PARAMS_OK
+static void CopyTwentyInt32(CallDescriptor* desc) {
+  const int kNumParams = 20;
+  int32_t input[kNumParams];
+  int32_t output[kNumParams];
+  Isolate* isolate = CcTest::InitIsolateOnce();
+  HandleScope scope(isolate);
+  Handle<Code> inner = Handle<Code>::null();
+  {
+    // Writes all parameters into the output buffer.
+    Zone zone;
+    Graph graph(&zone);
+    RawMachineAssembler raw(isolate, &graph, desc);
+    Node* base = raw.PointerConstant(output);
+    for (int i = 0; i < kNumParams; i++) {
+      Node* offset = raw.Int32Constant(i * sizeof(int32_t));
+      raw.Store(kMachInt32, base, offset, raw.Parameter(i));
+    }
+    raw.Return(raw.Int32Constant(42));
+    inner = CompileGraph("CopyTwentyInt32", desc, &graph, raw.Export());
+  }
+
+  CSignature0<int32_t> csig;
+  Handle<Code> wrapper = Handle<Code>::null();
+  {
+    // Loads parameters from the input buffer and calls the above code.
+    Zone zone;
+    Graph graph(&zone);
+    CallDescriptor* cdesc = Linkage::GetSimplifiedCDescriptor(&zone, &csig);
+    RawMachineAssembler raw(isolate, &graph, cdesc);
+    Node* base = raw.PointerConstant(input);
+    Unique<HeapObject> unique = Unique<HeapObject>::CreateUninitialized(inner);
+    Node* target = raw.HeapConstant(unique);
+    Node** args = zone.NewArray<Node*>(kNumParams);
+    for (int i = 0; i < kNumParams; i++) {
+      Node* offset = raw.Int32Constant(i * sizeof(int32_t));
+      args[i] = raw.Load(kMachInt32, base, offset);
+    }
+
+    Node* call = raw.CallN(desc, target, args);
+    raw.Return(call);
+    wrapper =
+        CompileGraph("CopyTwentyInt32-wrapper", cdesc, &graph, raw.Export());
+  }
+
+  CodeRunner<int32_t> runnable(isolate, wrapper, &csig);
+
+  // Run the code, checking it correctly implements the memcpy.
+  for (int i = 0; i < 5; i++) {
+    uint32_t base = 1111111111u * i;
+    for (int j = 0; j < kNumParams; j++) {
+      input[j] = static_cast<int32_t>(base + 13 * j);
+    }
+
+    memset(output, 0, sizeof(output));
+    CHECK_EQ(42, runnable.Call());
+
+    for (int j = 0; j < kNumParams; j++) {
+      CHECK_EQ(input[j], output[j]);
+    }
+  }
+}
+#endif  // NATIVE_STACK_PARAMS_OK
+
+
+static void Test_RunInt32SubWithRet(int retreg) {
+  Int32Signature sig(2);
+  Zone zone;
+  RegisterPairs pairs;
+  while (pairs.More()) {
+    int parray[2];
+    int rarray[] = {retreg};
+    pairs.Next(&parray[0], &parray[1], false);
+    Allocator params(parray, 2, nullptr, 0);
+    Allocator rets(rarray, 1, nullptr, 0);
+    RegisterConfig config(params, rets);
+    CallDescriptor* desc = config.Create(&zone, &sig);
+    TestInt32Sub(desc);
+  }
+}
+
+
+// Separate tests for parallelization.
+#define TEST_INT32_SUB_WITH_RET(x)                                             \
+  TEST(Run_Int32Sub_all_allocatable_pairs_##x) {                               \
+    if (Register::kMaxNumAllocatableRegisters > x) Test_RunInt32SubWithRet(x); \
+  }
+
+
+TEST_INT32_SUB_WITH_RET(0)
+TEST_INT32_SUB_WITH_RET(1)
+TEST_INT32_SUB_WITH_RET(2)
+TEST_INT32_SUB_WITH_RET(3)
+TEST_INT32_SUB_WITH_RET(4)
+TEST_INT32_SUB_WITH_RET(5)
+TEST_INT32_SUB_WITH_RET(6)
+TEST_INT32_SUB_WITH_RET(7)
+TEST_INT32_SUB_WITH_RET(8)
+TEST_INT32_SUB_WITH_RET(9)
+TEST_INT32_SUB_WITH_RET(10)
+TEST_INT32_SUB_WITH_RET(11)
+TEST_INT32_SUB_WITH_RET(12)
+TEST_INT32_SUB_WITH_RET(13)
+TEST_INT32_SUB_WITH_RET(14)
+TEST_INT32_SUB_WITH_RET(15)
+TEST_INT32_SUB_WITH_RET(16)
+TEST_INT32_SUB_WITH_RET(17)
+TEST_INT32_SUB_WITH_RET(18)
+TEST_INT32_SUB_WITH_RET(19)
+
+
+TEST(Run_Int32Sub_all_allocatable_single) {
+#ifdef NATIVE_STACK_PARAMS_OK
+  Int32Signature sig(2);
+  RegisterPairs pairs;
+  while (pairs.More()) {
+    Zone zone;
+    int parray[1];
+    int rarray[1];
+    pairs.Next(&rarray[0], &parray[0], true);
+    Allocator params(parray, 1, nullptr, 0);
+    Allocator rets(rarray, 1, nullptr, 0);
+    RegisterConfig config(params, rets);
+    CallDescriptor* desc = config.Create(&zone, &sig);
+    TestInt32Sub(desc);
+  }
+#endif  // NATIVE_STACK_PARAMS_OK
+}
+
+
+TEST(Run_CopyTwentyInt32_all_allocatable_pairs) {
+#ifdef NATIVE_STACK_PARAMS_OK
+  Int32Signature sig(20);
+  RegisterPairs pairs;
+  while (pairs.More()) {
+    Zone zone;
+    int parray[2];
+    int rarray[] = {0};
+    pairs.Next(&parray[0], &parray[1], false);
+    Allocator params(parray, 2, nullptr, 0);
+    Allocator rets(rarray, 1, nullptr, 0);
+    RegisterConfig config(params, rets);
+    CallDescriptor* desc = config.Create(&zone, &sig);
+    CopyTwentyInt32(desc);
+  }
+#endif  // NATIVE_STACK_PARAMS_OK
+}
+
+
+#ifdef NATIVE_STACK_PARAMS_OK
+int ParamCount(CallDescriptor* desc) {
+  return static_cast<int>(desc->GetMachineSignature()->parameter_count());
+}
+
+
+// Super mega helper routine to generate a computation with a given
+// call descriptor, compile the code, wrap the code, and pass various inputs,
+// comparing against a reference implementation.
+static void Run_Int32_Computation(
+    CallDescriptor* desc, void (*build)(CallDescriptor*, RawMachineAssembler&),
+    int32_t (*compute)(CallDescriptor*, int32_t* inputs), int seed = 1) {
+  int num_params = ParamCount(desc);
+  CHECK_LE(num_params, kMaxParamCount);
+  int32_t input[kMaxParamCount];
+  Isolate* isolate = CcTest::InitIsolateOnce();
+  HandleScope scope(isolate);
+  Handle<Code> inner = Handle<Code>::null();
+  {
+    // Build the graph for the computation.
+    Zone zone;
+    Graph graph(&zone);
+    RawMachineAssembler raw(isolate, &graph, desc);
+    build(desc, raw);
+    inner = CompileGraph("Compute", desc, &graph, raw.Export());
+  }
+
+  CSignature0<int32_t> csig;
+
+  if (false) {
+    // constant mode.
+    Handle<Code> wrapper = Handle<Code>::null();
+    {
+      // Wrap the above code with a callable function that passes constants.
+      Zone zone;
+      Graph graph(&zone);
+      CallDescriptor* cdesc = Linkage::GetSimplifiedCDescriptor(&zone, &csig);
+      RawMachineAssembler raw(isolate, &graph, cdesc);
+      Unique<HeapObject> unique =
+          Unique<HeapObject>::CreateUninitialized(inner);
+      Node* target = raw.HeapConstant(unique);
+      Node** args = zone.NewArray<Node*>(kMaxParamCount);
+      for (int i = 0; i < num_params; i++) {
+        args[i] = raw.Int32Constant(0x100 + i);
+      }
+
+      Node* call = raw.CallN(desc, target, args);
+      raw.Return(call);
+      wrapper = CompileGraph("Compute-wrapper", cdesc, &graph, raw.Export());
+    }
+
+    CodeRunner<int32_t> runnable(isolate, wrapper, &csig);
+
+    // Run the code, checking it against the reference.
+    for (int j = 0; j < kMaxParamCount; j++) {
+      input[j] = 0x100 + j;
+    }
+    int32_t expected = compute(desc, input);
+    int32_t result = runnable.Call();
+
+    CHECK_EQ(expected, result);
+  }
+
+  {
+    // buffer mode.
+    Handle<Code> wrapper = Handle<Code>::null();
+    {
+      // Wrap the above code with a callable function that loads from {input}.
+      Zone zone;
+      Graph graph(&zone);
+      CallDescriptor* cdesc = Linkage::GetSimplifiedCDescriptor(&zone, &csig);
+      RawMachineAssembler raw(isolate, &graph, cdesc);
+      Node* base = raw.PointerConstant(input);
+      Unique<HeapObject> unique =
+          Unique<HeapObject>::CreateUninitialized(inner);
+      Node* target = raw.HeapConstant(unique);
+      Node** args = zone.NewArray<Node*>(kMaxParamCount);
+      for (int i = 0; i < num_params; i++) {
+        Node* offset = raw.Int32Constant(i * sizeof(int32_t));
+        args[i] = raw.Load(kMachInt32, base, offset);
+      }
+
+      Node* call = raw.CallN(desc, target, args);
+      raw.Return(call);
+      wrapper = CompileGraph("Compute-wrapper", cdesc, &graph, raw.Export());
+    }
+
+    CodeRunner<int32_t> runnable(isolate, wrapper, &csig);
+
+    // Run the code, checking it against the reference.
+    for (int i = 0; i < 5; i++) {
+      // Use pseudo-random values for each run, but the first run gets args
+      // 100, 101, 102, 103... for easier diagnosis.
+      uint32_t base = 1111111111u * i * seed;
+      for (int j = 0; j < kMaxParamCount; j++) {
+        input[j] = static_cast<int32_t>(100 + base + j);
+      }
+      int32_t expected = compute(desc, input);
+      int32_t result = runnable.Call();
+
+      CHECK_EQ(expected, result);
+    }
+  }
+}
+
+
+static uint32_t coeff[] = {1,  2,  3,  5,  7,   11,  13,  17,  19, 23, 29,
+                           31, 37, 41, 43, 47,  53,  59,  61,  67, 71, 73,
+                           79, 83, 89, 97, 101, 103, 107, 109, 113};
+
+
+static void Build_Int32_WeightedSum(CallDescriptor* desc,
+                                    RawMachineAssembler& raw) {
+  Node* result = raw.Int32Constant(0);
+  for (int i = 0; i < ParamCount(desc); i++) {
+    Node* term = raw.Int32Mul(raw.Parameter(i), raw.Int32Constant(coeff[i]));
+    result = raw.Int32Add(result, term);
+  }
+  raw.Return(result);
+}
+
+
+static int32_t Compute_Int32_WeightedSum(CallDescriptor* desc, int32_t* input) {
+  uint32_t result = 0;
+  for (int i = 0; i < ParamCount(desc); i++) {
+    result += static_cast<uint32_t>(input[i]) * coeff[i];
+  }
+  return static_cast<int32_t>(result);
+}
+
+
+static void Test_Int32_WeightedSum_of_size(int count) {
+  Int32Signature sig(count);
+  for (int p0 = 0; p0 < Register::kMaxNumAllocatableRegisters; p0++) {
+    Zone zone;
+
+    int parray[] = {p0};
+    int rarray[] = {0};
+    Allocator params(parray, 1, nullptr, 0);
+    Allocator rets(rarray, 1, nullptr, 0);
+    RegisterConfig config(params, rets);
+    CallDescriptor* desc = config.Create(&zone, &sig);
+    Run_Int32_Computation(desc, Build_Int32_WeightedSum,
+                          Compute_Int32_WeightedSum, 257 + count);
+  }
+}
+
+
+// Separate tests for parallelization.
+#define TEST_INT32_WEIGHTEDSUM(x) \
+  TEST(Run_Int32_WeightedSum_##x) { Test_Int32_WeightedSum_of_size(x); }
+
+
+TEST_INT32_WEIGHTEDSUM(1)
+TEST_INT32_WEIGHTEDSUM(2)
+TEST_INT32_WEIGHTEDSUM(3)
+TEST_INT32_WEIGHTEDSUM(4)
+TEST_INT32_WEIGHTEDSUM(5)
+TEST_INT32_WEIGHTEDSUM(7)
+TEST_INT32_WEIGHTEDSUM(9)
+TEST_INT32_WEIGHTEDSUM(11)
+TEST_INT32_WEIGHTEDSUM(17)
+TEST_INT32_WEIGHTEDSUM(19)
+
+
+template <int which>
+static void Build_Int32_Select(CallDescriptor* desc, RawMachineAssembler& raw) {
+  raw.Return(raw.Parameter(which));
+}
+
+
+template <int which>
+static int32_t Compute_Int32_Select(CallDescriptor* desc, int32_t* inputs) {
+  return inputs[which];
+}
+
+
+template <int which>
+void Test_Int32_Select() {
+  int parray[] = {0};
+  int rarray[] = {0};
+  Allocator params(parray, 1, nullptr, 0);
+  Allocator rets(rarray, 1, nullptr, 0);
+  RegisterConfig config(params, rets);
+
+  Zone zone;
+
+  for (int i = which + 1; i <= 64; i++) {
+    Int32Signature sig(i);
+    CallDescriptor* desc = config.Create(&zone, &sig);
+    Run_Int32_Computation(desc, Build_Int32_Select<which>,
+                          Compute_Int32_Select<which>, 1025 + which);
+  }
+}
+
+
+// Separate tests for parallelization.
+#define TEST_INT32_SELECT(x) \
+  TEST(Run_Int32_Select_##x) { Test_Int32_Select<x>(); }
+
+
+TEST_INT32_SELECT(0)
+TEST_INT32_SELECT(1)
+TEST_INT32_SELECT(2)
+TEST_INT32_SELECT(3)
+TEST_INT32_SELECT(4)
+TEST_INT32_SELECT(5)
+TEST_INT32_SELECT(6)
+TEST_INT32_SELECT(11)
+TEST_INT32_SELECT(15)
+TEST_INT32_SELECT(19)
+TEST_INT32_SELECT(45)
+TEST_INT32_SELECT(62)
+TEST_INT32_SELECT(63)
+#endif  // NATIVE_STACK_PARAMS_OK
+
+
+TEST(TheLastTestForLint) {
+  // Yes, thank you.
+}