MFC r309124:

Upgrade our copies of clang, llvm, lldb, compiler-rt and libc++ to 3.9.0
release, and add lld 3.9.0.  Also completely revamp the build system for
clang, llvm, lldb and their related tools.

Please note that from 3.5.0 onwards, clang, llvm and lldb require C++11
support to build; see UPDATING for more information.

Release notes for llvm, clang and lld are available here:
<http://llvm.org/releases/3.9.0/docs/ReleaseNotes.html>
<http://llvm.org/releases/3.9.0/tools/clang/docs/ReleaseNotes.html>
<http://llvm.org/releases/3.9.0/tools/lld/docs/ReleaseNotes.html>

Thanks to Ed Maste, Bryan Drewery, Andrew Turner, Antoine Brodin and Jan
Beich for their help.

Relnotes:	yes

MFC r309147:

Pull in r282174 from upstream llvm trunk (by Krzysztof Parzyszek):

  [PPC] Set SP after loading data from stack frame, if no red zone is
  present

  Follow-up to r280705: Make sure that the SP is only restored after
  all data is loaded from the stack frame, if there is no red zone.

  This completes the fix for
  https://llvm.org/bugs/show_bug.cgi?id=26519.

  Differential Revision: https://reviews.llvm.org/D24466

Reported by:    Mark Millard
PR:             214433

MFC r309149:

Pull in r283060 from upstream llvm trunk (by Hal Finkel):

  [PowerPC] Refactor soft-float support, and enable PPC64 soft float

  This change enables soft-float for PowerPC64, and also makes
  soft-float disable all vector instruction sets for both 32-bit and
  64-bit modes. This latter part is necessary because the PPC backend
  canonicalizes many Altivec vector types to floating-point types, and
  so soft-float breaks scalarization support for many operations. Both
  for embedded targets and for operating-system kernels desiring
  soft-float support, it seems reasonable that disabling hardware
  floating-point also disables vector instructions (embedded targets
  without hardware floating point support are unlikely to have Altivec,
  etc. and operating system kernels desiring not to use floating-point
  registers to lower syscall cost are unlikely to want to use vector
  registers either). If someone needs this to work, we'll need to
  change the fact that we promote many Altivec operations to act on
  v4f32. To make it possible to disable Altivec when soft-float is
  enabled, hardware floating-point support needs to be expressed as a
  positive feature, like the others, and not a negative feature,
  because target features cannot have dependencies on the disabling of
  some other feature. So +soft-float has now become -hard-float.

  Fixes PR26970.

Pull in r283061 from upstream clang trunk (by Hal Finkel):

  [PowerPC] Enable soft-float for PPC64, and +soft-float -> -hard-float

  Enable soft-float support on PPC64, as the backend now supports it.
  Also, the backend now uses -hard-float instead of +soft-float, so set
  the target features accordingly.

  Fixes PR26970.

Reported by:	Mark Millard
PR:		214433

MFC r309212:

Add a few missed clang 3.9.0 files to OptionalObsoleteFiles.

MFC r309262:

Fix packaging for clang, lldb and lld 3.9.0

During the upgrade of clang/llvm etc to 3.9.0 in r309124, the PACKAGE
directive in the usr.bin/clang/*.mk files got dropped accidentally.

Restore it, with a few minor changes and additions:
* Correct license in clang.ucl to NCSA
* Add PACKAGE=clang for clang and most of the "ll" tools
* Put lldb in its own package
* Put lld in its own package

Reviewed by:	gjb, jmallett
Differential Revision: https://reviews.freebsd.org/D8666

MFC r309656:

During the bootstrap phase, when building the minimal llvm library on
PowerPC, add lib/Support/Atomic.cpp.  This is needed because upstream
llvm revision r271821 disabled the use of std::call_once, which causes
some fallback functions from Atomic.cpp to be used instead.

Reported by:	Mark Millard
PR:		214902

MFC r309835:

Tentatively apply https://reviews.llvm.org/D18730 to work around gcc PR
70528 (bogus error: constructor required before non-static data member).
This should fix buildworld with the external gcc package.

Reported by:	https://jenkins.freebsd.org/job/FreeBSD_HEAD_amd64_gcc/

MFC r310194:

Upgrade our copies of clang, llvm, lld, lldb, compiler-rt and libc++ to
3.9.1 release.

Please note that from 3.5.0 onwards, clang, llvm and lldb require C++11
support to build; see UPDATING for more information.

Release notes for llvm, clang and lld will be available here:
<http://releases.llvm.org/3.9.1/docs/ReleaseNotes.html>
<http://releases.llvm.org/3.9.1/tools/clang/docs/ReleaseNotes.html>
<http://releases.llvm.org/3.9.1/tools/lld/docs/ReleaseNotes.html>

Relnotes:	yes

1 files changed, 266 insertions, 533 deletions

diff --git a/contrib/llvm/lib/Transforms/Utils/SimplifyLibCalls.cpp b/contrib/llvm/lib/Transforms/Utils/SimplifyLibCalls.cpp
index 2f3c311..c298695 100644
--- a/contrib/llvm/lib/Transforms/Utils/SimplifyLibCalls.cpp
+++ b/contrib/llvm/lib/Transforms/Utils/SimplifyLibCalls.cpp
@@ -29,7 +29,6 @@
 #include "llvm/IR/LLVMContext.h"
 #include "llvm/IR/Module.h"
 #include "llvm/IR/PatternMatch.h"
-#include "llvm/Support/Allocator.h"
 #include "llvm/Support/CommandLine.h"
 #include "llvm/Transforms/Utils/BuildLibCalls.h"
 #include "llvm/Transforms/Utils/Local.h"
@@ -104,101 +103,11 @@ static bool hasUnaryFloatFn(const TargetLibraryInfo *TLI, Type *Ty,
   }
 }
 
-/// \brief Check whether we can use unsafe floating point math for
-/// the function passed as input.
-static bool canUseUnsafeFPMath(Function *F) {
-
-  // FIXME: For finer-grain optimization, we need intrinsics to have the same
-  // fast-math flag decorations that are applied to FP instructions. For now,
-  // we have to rely on the function-level unsafe-fp-math attribute to do this
-  // optimization because there's no other way to express that the call can be
-  // relaxed.
-  if (F->hasFnAttribute("unsafe-fp-math")) {
-    Attribute Attr = F->getFnAttribute("unsafe-fp-math");
-    if (Attr.getValueAsString() == "true")
-      return true;
-  }
-  return false;
-}
-
-/// \brief Returns whether \p F matches the signature expected for the
-/// string/memory copying library function \p Func.
-/// Acceptable functions are st[rp][n]?cpy, memove, memcpy, and memset.
-/// Their fortified (_chk) counterparts are also accepted.
-static bool checkStringCopyLibFuncSignature(Function *F, LibFunc::Func Func) {
-  const DataLayout &DL = F->getParent()->getDataLayout();
-  FunctionType *FT = F->getFunctionType();
-  LLVMContext &Context = F->getContext();
-  Type *PCharTy = Type::getInt8PtrTy(Context);
-  Type *SizeTTy = DL.getIntPtrType(Context);
-  unsigned NumParams = FT->getNumParams();
-
-  // All string libfuncs return the same type as the first parameter.
-  if (FT->getReturnType() != FT->getParamType(0))
-    return false;
-
-  switch (Func) {
-  default:
-    llvm_unreachable("Can't check signature for non-string-copy libfunc.");
-  case LibFunc::stpncpy_chk:
-  case LibFunc::strncpy_chk:
-    --NumParams; // fallthrough
-  case LibFunc::stpncpy:
-  case LibFunc::strncpy: {
-    if (NumParams != 3 || FT->getParamType(0) != FT->getParamType(1) ||
-        FT->getParamType(0) != PCharTy || !FT->getParamType(2)->isIntegerTy())
-      return false;
-    break;
-  }
-  case LibFunc::strcpy_chk:
-  case LibFunc::stpcpy_chk:
-    --NumParams; // fallthrough
-  case LibFunc::stpcpy:
-  case LibFunc::strcpy: {
-    if (NumParams != 2 || FT->getParamType(0) != FT->getParamType(1) ||
-        FT->getParamType(0) != PCharTy)
-      return false;
-    break;
-  }
-  case LibFunc::memmove_chk:
-  case LibFunc::memcpy_chk:
-    --NumParams; // fallthrough
-  case LibFunc::memmove:
-  case LibFunc::memcpy: {
-    if (NumParams != 3 || !FT->getParamType(0)->isPointerTy() ||
-        !FT->getParamType(1)->isPointerTy() || FT->getParamType(2) != SizeTTy)
-      return false;
-    break;
-  }
-  case LibFunc::memset_chk:
-    --NumParams; // fallthrough
-  case LibFunc::memset: {
-    if (NumParams != 3 || !FT->getParamType(0)->isPointerTy() ||
-        !FT->getParamType(1)->isIntegerTy() || FT->getParamType(2) != SizeTTy)
-      return false;
-    break;
-  }
-  }
-  // If this is a fortified libcall, the last parameter is a size_t.
-  if (NumParams == FT->getNumParams() - 1)
-    return FT->getParamType(FT->getNumParams() - 1) == SizeTTy;
-  return true;
-}
-
 //===----------------------------------------------------------------------===//
 // String and Memory Library Call Optimizations
 //===----------------------------------------------------------------------===//
 
 Value *LibCallSimplifier::optimizeStrCat(CallInst *CI, IRBuilder<> &B) {
-  Function *Callee = CI->getCalledFunction();
-  // Verify the "strcat" function prototype.
-  FunctionType *FT = Callee->getFunctionType();
-  if (FT->getNumParams() != 2||
-      FT->getReturnType() != B.getInt8PtrTy() ||
-      FT->getParamType(0) != FT->getReturnType() ||
-      FT->getParamType(1) != FT->getReturnType())
-    return nullptr;
-
   // Extract some information from the instruction
   Value *Dst = CI->getArgOperand(0);
   Value *Src = CI->getArgOperand(1);
@@ -220,7 +129,7 @@ Value *LibCallSimplifier::emitStrLenMemCpy(Value *Src, Value *Dst, uint64_t Len,
                                            IRBuilder<> &B) {
   // We need to find the end of the destination string.  That's where the
   // memory is to be moved to. We just generate a call to strlen.
-  Value *DstLen = EmitStrLen(Dst, B, DL, TLI);
+  Value *DstLen = emitStrLen(Dst, B, DL, TLI);
   if (!DstLen)
     return nullptr;
 
@@ -238,15 +147,6 @@ Value *LibCallSimplifier::emitStrLenMemCpy(Value *Src, Value *Dst, uint64_t Len,
 }
 
 Value *LibCallSimplifier::optimizeStrNCat(CallInst *CI, IRBuilder<> &B) {
-  Function *Callee = CI->getCalledFunction();
-  // Verify the "strncat" function prototype.
-  FunctionType *FT = Callee->getFunctionType();
-  if (FT->getNumParams() != 3 || FT->getReturnType() != B.getInt8PtrTy() ||
-      FT->getParamType(0) != FT->getReturnType() ||
-      FT->getParamType(1) != FT->getReturnType() ||
-      !FT->getParamType(2)->isIntegerTy())
-    return nullptr;
-
   // Extract some information from the instruction.
   Value *Dst = CI->getArgOperand(0);
   Value *Src = CI->getArgOperand(1);
@@ -281,13 +181,7 @@ Value *LibCallSimplifier::optimizeStrNCat(CallInst *CI, IRBuilder<> &B) {
 
 Value *LibCallSimplifier::optimizeStrChr(CallInst *CI, IRBuilder<> &B) {
   Function *Callee = CI->getCalledFunction();
-  // Verify the "strchr" function prototype.
   FunctionType *FT = Callee->getFunctionType();
-  if (FT->getNumParams() != 2 || FT->getReturnType() != B.getInt8PtrTy() ||
-      FT->getParamType(0) != FT->getReturnType() ||
-      !FT->getParamType(1)->isIntegerTy(32))
-    return nullptr;
-
   Value *SrcStr = CI->getArgOperand(0);
 
   // If the second operand is non-constant, see if we can compute the length
@@ -298,7 +192,7 @@ Value *LibCallSimplifier::optimizeStrChr(CallInst *CI, IRBuilder<> &B) {
     if (Len == 0 || !FT->getParamType(1)->isIntegerTy(32)) // memchr needs i32.
       return nullptr;
 
-    return EmitMemChr(SrcStr, CI->getArgOperand(1), // include nul.
+    return emitMemChr(SrcStr, CI->getArgOperand(1), // include nul.
                       ConstantInt::get(DL.getIntPtrType(CI->getContext()), Len),
                       B, DL, TLI);
   }
@@ -308,7 +202,7 @@ Value *LibCallSimplifier::optimizeStrChr(CallInst *CI, IRBuilder<> &B) {
   StringRef Str;
   if (!getConstantStringInfo(SrcStr, Str)) {
     if (CharC->isZero()) // strchr(p, 0) -> p + strlen(p)
-      return B.CreateGEP(B.getInt8Ty(), SrcStr, EmitStrLen(SrcStr, B, DL, TLI),
+      return B.CreateGEP(B.getInt8Ty(), SrcStr, emitStrLen(SrcStr, B, DL, TLI),
                          "strchr");
     return nullptr;
   }
@@ -326,14 +220,6 @@ Value *LibCallSimplifier::optimizeStrChr(CallInst *CI, IRBuilder<> &B) {
 }
 
 Value *LibCallSimplifier::optimizeStrRChr(CallInst *CI, IRBuilder<> &B) {
-  Function *Callee = CI->getCalledFunction();
-  // Verify the "strrchr" function prototype.
-  FunctionType *FT = Callee->getFunctionType();
-  if (FT->getNumParams() != 2 || FT->getReturnType() != B.getInt8PtrTy() ||
-      FT->getParamType(0) != FT->getReturnType() ||
-      !FT->getParamType(1)->isIntegerTy(32))
-    return nullptr;
-
   Value *SrcStr = CI->getArgOperand(0);
   ConstantInt *CharC = dyn_cast<ConstantInt>(CI->getArgOperand(1));
 
@@ -345,7 +231,7 @@ Value *LibCallSimplifier::optimizeStrRChr(CallInst *CI, IRBuilder<> &B) {
   if (!getConstantStringInfo(SrcStr, Str)) {
     // strrchr(s, 0) -> strchr(s, 0)
     if (CharC->isZero())
-      return EmitStrChr(SrcStr, '\0', B, TLI);
+      return emitStrChr(SrcStr, '\0', B, TLI);
     return nullptr;
   }
 
@@ -361,14 +247,6 @@ Value *LibCallSimplifier::optimizeStrRChr(CallInst *CI, IRBuilder<> &B) {
 }
 
 Value *LibCallSimplifier::optimizeStrCmp(CallInst *CI, IRBuilder<> &B) {
-  Function *Callee = CI->getCalledFunction();
-  // Verify the "strcmp" function prototype.
-  FunctionType *FT = Callee->getFunctionType();
-  if (FT->getNumParams() != 2 || !FT->getReturnType()->isIntegerTy(32) ||
-      FT->getParamType(0) != FT->getParamType(1) ||
-      FT->getParamType(0) != B.getInt8PtrTy())
-    return nullptr;
-
   Value *Str1P = CI->getArgOperand(0), *Str2P = CI->getArgOperand(1);
   if (Str1P == Str2P) // strcmp(x,x)  -> 0
     return ConstantInt::get(CI->getType(), 0);
@@ -392,7 +270,7 @@ Value *LibCallSimplifier::optimizeStrCmp(CallInst *CI, IRBuilder<> &B) {
   uint64_t Len1 = GetStringLength(Str1P);
   uint64_t Len2 = GetStringLength(Str2P);
   if (Len1 && Len2) {
-    return EmitMemCmp(Str1P, Str2P,
+    return emitMemCmp(Str1P, Str2P,
                       ConstantInt::get(DL.getIntPtrType(CI->getContext()),
                                        std::min(Len1, Len2)),
                       B, DL, TLI);
@@ -402,15 +280,6 @@ Value *LibCallSimplifier::optimizeStrCmp(CallInst *CI, IRBuilder<> &B) {
 }
 
 Value *LibCallSimplifier::optimizeStrNCmp(CallInst *CI, IRBuilder<> &B) {
-  Function *Callee = CI->getCalledFunction();
-  // Verify the "strncmp" function prototype.
-  FunctionType *FT = Callee->getFunctionType();
-  if (FT->getNumParams() != 3 || !FT->getReturnType()->isIntegerTy(32) ||
-      FT->getParamType(0) != FT->getParamType(1) ||
-      FT->getParamType(0) != B.getInt8PtrTy() ||
-      !FT->getParamType(2)->isIntegerTy())
-    return nullptr;
-
   Value *Str1P = CI->getArgOperand(0), *Str2P = CI->getArgOperand(1);
   if (Str1P == Str2P) // strncmp(x,x,n)  -> 0
     return ConstantInt::get(CI->getType(), 0);
@@ -426,7 +295,7 @@ Value *LibCallSimplifier::optimizeStrNCmp(CallInst *CI, IRBuilder<> &B) {
     return ConstantInt::get(CI->getType(), 0);
 
   if (Length == 1) // strncmp(x,y,1) -> memcmp(x,y,1)
-    return EmitMemCmp(Str1P, Str2P, CI->getArgOperand(2), B, DL, TLI);
+    return emitMemCmp(Str1P, Str2P, CI->getArgOperand(2), B, DL, TLI);
 
   StringRef Str1, Str2;
   bool HasStr1 = getConstantStringInfo(Str1P, Str1);
@@ -450,11 +319,6 @@ Value *LibCallSimplifier::optimizeStrNCmp(CallInst *CI, IRBuilder<> &B) {
 }
 
 Value *LibCallSimplifier::optimizeStrCpy(CallInst *CI, IRBuilder<> &B) {
-  Function *Callee = CI->getCalledFunction();
-
-  if (!checkStringCopyLibFuncSignature(Callee, LibFunc::strcpy))
-    return nullptr;
-
   Value *Dst = CI->getArgOperand(0), *Src = CI->getArgOperand(1);
   if (Dst == Src) // strcpy(x,x)  -> x
     return Src;
@@ -473,12 +337,9 @@ Value *LibCallSimplifier::optimizeStrCpy(CallInst *CI, IRBuilder<> &B) {
 
 Value *LibCallSimplifier::optimizeStpCpy(CallInst *CI, IRBuilder<> &B) {
   Function *Callee = CI->getCalledFunction();
-  if (!checkStringCopyLibFuncSignature(Callee, LibFunc::stpcpy))
-    return nullptr;
-
   Value *Dst = CI->getArgOperand(0), *Src = CI->getArgOperand(1);
   if (Dst == Src) { // stpcpy(x,x)  -> x+strlen(x)
-    Value *StrLen = EmitStrLen(Src, B, DL, TLI);
+    Value *StrLen = emitStrLen(Src, B, DL, TLI);
     return StrLen ? B.CreateInBoundsGEP(B.getInt8Ty(), Dst, StrLen) : nullptr;
   }
 
@@ -500,9 +361,6 @@ Value *LibCallSimplifier::optimizeStpCpy(CallInst *CI, IRBuilder<> &B) {
 
 Value *LibCallSimplifier::optimizeStrNCpy(CallInst *CI, IRBuilder<> &B) {
   Function *Callee = CI->getCalledFunction();
-  if (!checkStringCopyLibFuncSignature(Callee, LibFunc::strncpy))
-    return nullptr;
-
   Value *Dst = CI->getArgOperand(0);
   Value *Src = CI->getArgOperand(1);
   Value *LenOp = CI->getArgOperand(2);
@@ -540,18 +398,63 @@ Value *LibCallSimplifier::optimizeStrNCpy(CallInst *CI, IRBuilder<> &B) {
 }
 
 Value *LibCallSimplifier::optimizeStrLen(CallInst *CI, IRBuilder<> &B) {
-  Function *Callee = CI->getCalledFunction();
-  FunctionType *FT = Callee->getFunctionType();
-  if (FT->getNumParams() != 1 || FT->getParamType(0) != B.getInt8PtrTy() ||
-      !FT->getReturnType()->isIntegerTy())
-    return nullptr;
-
   Value *Src = CI->getArgOperand(0);
 
   // Constant folding: strlen("xyz") -> 3
   if (uint64_t Len = GetStringLength(Src))
     return ConstantInt::get(CI->getType(), Len - 1);
 
+  // If s is a constant pointer pointing to a string literal, we can fold
+  // strlen(s + x) to strlen(s) - x, when x is known to be in the range 
+  // [0, strlen(s)] or the string has a single null terminator '\0' at the end.
+  // We only try to simplify strlen when the pointer s points to an array 
+  // of i8. Otherwise, we would need to scale the offset x before doing the
+  // subtraction. This will make the optimization more complex, and it's not 
+  // very useful because calling strlen for a pointer of other types is 
+  // very uncommon.
+  if (GEPOperator *GEP = dyn_cast<GEPOperator>(Src)) {
+    if (!isGEPBasedOnPointerToString(GEP))
+      return nullptr;
+
+    StringRef Str;
+    if (getConstantStringInfo(GEP->getOperand(0), Str, 0, false)) {
+      size_t NullTermIdx = Str.find('\0');
+      
+      // If the string does not have '\0', leave it to strlen to compute
+      // its length.
+      if (NullTermIdx == StringRef::npos)
+        return nullptr;
+     
+      Value *Offset = GEP->getOperand(2);
+      unsigned BitWidth = Offset->getType()->getIntegerBitWidth();
+      APInt KnownZero(BitWidth, 0);
+      APInt KnownOne(BitWidth, 0);
+      computeKnownBits(Offset, KnownZero, KnownOne, DL, 0, nullptr, CI, 
+                       nullptr);
+      KnownZero.flipAllBits();
+      size_t ArrSize = 
+             cast<ArrayType>(GEP->getSourceElementType())->getNumElements();
+
+      // KnownZero's bits are flipped, so zeros in KnownZero now represent 
+      // bits known to be zeros in Offset, and ones in KnowZero represent 
+      // bits unknown in Offset. Therefore, Offset is known to be in range
+      // [0, NullTermIdx] when the flipped KnownZero is non-negative and 
+      // unsigned-less-than NullTermIdx.
+      //
+      // If Offset is not provably in the range [0, NullTermIdx], we can still 
+      // optimize if we can prove that the program has undefined behavior when 
+      // Offset is outside that range. That is the case when GEP->getOperand(0) 
+      // is a pointer to an object whose memory extent is NullTermIdx+1.
+      if ((KnownZero.isNonNegative() && KnownZero.ule(NullTermIdx)) || 
+          (GEP->isInBounds() && isa<GlobalVariable>(GEP->getOperand(0)) &&
+           NullTermIdx == ArrSize - 1))
+        return B.CreateSub(ConstantInt::get(CI->getType(), NullTermIdx), 
+                           Offset);
+    }
+
+    return nullptr;
+  }
+
   // strlen(x?"foo":"bars") --> x ? 3 : 4
   if (SelectInst *SI = dyn_cast<SelectInst>(Src)) {
     uint64_t LenTrue = GetStringLength(SI->getTrueValue());
@@ -576,13 +479,6 @@ Value *LibCallSimplifier::optimizeStrLen(CallInst *CI, IRBuilder<> &B) {
 }
 
 Value *LibCallSimplifier::optimizeStrPBrk(CallInst *CI, IRBuilder<> &B) {
-  Function *Callee = CI->getCalledFunction();
-  FunctionType *FT = Callee->getFunctionType();
-  if (FT->getNumParams() != 2 || FT->getParamType(0) != B.getInt8PtrTy() ||
-      FT->getParamType(1) != FT->getParamType(0) ||
-      FT->getReturnType() != FT->getParamType(0))
-    return nullptr;
-
   StringRef S1, S2;
   bool HasS1 = getConstantStringInfo(CI->getArgOperand(0), S1);
   bool HasS2 = getConstantStringInfo(CI->getArgOperand(1), S2);
@@ -604,19 +500,12 @@ Value *LibCallSimplifier::optimizeStrPBrk(CallInst *CI, IRBuilder<> &B) {
 
   // strpbrk(s, "a") -> strchr(s, 'a')
   if (HasS2 && S2.size() == 1)
-    return EmitStrChr(CI->getArgOperand(0), S2[0], B, TLI);
+    return emitStrChr(CI->getArgOperand(0), S2[0], B, TLI);
 
   return nullptr;
 }
 
 Value *LibCallSimplifier::optimizeStrTo(CallInst *CI, IRBuilder<> &B) {
-  Function *Callee = CI->getCalledFunction();
-  FunctionType *FT = Callee->getFunctionType();
-  if ((FT->getNumParams() != 2 && FT->getNumParams() != 3) ||
-      !FT->getParamType(0)->isPointerTy() ||
-      !FT->getParamType(1)->isPointerTy())
-    return nullptr;
-
   Value *EndPtr = CI->getArgOperand(1);
   if (isa<ConstantPointerNull>(EndPtr)) {
     // With a null EndPtr, this function won't capture the main argument.
@@ -628,13 +517,6 @@ Value *LibCallSimplifier::optimizeStrTo(CallInst *CI, IRBuilder<> &B) {
 }
 
 Value *LibCallSimplifier::optimizeStrSpn(CallInst *CI, IRBuilder<> &B) {
-  Function *Callee = CI->getCalledFunction();
-  FunctionType *FT = Callee->getFunctionType();
-  if (FT->getNumParams() != 2 || FT->getParamType(0) != B.getInt8PtrTy() ||
-      FT->getParamType(1) != FT->getParamType(0) ||
-      !FT->getReturnType()->isIntegerTy())
-    return nullptr;
-
   StringRef S1, S2;
   bool HasS1 = getConstantStringInfo(CI->getArgOperand(0), S1);
   bool HasS2 = getConstantStringInfo(CI->getArgOperand(1), S2);
@@ -656,13 +538,6 @@ Value *LibCallSimplifier::optimizeStrSpn(CallInst *CI, IRBuilder<> &B) {
 }
 
 Value *LibCallSimplifier::optimizeStrCSpn(CallInst *CI, IRBuilder<> &B) {
-  Function *Callee = CI->getCalledFunction();
-  FunctionType *FT = Callee->getFunctionType();
-  if (FT->getNumParams() != 2 || FT->getParamType(0) != B.getInt8PtrTy() ||
-      FT->getParamType(1) != FT->getParamType(0) ||
-      !FT->getReturnType()->isIntegerTy())
-    return nullptr;
-
   StringRef S1, S2;
   bool HasS1 = getConstantStringInfo(CI->getArgOperand(0), S1);
   bool HasS2 = getConstantStringInfo(CI->getArgOperand(1), S2);
@@ -681,29 +556,22 @@ Value *LibCallSimplifier::optimizeStrCSpn(CallInst *CI, IRBuilder<> &B) {
 
   // strcspn(s, "") -> strlen(s)
   if (HasS2 && S2.empty())
-    return EmitStrLen(CI->getArgOperand(0), B, DL, TLI);
+    return emitStrLen(CI->getArgOperand(0), B, DL, TLI);
 
   return nullptr;
 }
 
 Value *LibCallSimplifier::optimizeStrStr(CallInst *CI, IRBuilder<> &B) {
-  Function *Callee = CI->getCalledFunction();
-  FunctionType *FT = Callee->getFunctionType();
-  if (FT->getNumParams() != 2 || !FT->getParamType(0)->isPointerTy() ||
-      !FT->getParamType(1)->isPointerTy() ||
-      !FT->getReturnType()->isPointerTy())
-    return nullptr;
-
   // fold strstr(x, x) -> x.
   if (CI->getArgOperand(0) == CI->getArgOperand(1))
     return B.CreateBitCast(CI->getArgOperand(0), CI->getType());
 
   // fold strstr(a, b) == a -> strncmp(a, b, strlen(b)) == 0
   if (isOnlyUsedInEqualityComparison(CI, CI->getArgOperand(0))) {
-    Value *StrLen = EmitStrLen(CI->getArgOperand(1), B, DL, TLI);
+    Value *StrLen = emitStrLen(CI->getArgOperand(1), B, DL, TLI);
     if (!StrLen)
       return nullptr;
-    Value *StrNCmp = EmitStrNCmp(CI->getArgOperand(0), CI->getArgOperand(1),
+    Value *StrNCmp = emitStrNCmp(CI->getArgOperand(0), CI->getArgOperand(1),
                                  StrLen, B, DL, TLI);
     if (!StrNCmp)
       return nullptr;
@@ -734,28 +602,20 @@ Value *LibCallSimplifier::optimizeStrStr(CallInst *CI, IRBuilder<> &B) {
       return Constant::getNullValue(CI->getType());
 
     // strstr("abcd", "bc") -> gep((char*)"abcd", 1)
-    Value *Result = CastToCStr(CI->getArgOperand(0), B);
+    Value *Result = castToCStr(CI->getArgOperand(0), B);
     Result = B.CreateConstInBoundsGEP1_64(Result, Offset, "strstr");
     return B.CreateBitCast(Result, CI->getType());
   }
 
   // fold strstr(x, "y") -> strchr(x, 'y').
   if (HasStr2 && ToFindStr.size() == 1) {
-    Value *StrChr = EmitStrChr(CI->getArgOperand(0), ToFindStr[0], B, TLI);
+    Value *StrChr = emitStrChr(CI->getArgOperand(0), ToFindStr[0], B, TLI);
     return StrChr ? B.CreateBitCast(StrChr, CI->getType()) : nullptr;
   }
   return nullptr;
 }
 
 Value *LibCallSimplifier::optimizeMemChr(CallInst *CI, IRBuilder<> &B) {
-  Function *Callee = CI->getCalledFunction();
-  FunctionType *FT = Callee->getFunctionType();
-  if (FT->getNumParams() != 3 || !FT->getParamType(0)->isPointerTy() ||
-      !FT->getParamType(1)->isIntegerTy(32) ||
-      !FT->getParamType(2)->isIntegerTy() ||
-      !FT->getReturnType()->isPointerTy())
-    return nullptr;
-
   Value *SrcStr = CI->getArgOperand(0);
   ConstantInt *CharC = dyn_cast<ConstantInt>(CI->getArgOperand(1));
   ConstantInt *LenC = dyn_cast<ConstantInt>(CI->getArgOperand(2));
@@ -834,13 +694,6 @@ Value *LibCallSimplifier::optimizeMemChr(CallInst *CI, IRBuilder<> &B) {
 }
 
 Value *LibCallSimplifier::optimizeMemCmp(CallInst *CI, IRBuilder<> &B) {
-  Function *Callee = CI->getCalledFunction();
-  FunctionType *FT = Callee->getFunctionType();
-  if (FT->getNumParams() != 3 || !FT->getParamType(0)->isPointerTy() ||
-      !FT->getParamType(1)->isPointerTy() ||
-      !FT->getReturnType()->isIntegerTy(32))
-    return nullptr;
-
   Value *LHS = CI->getArgOperand(0), *RHS = CI->getArgOperand(1);
 
   if (LHS == RHS) // memcmp(s,s,x) -> 0
@@ -857,9 +710,9 @@ Value *LibCallSimplifier::optimizeMemCmp(CallInst *CI, IRBuilder<> &B) {
 
   // memcmp(S1,S2,1) -> *(unsigned char*)LHS - *(unsigned char*)RHS
   if (Len == 1) {
-    Value *LHSV = B.CreateZExt(B.CreateLoad(CastToCStr(LHS, B), "lhsc"),
+    Value *LHSV = B.CreateZExt(B.CreateLoad(castToCStr(LHS, B), "lhsc"),
                                CI->getType(), "lhsv");
-    Value *RHSV = B.CreateZExt(B.CreateLoad(CastToCStr(RHS, B), "rhsc"),
+    Value *RHSV = B.CreateZExt(B.CreateLoad(castToCStr(RHS, B), "rhsc"),
                                CI->getType(), "rhsv");
     return B.CreateSub(LHSV, RHSV, "chardiff");
   }
@@ -909,11 +762,6 @@ Value *LibCallSimplifier::optimizeMemCmp(CallInst *CI, IRBuilder<> &B) {
 }
 
 Value *LibCallSimplifier::optimizeMemCpy(CallInst *CI, IRBuilder<> &B) {
-  Function *Callee = CI->getCalledFunction();
-
-  if (!checkStringCopyLibFuncSignature(Callee, LibFunc::memcpy))
-    return nullptr;
-
   // memcpy(x, y, n) -> llvm.memcpy(x, y, n, 1)
   B.CreateMemCpy(CI->getArgOperand(0), CI->getArgOperand(1),
                  CI->getArgOperand(2), 1);
@@ -921,23 +769,81 @@ Value *LibCallSimplifier::optimizeMemCpy(CallInst *CI, IRBuilder<> &B) {
 }
 
 Value *LibCallSimplifier::optimizeMemMove(CallInst *CI, IRBuilder<> &B) {
-  Function *Callee = CI->getCalledFunction();
-
-  if (!checkStringCopyLibFuncSignature(Callee, LibFunc::memmove))
-    return nullptr;
-
   // memmove(x, y, n) -> llvm.memmove(x, y, n, 1)
   B.CreateMemMove(CI->getArgOperand(0), CI->getArgOperand(1),
                   CI->getArgOperand(2), 1);
   return CI->getArgOperand(0);
 }
 
-Value *LibCallSimplifier::optimizeMemSet(CallInst *CI, IRBuilder<> &B) {
-  Function *Callee = CI->getCalledFunction();
+// TODO: Does this belong in BuildLibCalls or should all of those similar
+// functions be moved here?
+static Value *emitCalloc(Value *Num, Value *Size, const AttributeSet &Attrs,
+                         IRBuilder<> &B, const TargetLibraryInfo &TLI) {
+  LibFunc::Func Func;
+  if (!TLI.getLibFunc("calloc", Func) || !TLI.has(Func))
+    return nullptr;
+
+  Module *M = B.GetInsertBlock()->getModule();
+  const DataLayout &DL = M->getDataLayout();
+  IntegerType *PtrType = DL.getIntPtrType((B.GetInsertBlock()->getContext()));
+  Value *Calloc = M->getOrInsertFunction("calloc", Attrs, B.getInt8PtrTy(),
+                                         PtrType, PtrType, nullptr);
+  CallInst *CI = B.CreateCall(Calloc, { Num, Size }, "calloc");
+
+  if (const auto *F = dyn_cast<Function>(Calloc->stripPointerCasts()))
+    CI->setCallingConv(F->getCallingConv());
+
+  return CI;
+}
 
-  if (!checkStringCopyLibFuncSignature(Callee, LibFunc::memset))
+/// Fold memset[_chk](malloc(n), 0, n) --> calloc(1, n).
+static Value *foldMallocMemset(CallInst *Memset, IRBuilder<> &B,
+                               const TargetLibraryInfo &TLI) {
+  // This has to be a memset of zeros (bzero).
+  auto *FillValue = dyn_cast<ConstantInt>(Memset->getArgOperand(1));
+  if (!FillValue || FillValue->getZExtValue() != 0)
     return nullptr;
 
+  // TODO: We should handle the case where the malloc has more than one use.
+  // This is necessary to optimize common patterns such as when the result of
+  // the malloc is checked against null or when a memset intrinsic is used in
+  // place of a memset library call.
+  auto *Malloc = dyn_cast<CallInst>(Memset->getArgOperand(0));
+  if (!Malloc || !Malloc->hasOneUse())
+    return nullptr;
+
+  // Is the inner call really malloc()?
+  Function *InnerCallee = Malloc->getCalledFunction();
+  LibFunc::Func Func;
+  if (!TLI.getLibFunc(*InnerCallee, Func) || !TLI.has(Func) ||
+      Func != LibFunc::malloc)
+    return nullptr;
+
+  // The memset must cover the same number of bytes that are malloc'd.
+  if (Memset->getArgOperand(2) != Malloc->getArgOperand(0))
+    return nullptr;
+
+  // Replace the malloc with a calloc. We need the data layout to know what the
+  // actual size of a 'size_t' parameter is. 
+  B.SetInsertPoint(Malloc->getParent(), ++Malloc->getIterator());
+  const DataLayout &DL = Malloc->getModule()->getDataLayout();
+  IntegerType *SizeType = DL.getIntPtrType(B.GetInsertBlock()->getContext());
+  Value *Calloc = emitCalloc(ConstantInt::get(SizeType, 1),
+                             Malloc->getArgOperand(0), Malloc->getAttributes(),
+                             B, TLI);
+  if (!Calloc)
+    return nullptr;
+
+  Malloc->replaceAllUsesWith(Calloc);
+  Malloc->eraseFromParent();
+
+  return Calloc;
+}
+
+Value *LibCallSimplifier::optimizeMemSet(CallInst *CI, IRBuilder<> &B) {
+  if (auto *Calloc = foldMallocMemset(CI, B, *TLI))
+    return Calloc;
+
   // memset(p, v, n) -> llvm.memset(p, v, n, 1)
   Value *Val = B.CreateIntCast(CI->getArgOperand(1), B.getInt8Ty(), false);
   B.CreateMemSet(CI->getArgOperand(0), Val, CI->getArgOperand(2), 1);
@@ -970,34 +876,12 @@ static Value *valueHasFloatPrecision(Value *Val) {
   return nullptr;
 }
 
-/// Any floating-point library function that we're trying to simplify will have
-/// a signature of the form: fptype foo(fptype param1, fptype param2, ...).
-/// CheckDoubleTy indicates that 'fptype' must be 'double'.
-static bool matchesFPLibFunctionSignature(const Function *F, unsigned NumParams,
-                                          bool CheckDoubleTy) {
-  FunctionType *FT = F->getFunctionType();
-  if (FT->getNumParams() != NumParams)
-    return false;
-
-  // The return type must match what we're looking for.
-  Type *RetTy = FT->getReturnType();
-  if (CheckDoubleTy ? !RetTy->isDoubleTy() : !RetTy->isFloatingPointTy())
-    return false;
-
-  // Each parameter must match the return type, and therefore, match every other
-  // parameter too.
-  for (const Type *ParamTy : FT->params())
-    if (ParamTy != RetTy)
-      return false;
-
-  return true;
-}
-
 /// Shrink double -> float for unary functions like 'floor'.
 static Value *optimizeUnaryDoubleFP(CallInst *CI, IRBuilder<> &B,
                                     bool CheckRetType) {
   Function *Callee = CI->getCalledFunction();
-  if (!matchesFPLibFunctionSignature(Callee, 1, true))
+  // We know this libcall has a valid prototype, but we don't know which.
+  if (!CI->getType()->isDoubleTy())
     return nullptr;
 
   if (CheckRetType) {
@@ -1026,7 +910,7 @@ static Value *optimizeUnaryDoubleFP(CallInst *CI, IRBuilder<> &B,
     V = B.CreateCall(F, V);
   } else {
     // The call is a library call rather than an intrinsic.
-    V = EmitUnaryFloatFnCall(V, Callee->getName(), B, Callee->getAttributes());
+    V = emitUnaryFloatFnCall(V, Callee->getName(), B, Callee->getAttributes());
   }
 
   return B.CreateFPExt(V, B.getDoubleTy());
@@ -1035,7 +919,8 @@ static Value *optimizeUnaryDoubleFP(CallInst *CI, IRBuilder<> &B,
 /// Shrink double -> float for binary functions like 'fmin/fmax'.
 static Value *optimizeBinaryDoubleFP(CallInst *CI, IRBuilder<> &B) {
   Function *Callee = CI->getCalledFunction();
-  if (!matchesFPLibFunctionSignature(Callee, 2, true))
+  // We know this libcall has a valid prototype, but we don't know which.
+  if (!CI->getType()->isDoubleTy())
     return nullptr;
 
   // If this is something like 'fmin((double)floatval1, (double)floatval2)',
@@ -1054,7 +939,7 @@ static Value *optimizeBinaryDoubleFP(CallInst *CI, IRBuilder<> &B) {
   // fmin((double)floatval1, (double)floatval2)
   //                      -> (double)fminf(floatval1, floatval2)
   // TODO: Handle intrinsics in the same way as in optimizeUnaryDoubleFP().
-  Value *V = EmitBinaryFloatFnCall(V1, V2, Callee->getName(), B,
+  Value *V = emitBinaryFloatFnCall(V1, V2, Callee->getName(), B,
                                    Callee->getAttributes());
   return B.CreateFPExt(V, B.getDoubleTy());
 }
@@ -1066,13 +951,6 @@ Value *LibCallSimplifier::optimizeCos(CallInst *CI, IRBuilder<> &B) {
   if (UnsafeFPShrink && Name == "cos" && hasFloatVersion(Name))
     Ret = optimizeUnaryDoubleFP(CI, B, true);
 
-  FunctionType *FT = Callee->getFunctionType();
-  // Just make sure this has 1 argument of FP type, which matches the
-  // result type.
-  if (FT->getNumParams() != 1 || FT->getReturnType() != FT->getParamType(0) ||
-      !FT->getParamType(0)->isFloatingPointTy())
-    return Ret;
-
   // cos(-x) -> cos(x)
   Value *Op1 = CI->getArgOperand(0);
   if (BinaryOperator::isFNeg(Op1)) {
@@ -1114,14 +992,6 @@ Value *LibCallSimplifier::optimizePow(CallInst *CI, IRBuilder<> &B) {
   if (UnsafeFPShrink && Name == "pow" && hasFloatVersion(Name))
     Ret = optimizeUnaryDoubleFP(CI, B, true);
 
-  FunctionType *FT = Callee->getFunctionType();
-  // Just make sure this has 2 arguments of the same FP type, which match the
-  // result type.
-  if (FT->getNumParams() != 2 || FT->getReturnType() != FT->getParamType(0) ||
-      FT->getParamType(0) != FT->getParamType(1) ||
-      !FT->getParamType(0)->isFloatingPointTy())
-    return Ret;
-
   Value *Op1 = CI->getArgOperand(0), *Op2 = CI->getArgOperand(1);
   if (ConstantFP *Op1C = dyn_cast<ConstantFP>(Op1)) {
     // pow(1.0, x) -> 1.0
@@ -1131,19 +1001,16 @@ Value *LibCallSimplifier::optimizePow(CallInst *CI, IRBuilder<> &B) {
     if (Op1C->isExactlyValue(2.0) &&
         hasUnaryFloatFn(TLI, Op1->getType(), LibFunc::exp2, LibFunc::exp2f,
                         LibFunc::exp2l))
-      return EmitUnaryFloatFnCall(Op2, TLI->getName(LibFunc::exp2), B,
+      return emitUnaryFloatFnCall(Op2, TLI->getName(LibFunc::exp2), B,
                                   Callee->getAttributes());
     // pow(10.0, x) -> exp10(x)
     if (Op1C->isExactlyValue(10.0) &&
         hasUnaryFloatFn(TLI, Op1->getType(), LibFunc::exp10, LibFunc::exp10f,
                         LibFunc::exp10l))
-      return EmitUnaryFloatFnCall(Op2, TLI->getName(LibFunc::exp10), B,
+      return emitUnaryFloatFnCall(Op2, TLI->getName(LibFunc::exp10), B,
                                   Callee->getAttributes());
   }
 
-  // FIXME: Use instruction-level FMF.
-  bool UnsafeFPMath = canUseUnsafeFPMath(CI->getParent()->getParent());
-
   // pow(exp(x), y) -> exp(x * y)
   // pow(exp2(x), y) -> exp2(x * y)
   // We enable these only with fast-math. Besides rounding differences, the
@@ -1159,7 +1026,7 @@ Value *LibCallSimplifier::optimizePow(CallInst *CI, IRBuilder<> &B) {
       IRBuilder<>::FastMathFlagGuard Guard(B);
       B.setFastMathFlags(CI->getFastMathFlags());
       Value *FMul = B.CreateFMul(OpC->getArgOperand(0), Op2, "mul");
-      return EmitUnaryFloatFnCall(FMul, OpCCallee->getName(), B,
+      return emitUnaryFloatFnCall(FMul, OpCCallee->getName(), B,
                                   OpCCallee->getAttributes());
     }
   }
@@ -1181,7 +1048,7 @@ Value *LibCallSimplifier::optimizePow(CallInst *CI, IRBuilder<> &B) {
     if (CI->hasUnsafeAlgebra()) {
       IRBuilder<>::FastMathFlagGuard Guard(B);
       B.setFastMathFlags(CI->getFastMathFlags());
-      return EmitUnaryFloatFnCall(Op1, TLI->getName(LibFunc::sqrt), B,
+      return emitUnaryFloatFnCall(Op1, TLI->getName(LibFunc::sqrt), B,
                                   Callee->getAttributes());
     }
 
@@ -1191,9 +1058,9 @@ Value *LibCallSimplifier::optimizePow(CallInst *CI, IRBuilder<> &B) {
     // TODO: In finite-only mode, this could be just fabs(sqrt(x)).
     Value *Inf = ConstantFP::getInfinity(CI->getType());
     Value *NegInf = ConstantFP::getInfinity(CI->getType(), true);
-    Value *Sqrt = EmitUnaryFloatFnCall(Op1, "sqrt", B, Callee->getAttributes());
+    Value *Sqrt = emitUnaryFloatFnCall(Op1, "sqrt", B, Callee->getAttributes());
     Value *FAbs =
-        EmitUnaryFloatFnCall(Sqrt, "fabs", B, Callee->getAttributes());
+        emitUnaryFloatFnCall(Sqrt, "fabs", B, Callee->getAttributes());
     Value *FCmp = B.CreateFCmpOEQ(Op1, NegInf);
     Value *Sel = B.CreateSelect(FCmp, Inf, FAbs);
     return Sel;
@@ -1207,7 +1074,7 @@ Value *LibCallSimplifier::optimizePow(CallInst *CI, IRBuilder<> &B) {
     return B.CreateFDiv(ConstantFP::get(CI->getType(), 1.0), Op1, "powrecip");
 
   // In -ffast-math, generate repeated fmul instead of generating pow(x, n).
-  if (UnsafeFPMath) {
+  if (CI->hasUnsafeAlgebra()) {
     APFloat V = abs(Op2C->getValueAPF());
     // We limit to a max of 7 fmul(s). Thus max exponent is 32.
     // This transformation applies to integer exponents only.
@@ -1224,6 +1091,8 @@ Value *LibCallSimplifier::optimizePow(CallInst *CI, IRBuilder<> &B) {
     // So we first convert V to something which could be converted to double.
     bool ignored;
     V.convert(APFloat::IEEEdouble, APFloat::rmTowardZero, &ignored);
+    
+    // TODO: Should the new instructions propagate the 'fast' flag of the pow()?
     Value *FMul = getPow(InnerChain, V.convertToDouble(), B);
     // For negative exponents simply compute the reciprocal.
     if (Op2C->isNegative())
@@ -1236,19 +1105,11 @@ Value *LibCallSimplifier::optimizePow(CallInst *CI, IRBuilder<> &B) {
 
 Value *LibCallSimplifier::optimizeExp2(CallInst *CI, IRBuilder<> &B) {
   Function *Callee = CI->getCalledFunction();
-  Function *Caller = CI->getParent()->getParent();
   Value *Ret = nullptr;
   StringRef Name = Callee->getName();
   if (UnsafeFPShrink && Name == "exp2" && hasFloatVersion(Name))
     Ret = optimizeUnaryDoubleFP(CI, B, true);
 
-  FunctionType *FT = Callee->getFunctionType();
-  // Just make sure this has 1 argument of FP type, which matches the
-  // result type.
-  if (FT->getNumParams() != 1 || FT->getReturnType() != FT->getParamType(0) ||
-      !FT->getParamType(0)->isFloatingPointTy())
-    return Ret;
-
   Value *Op = CI->getArgOperand(0);
   // Turn exp2(sitofp(x)) -> ldexp(1.0, sext(x))  if sizeof(x) <= 32
   // Turn exp2(uitofp(x)) -> ldexp(1.0, zext(x))  if sizeof(x) < 32
@@ -1273,11 +1134,11 @@ Value *LibCallSimplifier::optimizeExp2(CallInst *CI, IRBuilder<> &B) {
       if (!Op->getType()->isFloatTy())
         One = ConstantExpr::getFPExtend(One, Op->getType());
 
-      Module *M = Caller->getParent();
-      Value *Callee =
+      Module *M = CI->getModule();
+      Value *NewCallee =
           M->getOrInsertFunction(TLI->getName(LdExp), Op->getType(),
                                  Op->getType(), B.getInt32Ty(), nullptr);
-      CallInst *CI = B.CreateCall(Callee, {One, LdExpArg});
+      CallInst *CI = B.CreateCall(NewCallee, {One, LdExpArg});
       if (const Function *F = dyn_cast<Function>(Callee->stripPointerCasts()))
         CI->setCallingConv(F->getCallingConv());
 
@@ -1294,12 +1155,6 @@ Value *LibCallSimplifier::optimizeFabs(CallInst *CI, IRBuilder<> &B) {
   if (Name == "fabs" && hasFloatVersion(Name))
     Ret = optimizeUnaryDoubleFP(CI, B, false);
 
-  FunctionType *FT = Callee->getFunctionType();
-  // Make sure this has 1 argument of FP type which matches the result type.
-  if (FT->getNumParams() != 1 || FT->getReturnType() != FT->getParamType(0) ||
-      !FT->getParamType(0)->isFloatingPointTy())
-    return Ret;
-
   Value *Op = CI->getArgOperand(0);
   if (Instruction *I = dyn_cast<Instruction>(Op)) {
     // Fold fabs(x * x) -> x * x; any squared FP value must already be positive.
@@ -1311,21 +1166,14 @@ Value *LibCallSimplifier::optimizeFabs(CallInst *CI, IRBuilder<> &B) {
 }
 
 Value *LibCallSimplifier::optimizeFMinFMax(CallInst *CI, IRBuilder<> &B) {
+  Function *Callee = CI->getCalledFunction();
   // If we can shrink the call to a float function rather than a double
   // function, do that first.
-  Function *Callee = CI->getCalledFunction();
   StringRef Name = Callee->getName();
   if ((Name == "fmin" || Name == "fmax") && hasFloatVersion(Name))
     if (Value *Ret = optimizeBinaryDoubleFP(CI, B))
       return Ret;
 
-  // Make sure this has 2 arguments of FP type which match the result type.
-  FunctionType *FT = Callee->getFunctionType();
-  if (FT->getNumParams() != 2 || FT->getReturnType() != FT->getParamType(0) ||
-      FT->getParamType(0) != FT->getParamType(1) ||
-      !FT->getParamType(0)->isFloatingPointTy())
-    return nullptr;
-
   IRBuilder<>::FastMathFlagGuard Guard(B);
   FastMathFlags FMF;
   if (CI->hasUnsafeAlgebra()) {
@@ -1360,13 +1208,6 @@ Value *LibCallSimplifier::optimizeLog(CallInst *CI, IRBuilder<> &B) {
   StringRef Name = Callee->getName();
   if (UnsafeFPShrink && hasFloatVersion(Name))
     Ret = optimizeUnaryDoubleFP(CI, B, true);
-  FunctionType *FT = Callee->getFunctionType();
-
-  // Just make sure this has 1 argument of FP type, which matches the
-  // result type.
-  if (FT->getNumParams() != 1 || FT->getReturnType() != FT->getParamType(0) ||
-      !FT->getParamType(0)->isFloatingPointTy())
-    return Ret;
 
   if (!CI->hasUnsafeAlgebra())
     return Ret;
@@ -1392,7 +1233,7 @@ Value *LibCallSimplifier::optimizeLog(CallInst *CI, IRBuilder<> &B) {
   if (F && ((TLI->getLibFunc(F->getName(), Func) && TLI->has(Func) &&
       Func == LibFunc::pow) || F->getIntrinsicID() == Intrinsic::pow))
     return B.CreateFMul(OpC->getArgOperand(1),
-      EmitUnaryFloatFnCall(OpC->getOperand(0), Callee->getName(), B,
+      emitUnaryFloatFnCall(OpC->getOperand(0), Callee->getName(), B,
                            Callee->getAttributes()), "mul");
 
   // log(exp2(y)) -> y*log(2)
@@ -1400,7 +1241,7 @@ Value *LibCallSimplifier::optimizeLog(CallInst *CI, IRBuilder<> &B) {
       TLI->has(Func) && Func == LibFunc::exp2)
     return B.CreateFMul(
         OpC->getArgOperand(0),
-        EmitUnaryFloatFnCall(ConstantFP::get(CI->getType(), 2.0),
+        emitUnaryFloatFnCall(ConstantFP::get(CI->getType(), 2.0),
                              Callee->getName(), B, Callee->getAttributes()),
         "logmul");
   return Ret;
@@ -1408,21 +1249,11 @@ Value *LibCallSimplifier::optimizeLog(CallInst *CI, IRBuilder<> &B) {
 
 Value *LibCallSimplifier::optimizeSqrt(CallInst *CI, IRBuilder<> &B) {
   Function *Callee = CI->getCalledFunction();
-
   Value *Ret = nullptr;
   if (TLI->has(LibFunc::sqrtf) && (Callee->getName() == "sqrt" ||
                                    Callee->getIntrinsicID() == Intrinsic::sqrt))
     Ret = optimizeUnaryDoubleFP(CI, B, true);
 
-  // FIXME: Refactor - this check is repeated all over this file and even in the
-  // preceding call to shrink double -> float.
-
-  // Make sure this has 1 argument of FP type, which matches the result type.
-  FunctionType *FT = Callee->getFunctionType();
-  if (FT->getNumParams() != 1 || FT->getReturnType() != FT->getParamType(0) ||
-      !FT->getParamType(0)->isFloatingPointTy())
-    return Ret;
-
   if (!CI->hasUnsafeAlgebra())
     return Ret;
 
@@ -1489,13 +1320,6 @@ Value *LibCallSimplifier::optimizeTan(CallInst *CI, IRBuilder<> &B) {
   StringRef Name = Callee->getName();
   if (UnsafeFPShrink && Name == "tan" && hasFloatVersion(Name))
     Ret = optimizeUnaryDoubleFP(CI, B, true);
-  FunctionType *FT = Callee->getFunctionType();
-
-  // Just make sure this has 1 argument of FP type, which matches the
-  // result type.
-  if (FT->getNumParams() != 1 || FT->getReturnType() != FT->getParamType(0) ||
-      !FT->getParamType(0)->isFloatingPointTy())
-    return Ret;
 
   Value *Op1 = CI->getArgOperand(0);
   auto *OpC = dyn_cast<CallInst>(Op1);
@@ -1519,13 +1343,65 @@ Value *LibCallSimplifier::optimizeTan(CallInst *CI, IRBuilder<> &B) {
   return Ret;
 }
 
-static bool isTrigLibCall(CallInst *CI);
+static bool isTrigLibCall(CallInst *CI) {
+  // We can only hope to do anything useful if we can ignore things like errno
+  // and floating-point exceptions.
+  // We already checked the prototype.
+  return CI->hasFnAttr(Attribute::NoUnwind) &&
+         CI->hasFnAttr(Attribute::ReadNone);
+}
+
 static void insertSinCosCall(IRBuilder<> &B, Function *OrigCallee, Value *Arg,
                              bool UseFloat, Value *&Sin, Value *&Cos,
-                             Value *&SinCos);
+                             Value *&SinCos) {
+  Type *ArgTy = Arg->getType();
+  Type *ResTy;
+  StringRef Name;
 
-Value *LibCallSimplifier::optimizeSinCosPi(CallInst *CI, IRBuilder<> &B) {
+  Triple T(OrigCallee->getParent()->getTargetTriple());
+  if (UseFloat) {
+    Name = "__sincospif_stret";
+
+    assert(T.getArch() != Triple::x86 && "x86 messy and unsupported for now");
+    // x86_64 can't use {float, float} since that would be returned in both
+    // xmm0 and xmm1, which isn't what a real struct would do.
+    ResTy = T.getArch() == Triple::x86_64
+    ? static_cast<Type *>(VectorType::get(ArgTy, 2))
+    : static_cast<Type *>(StructType::get(ArgTy, ArgTy, nullptr));
+  } else {
+    Name = "__sincospi_stret";
+    ResTy = StructType::get(ArgTy, ArgTy, nullptr);
+  }
+
+  Module *M = OrigCallee->getParent();
+  Value *Callee = M->getOrInsertFunction(Name, OrigCallee->getAttributes(),
+                                         ResTy, ArgTy, nullptr);
+
+  if (Instruction *ArgInst = dyn_cast<Instruction>(Arg)) {
+    // If the argument is an instruction, it must dominate all uses so put our
+    // sincos call there.
+    B.SetInsertPoint(ArgInst->getParent(), ++ArgInst->getIterator());
+  } else {
+    // Otherwise (e.g. for a constant) the beginning of the function is as
+    // good a place as any.
+    BasicBlock &EntryBB = B.GetInsertBlock()->getParent()->getEntryBlock();
+    B.SetInsertPoint(&EntryBB, EntryBB.begin());
+  }
+
+  SinCos = B.CreateCall(Callee, Arg, "sincospi");
+
+  if (SinCos->getType()->isStructTy()) {
+    Sin = B.CreateExtractValue(SinCos, 0, "sinpi");
+    Cos = B.CreateExtractValue(SinCos, 1, "cospi");
+  } else {
+    Sin = B.CreateExtractElement(SinCos, ConstantInt::get(B.getInt32Ty(), 0),
+                                 "sinpi");
+    Cos = B.CreateExtractElement(SinCos, ConstantInt::get(B.getInt32Ty(), 1),
+                                 "cospi");
+  }
+}
 
+Value *LibCallSimplifier::optimizeSinCosPi(CallInst *CI, IRBuilder<> &B) {
   // Make sure the prototype is as expected, otherwise the rest of the
   // function is probably invalid and likely to abort.
   if (!isTrigLibCall(CI))
@@ -1541,9 +1417,9 @@ Value *LibCallSimplifier::optimizeSinCosPi(CallInst *CI, IRBuilder<> &B) {
   // Look for all compatible sinpi, cospi and sincospi calls with the same
   // argument. If there are enough (in some sense) we can make the
   // substitution.
+  Function *F = CI->getFunction();
   for (User *U : Arg->users())
-    classifyArgUse(U, CI->getParent(), IsFloat, SinCalls, CosCalls,
-                   SinCosCalls);
+    classifyArgUse(U, F, IsFloat, SinCalls, CosCalls, SinCosCalls);
 
   // It's only worthwhile if both sinpi and cospi are actually used.
   if (SinCosCalls.empty() && (SinCalls.empty() || CosCalls.empty()))
@@ -1559,35 +1435,23 @@ Value *LibCallSimplifier::optimizeSinCosPi(CallInst *CI, IRBuilder<> &B) {
   return nullptr;
 }
 
-static bool isTrigLibCall(CallInst *CI) {
-  Function *Callee = CI->getCalledFunction();
-  FunctionType *FT = Callee->getFunctionType();
-
-  // We can only hope to do anything useful if we can ignore things like errno
-  // and floating-point exceptions.
-  bool AttributesSafe =
-      CI->hasFnAttr(Attribute::NoUnwind) && CI->hasFnAttr(Attribute::ReadNone);
-
-  // Other than that we need float(float) or double(double)
-  return AttributesSafe && FT->getNumParams() == 1 &&
-         FT->getReturnType() == FT->getParamType(0) &&
-         (FT->getParamType(0)->isFloatTy() ||
-          FT->getParamType(0)->isDoubleTy());
-}
-
-void
-LibCallSimplifier::classifyArgUse(Value *Val, BasicBlock *BB, bool IsFloat,
-                                  SmallVectorImpl<CallInst *> &SinCalls,
-                                  SmallVectorImpl<CallInst *> &CosCalls,
-                                  SmallVectorImpl<CallInst *> &SinCosCalls) {
+void LibCallSimplifier::classifyArgUse(
+    Value *Val, Function *F, bool IsFloat,
+    SmallVectorImpl<CallInst *> &SinCalls,
+    SmallVectorImpl<CallInst *> &CosCalls,
+    SmallVectorImpl<CallInst *> &SinCosCalls) {
   CallInst *CI = dyn_cast<CallInst>(Val);
 
   if (!CI)
     return;
 
+  // Don't consider calls in other functions.
+  if (CI->getFunction() != F)
+    return;
+
   Function *Callee = CI->getCalledFunction();
   LibFunc::Func Func;
-  if (!Callee || !TLI->getLibFunc(Callee->getName(), Func) || !TLI->has(Func) ||
+  if (!Callee || !TLI->getLibFunc(*Callee, Func) || !TLI->has(Func) ||
       !isTrigLibCall(CI))
     return;
 
@@ -1614,69 +1478,12 @@ void LibCallSimplifier::replaceTrigInsts(SmallVectorImpl<CallInst *> &Calls,
     replaceAllUsesWith(C, Res);
 }
 
-void insertSinCosCall(IRBuilder<> &B, Function *OrigCallee, Value *Arg,
-                      bool UseFloat, Value *&Sin, Value *&Cos, Value *&SinCos) {
-  Type *ArgTy = Arg->getType();
-  Type *ResTy;
-  StringRef Name;
-
-  Triple T(OrigCallee->getParent()->getTargetTriple());
-  if (UseFloat) {
-    Name = "__sincospif_stret";
-
-    assert(T.getArch() != Triple::x86 && "x86 messy and unsupported for now");
-    // x86_64 can't use {float, float} since that would be returned in both
-    // xmm0 and xmm1, which isn't what a real struct would do.
-    ResTy = T.getArch() == Triple::x86_64
-                ? static_cast<Type *>(VectorType::get(ArgTy, 2))
-                : static_cast<Type *>(StructType::get(ArgTy, ArgTy, nullptr));
-  } else {
-    Name = "__sincospi_stret";
-    ResTy = StructType::get(ArgTy, ArgTy, nullptr);
-  }
-
-  Module *M = OrigCallee->getParent();
-  Value *Callee = M->getOrInsertFunction(Name, OrigCallee->getAttributes(),
-                                         ResTy, ArgTy, nullptr);
-
-  if (Instruction *ArgInst = dyn_cast<Instruction>(Arg)) {
-    // If the argument is an instruction, it must dominate all uses so put our
-    // sincos call there.
-    B.SetInsertPoint(ArgInst->getParent(), ++ArgInst->getIterator());
-  } else {
-    // Otherwise (e.g. for a constant) the beginning of the function is as
-    // good a place as any.
-    BasicBlock &EntryBB = B.GetInsertBlock()->getParent()->getEntryBlock();
-    B.SetInsertPoint(&EntryBB, EntryBB.begin());
-  }
-
-  SinCos = B.CreateCall(Callee, Arg, "sincospi");
-
-  if (SinCos->getType()->isStructTy()) {
-    Sin = B.CreateExtractValue(SinCos, 0, "sinpi");
-    Cos = B.CreateExtractValue(SinCos, 1, "cospi");
-  } else {
-    Sin = B.CreateExtractElement(SinCos, ConstantInt::get(B.getInt32Ty(), 0),
-                                 "sinpi");
-    Cos = B.CreateExtractElement(SinCos, ConstantInt::get(B.getInt32Ty(), 1),
-                                 "cospi");
-  }
-}
-
 //===----------------------------------------------------------------------===//
 // Integer Library Call Optimizations
 //===----------------------------------------------------------------------===//
 
-static bool checkIntUnaryReturnAndParam(Function *Callee) {
-  FunctionType *FT = Callee->getFunctionType();
-  return FT->getNumParams() == 1 && FT->getReturnType()->isIntegerTy(32) &&
-    FT->getParamType(0)->isIntegerTy();
-}
-
 Value *LibCallSimplifier::optimizeFFS(CallInst *CI, IRBuilder<> &B) {
   Function *Callee = CI->getCalledFunction();
-  if (!checkIntUnaryReturnAndParam(Callee))
-    return nullptr;
   Value *Op = CI->getArgOperand(0);
 
   // Constant fold.
@@ -1700,13 +1507,6 @@ Value *LibCallSimplifier::optimizeFFS(CallInst *CI, IRBuilder<> &B) {
 }
 
 Value *LibCallSimplifier::optimizeAbs(CallInst *CI, IRBuilder<> &B) {
-  Function *Callee = CI->getCalledFunction();
-  FunctionType *FT = Callee->getFunctionType();
-  // We require integer(integer) where the types agree.
-  if (FT->getNumParams() != 1 || !FT->getReturnType()->isIntegerTy() ||
-      FT->getParamType(0) != FT->getReturnType())
-    return nullptr;
-
   // abs(x) -> x >s -1 ? x : -x
   Value *Op = CI->getArgOperand(0);
   Value *Pos =
@@ -1716,9 +1516,6 @@ Value *LibCallSimplifier::optimizeAbs(CallInst *CI, IRBuilder<> &B) {
 }
 
 Value *LibCallSimplifier::optimizeIsDigit(CallInst *CI, IRBuilder<> &B) {
-  if (!checkIntUnaryReturnAndParam(CI->getCalledFunction()))
-    return nullptr;
-
   // isdigit(c) -> (c-'0') <u 10
   Value *Op = CI->getArgOperand(0);
   Op = B.CreateSub(Op, B.getInt32('0'), "isdigittmp");
@@ -1727,9 +1524,6 @@ Value *LibCallSimplifier::optimizeIsDigit(CallInst *CI, IRBuilder<> &B) {
 }
 
 Value *LibCallSimplifier::optimizeIsAscii(CallInst *CI, IRBuilder<> &B) {
-  if (!checkIntUnaryReturnAndParam(CI->getCalledFunction()))
-    return nullptr;
-
   // isascii(c) -> c <u 128
   Value *Op = CI->getArgOperand(0);
   Op = B.CreateICmpULT(Op, B.getInt32(128), "isascii");
@@ -1737,9 +1531,6 @@ Value *LibCallSimplifier::optimizeIsAscii(CallInst *CI, IRBuilder<> &B) {
 }
 
 Value *LibCallSimplifier::optimizeToAscii(CallInst *CI, IRBuilder<> &B) {
-  if (!checkIntUnaryReturnAndParam(CI->getCalledFunction()))
-    return nullptr;
-
   // toascii(c) -> c & 0x7f
   return B.CreateAnd(CI->getArgOperand(0),
                      ConstantInt::get(CI->getType(), 0x7F));
@@ -1753,6 +1544,7 @@ static bool isReportingError(Function *Callee, CallInst *CI, int StreamArg);
 
 Value *LibCallSimplifier::optimizeErrorReporting(CallInst *CI, IRBuilder<> &B,
                                                  int StreamArg) {
+  Function *Callee = CI->getCalledFunction();
   // Error reporting calls should be cold, mark them as such.
   // This applies even to non-builtin calls: it is only a hint and applies to
   // functions that the frontend might not understand as builtins.
@@ -1761,8 +1553,6 @@ Value *LibCallSimplifier::optimizeErrorReporting(CallInst *CI, IRBuilder<> &B,
   // Improving Static Branch Prediction in a Compiler
   // Brian L. Deitrich, Ben-Chung Cheng, Wen-mei W. Hwu
   // Proceedings of PACT'98, Oct. 1998, IEEE
-  Function *Callee = CI->getCalledFunction();
-
   if (!CI->hasFnAttr(Attribute::Cold) &&
       isReportingError(Callee, CI, StreamArg)) {
     CI->addAttribute(AttributeSet::FunctionIndex, Attribute::Cold);
@@ -1808,12 +1598,18 @@ Value *LibCallSimplifier::optimizePrintFString(CallInst *CI, IRBuilder<> &B) {
   if (!CI->use_empty())
     return nullptr;
 
-  // printf("x") -> putchar('x'), even for '%'.
-  if (FormatStr.size() == 1) {
-    Value *Res = EmitPutChar(B.getInt32(FormatStr[0]), B, TLI);
-    if (CI->use_empty() || !Res)
-      return Res;
-    return B.CreateIntCast(Res, CI->getType(), true);
+  // printf("x") -> putchar('x'), even for "%" and "%%".
+  if (FormatStr.size() == 1 || FormatStr == "%%")
+    return emitPutChar(B.getInt32(FormatStr[0]), B, TLI);
+
+  // printf("%s", "a") --> putchar('a')
+  if (FormatStr == "%s" && CI->getNumArgOperands() > 1) {
+    StringRef ChrStr;
+    if (!getConstantStringInfo(CI->getOperand(1), ChrStr))
+      return nullptr;
+    if (ChrStr.size() != 1)
+      return nullptr;
+    return emitPutChar(B.getInt32(ChrStr[0]), B, TLI);
   }
 
   // printf("foo\n") --> puts("foo")
@@ -1823,40 +1619,26 @@ Value *LibCallSimplifier::optimizePrintFString(CallInst *CI, IRBuilder<> &B) {
     // pass to be run after this pass, to merge duplicate strings.
     FormatStr = FormatStr.drop_back();
     Value *GV = B.CreateGlobalString(FormatStr, "str");
-    Value *NewCI = EmitPutS(GV, B, TLI);
-    return (CI->use_empty() || !NewCI)
-               ? NewCI
-               : ConstantInt::get(CI->getType(), FormatStr.size() + 1);
+    return emitPutS(GV, B, TLI);
   }
 
   // Optimize specific format strings.
   // printf("%c", chr) --> putchar(chr)
   if (FormatStr == "%c" && CI->getNumArgOperands() > 1 &&
-      CI->getArgOperand(1)->getType()->isIntegerTy()) {
-    Value *Res = EmitPutChar(CI->getArgOperand(1), B, TLI);
-
-    if (CI->use_empty() || !Res)
-      return Res;
-    return B.CreateIntCast(Res, CI->getType(), true);
-  }
+      CI->getArgOperand(1)->getType()->isIntegerTy())
+    return emitPutChar(CI->getArgOperand(1), B, TLI);
 
   // printf("%s\n", str) --> puts(str)
   if (FormatStr == "%s\n" && CI->getNumArgOperands() > 1 &&
-      CI->getArgOperand(1)->getType()->isPointerTy()) {
-    return EmitPutS(CI->getArgOperand(1), B, TLI);
-  }
+      CI->getArgOperand(1)->getType()->isPointerTy())
+    return emitPutS(CI->getArgOperand(1), B, TLI);
   return nullptr;
 }
 
 Value *LibCallSimplifier::optimizePrintF(CallInst *CI, IRBuilder<> &B) {
 
   Function *Callee = CI->getCalledFunction();
-  // Require one fixed pointer argument and an integer/void result.
   FunctionType *FT = Callee->getFunctionType();
-  if (FT->getNumParams() < 1 || !FT->getParamType(0)->isPointerTy() ||
-      !(FT->getReturnType()->isIntegerTy() || FT->getReturnType()->isVoidTy()))
-    return nullptr;
-
   if (Value *V = optimizePrintFString(CI, B)) {
     return V;
   }
@@ -1909,7 +1691,7 @@ Value *LibCallSimplifier::optimizeSPrintFString(CallInst *CI, IRBuilder<> &B) {
     if (!CI->getArgOperand(2)->getType()->isIntegerTy())
       return nullptr;
     Value *V = B.CreateTrunc(CI->getArgOperand(2), B.getInt8Ty(), "char");
-    Value *Ptr = CastToCStr(CI->getArgOperand(0), B);
+    Value *Ptr = castToCStr(CI->getArgOperand(0), B);
     B.CreateStore(V, Ptr);
     Ptr = B.CreateGEP(B.getInt8Ty(), Ptr, B.getInt32(1), "nul");
     B.CreateStore(B.getInt8(0), Ptr);
@@ -1922,7 +1704,7 @@ Value *LibCallSimplifier::optimizeSPrintFString(CallInst *CI, IRBuilder<> &B) {
     if (!CI->getArgOperand(2)->getType()->isPointerTy())
       return nullptr;
 
-    Value *Len = EmitStrLen(CI->getArgOperand(2), B, DL, TLI);
+    Value *Len = emitStrLen(CI->getArgOperand(2), B, DL, TLI);
     if (!Len)
       return nullptr;
     Value *IncLen =
@@ -1937,13 +1719,7 @@ Value *LibCallSimplifier::optimizeSPrintFString(CallInst *CI, IRBuilder<> &B) {
 
 Value *LibCallSimplifier::optimizeSPrintF(CallInst *CI, IRBuilder<> &B) {
   Function *Callee = CI->getCalledFunction();
-  // Require two fixed pointer arguments and an integer result.
   FunctionType *FT = Callee->getFunctionType();
-  if (FT->getNumParams() != 2 || !FT->getParamType(0)->isPointerTy() ||
-      !FT->getParamType(1)->isPointerTy() ||
-      !FT->getReturnType()->isIntegerTy())
-    return nullptr;
-
   if (Value *V = optimizeSPrintFString(CI, B)) {
     return V;
   }
@@ -1982,7 +1758,7 @@ Value *LibCallSimplifier::optimizeFPrintFString(CallInst *CI, IRBuilder<> &B) {
       if (FormatStr[i] == '%') // Could handle %% -> % if we cared.
         return nullptr;        // We found a format specifier.
 
-    return EmitFWrite(
+    return emitFWrite(
         CI->getArgOperand(1),
         ConstantInt::get(DL.getIntPtrType(CI->getContext()), FormatStr.size()),
         CI->getArgOperand(0), B, DL, TLI);
@@ -1999,27 +1775,21 @@ Value *LibCallSimplifier::optimizeFPrintFString(CallInst *CI, IRBuilder<> &B) {
     // fprintf(F, "%c", chr) --> fputc(chr, F)
     if (!CI->getArgOperand(2)->getType()->isIntegerTy())
       return nullptr;
-    return EmitFPutC(CI->getArgOperand(2), CI->getArgOperand(0), B, TLI);
+    return emitFPutC(CI->getArgOperand(2), CI->getArgOperand(0), B, TLI);
   }
 
   if (FormatStr[1] == 's') {
     // fprintf(F, "%s", str) --> fputs(str, F)
     if (!CI->getArgOperand(2)->getType()->isPointerTy())
       return nullptr;
-    return EmitFPutS(CI->getArgOperand(2), CI->getArgOperand(0), B, TLI);
+    return emitFPutS(CI->getArgOperand(2), CI->getArgOperand(0), B, TLI);
   }
   return nullptr;
 }
 
 Value *LibCallSimplifier::optimizeFPrintF(CallInst *CI, IRBuilder<> &B) {
   Function *Callee = CI->getCalledFunction();
-  // Require two fixed paramters as pointers and integer result.
   FunctionType *FT = Callee->getFunctionType();
-  if (FT->getNumParams() != 2 || !FT->getParamType(0)->isPointerTy() ||
-      !FT->getParamType(1)->isPointerTy() ||
-      !FT->getReturnType()->isIntegerTy())
-    return nullptr;
-
   if (Value *V = optimizeFPrintFString(CI, B)) {
     return V;
   }
@@ -2041,16 +1811,6 @@ Value *LibCallSimplifier::optimizeFPrintF(CallInst *CI, IRBuilder<> &B) {
 Value *LibCallSimplifier::optimizeFWrite(CallInst *CI, IRBuilder<> &B) {
   optimizeErrorReporting(CI, B, 3);
 
-  Function *Callee = CI->getCalledFunction();
-  // Require a pointer, an integer, an integer, a pointer, returning integer.
-  FunctionType *FT = Callee->getFunctionType();
-  if (FT->getNumParams() != 4 || !FT->getParamType(0)->isPointerTy() ||
-      !FT->getParamType(1)->isIntegerTy() ||
-      !FT->getParamType(2)->isIntegerTy() ||
-      !FT->getParamType(3)->isPointerTy() ||
-      !FT->getReturnType()->isIntegerTy())
-    return nullptr;
-
   // Get the element size and count.
   ConstantInt *SizeC = dyn_cast<ConstantInt>(CI->getArgOperand(1));
   ConstantInt *CountC = dyn_cast<ConstantInt>(CI->getArgOperand(2));
@@ -2065,8 +1825,8 @@ Value *LibCallSimplifier::optimizeFWrite(CallInst *CI, IRBuilder<> &B) {
   // If this is writing one byte, turn it into fputc.
   // This optimisation is only valid, if the return value is unused.
   if (Bytes == 1 && CI->use_empty()) { // fwrite(S,1,1,F) -> fputc(S[0],F)
-    Value *Char = B.CreateLoad(CastToCStr(CI->getArgOperand(0), B), "char");
-    Value *NewCI = EmitFPutC(Char, CI->getArgOperand(3), B, TLI);
+    Value *Char = B.CreateLoad(castToCStr(CI->getArgOperand(0), B), "char");
+    Value *NewCI = emitFPutC(Char, CI->getArgOperand(3), B, TLI);
     return NewCI ? ConstantInt::get(CI->getType(), 1) : nullptr;
   }
 
@@ -2076,12 +1836,13 @@ Value *LibCallSimplifier::optimizeFWrite(CallInst *CI, IRBuilder<> &B) {
 Value *LibCallSimplifier::optimizeFPuts(CallInst *CI, IRBuilder<> &B) {
   optimizeErrorReporting(CI, B, 1);
 
-  Function *Callee = CI->getCalledFunction();
+  // Don't rewrite fputs to fwrite when optimising for size because fwrite
+  // requires more arguments and thus extra MOVs are required.
+  if (CI->getParent()->getParent()->optForSize())
+    return nullptr;
 
-  // Require two pointers.  Also, we can't optimize if return value is used.
-  FunctionType *FT = Callee->getFunctionType();
-  if (FT->getNumParams() != 2 || !FT->getParamType(0)->isPointerTy() ||
-      !FT->getParamType(1)->isPointerTy() || !CI->use_empty())
+  // We can't optimize if return value is used.
+  if (!CI->use_empty())
     return nullptr;
 
   // fputs(s,F) --> fwrite(s,1,strlen(s),F)
@@ -2090,20 +1851,13 @@ Value *LibCallSimplifier::optimizeFPuts(CallInst *CI, IRBuilder<> &B) {
     return nullptr;
 
   // Known to have no uses (see above).
-  return EmitFWrite(
+  return emitFWrite(
       CI->getArgOperand(0),
       ConstantInt::get(DL.getIntPtrType(CI->getContext()), Len - 1),
       CI->getArgOperand(1), B, DL, TLI);
 }
 
 Value *LibCallSimplifier::optimizePuts(CallInst *CI, IRBuilder<> &B) {
-  Function *Callee = CI->getCalledFunction();
-  // Require one fixed pointer argument and an integer/void result.
-  FunctionType *FT = Callee->getFunctionType();
-  if (FT->getNumParams() < 1 || !FT->getParamType(0)->isPointerTy() ||
-      !(FT->getReturnType()->isIntegerTy() || FT->getReturnType()->isVoidTy()))
-    return nullptr;
-
   // Check for a constant string.
   StringRef Str;
   if (!getConstantStringInfo(CI->getArgOperand(0), Str))
@@ -2111,7 +1865,7 @@ Value *LibCallSimplifier::optimizePuts(CallInst *CI, IRBuilder<> &B) {
 
   if (Str.empty() && CI->use_empty()) {
     // puts("") -> putchar('\n')
-    Value *Res = EmitPutChar(B.getInt32('\n'), B, TLI);
+    Value *Res = emitPutChar(B.getInt32('\n'), B, TLI);
     if (CI->use_empty() || !Res)
       return Res;
     return B.CreateIntCast(Res, CI->getType(), true);
@@ -2133,10 +1887,8 @@ Value *LibCallSimplifier::optimizeStringMemoryLibCall(CallInst *CI,
                                                       IRBuilder<> &Builder) {
   LibFunc::Func Func;
   Function *Callee = CI->getCalledFunction();
-  StringRef FuncName = Callee->getName();
-
   // Check for string/memory library functions.
-  if (TLI->getLibFunc(FuncName, Func) && TLI->has(Func)) {
+  if (TLI->getLibFunc(*Callee, Func) && TLI->has(Func)) {
     // Make sure we never change the calling convention.
     assert((ignoreCallingConv(Func) ||
             CI->getCallingConv() == llvm::CallingConv::C) &&
@@ -2208,10 +1960,10 @@ Value *LibCallSimplifier::optimizeCall(CallInst *CI) {
   IRBuilder<> Builder(CI, /*FPMathTag=*/nullptr, OpBundles);
   bool isCallingConvC = CI->getCallingConv() == llvm::CallingConv::C;
 
-  // Command-line parameter overrides function attribute.
+  // Command-line parameter overrides instruction attribute.
   if (EnableUnsafeFPShrink.getNumOccurrences() > 0)
     UnsafeFPShrink = EnableUnsafeFPShrink;
-  else if (canUseUnsafeFPMath(Callee))
+  else if (isa<FPMathOperator>(CI) && CI->hasUnsafeAlgebra())
     UnsafeFPShrink = true;
 
   // First, check for intrinsics.
@@ -2229,6 +1981,7 @@ Value *LibCallSimplifier::optimizeCall(CallInst *CI) {
       return optimizeLog(CI, Builder);
     case Intrinsic::sqrt:
       return optimizeSqrt(CI, Builder);
+    // TODO: Use foldMallocMemset() with memset intrinsic.
     default:
       return nullptr;
     }
@@ -2253,7 +2006,7 @@ Value *LibCallSimplifier::optimizeCall(CallInst *CI) {
   }
 
   // Then check for known library functions.
-  if (TLI->getLibFunc(FuncName, Func) && TLI->has(Func)) {
+  if (TLI->getLibFunc(*Callee, Func) && TLI->has(Func)) {
     // We never change the calling convention.
     if (!ignoreCallingConv(Func) && !isCallingConvC)
       return nullptr;
@@ -2457,11 +2210,6 @@ bool FortifiedLibCallSimplifier::isFortifiedCallFoldable(CallInst *CI,
 
 Value *FortifiedLibCallSimplifier::optimizeMemCpyChk(CallInst *CI,
                                                      IRBuilder<> &B) {
-  Function *Callee = CI->getCalledFunction();
-
-  if (!checkStringCopyLibFuncSignature(Callee, LibFunc::memcpy_chk))
-    return nullptr;
-
   if (isFortifiedCallFoldable(CI, 3, 2, false)) {
     B.CreateMemCpy(CI->getArgOperand(0), CI->getArgOperand(1),
                    CI->getArgOperand(2), 1);
@@ -2472,11 +2220,6 @@ Value *FortifiedLibCallSimplifier::optimizeMemCpyChk(CallInst *CI,
 
 Value *FortifiedLibCallSimplifier::optimizeMemMoveChk(CallInst *CI,
                                                       IRBuilder<> &B) {
-  Function *Callee = CI->getCalledFunction();
-
-  if (!checkStringCopyLibFuncSignature(Callee, LibFunc::memmove_chk))
-    return nullptr;
-
   if (isFortifiedCallFoldable(CI, 3, 2, false)) {
     B.CreateMemMove(CI->getArgOperand(0), CI->getArgOperand(1),
                     CI->getArgOperand(2), 1);
@@ -2487,10 +2230,7 @@ Value *FortifiedLibCallSimplifier::optimizeMemMoveChk(CallInst *CI,
 
 Value *FortifiedLibCallSimplifier::optimizeMemSetChk(CallInst *CI,
                                                      IRBuilder<> &B) {
-  Function *Callee = CI->getCalledFunction();
-
-  if (!checkStringCopyLibFuncSignature(Callee, LibFunc::memset_chk))
-    return nullptr;
+  // TODO: Try foldMallocMemset() here.
 
   if (isFortifiedCallFoldable(CI, 3, 2, false)) {
     Value *Val = B.CreateIntCast(CI->getArgOperand(1), B.getInt8Ty(), false);
@@ -2506,16 +2246,12 @@ Value *FortifiedLibCallSimplifier::optimizeStrpCpyChk(CallInst *CI,
   Function *Callee = CI->getCalledFunction();
   StringRef Name = Callee->getName();
   const DataLayout &DL = CI->getModule()->getDataLayout();
-
-  if (!checkStringCopyLibFuncSignature(Callee, Func))
-    return nullptr;
-
   Value *Dst = CI->getArgOperand(0), *Src = CI->getArgOperand(1),
         *ObjSize = CI->getArgOperand(2);
 
   // __stpcpy_chk(x,x,...)  -> x+strlen(x)
   if (Func == LibFunc::stpcpy_chk && !OnlyLowerUnknownSize && Dst == Src) {
-    Value *StrLen = EmitStrLen(Src, B, DL, TLI);
+    Value *StrLen = emitStrLen(Src, B, DL, TLI);
     return StrLen ? B.CreateInBoundsGEP(B.getInt8Ty(), Dst, StrLen) : nullptr;
   }
 
@@ -2525,7 +2261,7 @@ Value *FortifiedLibCallSimplifier::optimizeStrpCpyChk(CallInst *CI,
   // TODO: It might be nice to get a maximum length out of the possible
   // string lengths for varying.
   if (isFortifiedCallFoldable(CI, 2, 1, true))
-    return EmitStrCpy(Dst, Src, B, TLI, Name.substr(2, 6));
+    return emitStrCpy(Dst, Src, B, TLI, Name.substr(2, 6));
 
   if (OnlyLowerUnknownSize)
     return nullptr;
@@ -2537,7 +2273,7 @@ Value *FortifiedLibCallSimplifier::optimizeStrpCpyChk(CallInst *CI,
 
   Type *SizeTTy = DL.getIntPtrType(CI->getContext());
   Value *LenV = ConstantInt::get(SizeTTy, Len);
-  Value *Ret = EmitMemCpyChk(Dst, Src, LenV, ObjSize, B, DL, TLI);
+  Value *Ret = emitMemCpyChk(Dst, Src, LenV, ObjSize, B, DL, TLI);
   // If the function was an __stpcpy_chk, and we were able to fold it into
   // a __memcpy_chk, we still need to return the correct end pointer.
   if (Ret && Func == LibFunc::stpcpy_chk)
@@ -2550,11 +2286,8 @@ Value *FortifiedLibCallSimplifier::optimizeStrpNCpyChk(CallInst *CI,
                                                        LibFunc::Func Func) {
   Function *Callee = CI->getCalledFunction();
   StringRef Name = Callee->getName();
-
-  if (!checkStringCopyLibFuncSignature(Callee, Func))
-    return nullptr;
   if (isFortifiedCallFoldable(CI, 3, 2, false)) {
-    Value *Ret = EmitStrNCpy(CI->getArgOperand(0), CI->getArgOperand(1),
+    Value *Ret = emitStrNCpy(CI->getArgOperand(0), CI->getArgOperand(1),
                              CI->getArgOperand(2), B, TLI, Name.substr(2, 7));
     return Ret;
   }
@@ -2577,15 +2310,15 @@ Value *FortifiedLibCallSimplifier::optimizeCall(CallInst *CI) {
 
   LibFunc::Func Func;
   Function *Callee = CI->getCalledFunction();
-  StringRef FuncName = Callee->getName();
 
   SmallVector<OperandBundleDef, 2> OpBundles;
   CI->getOperandBundlesAsDefs(OpBundles);
   IRBuilder<> Builder(CI, /*FPMathTag=*/nullptr, OpBundles);
   bool isCallingConvC = CI->getCallingConv() == llvm::CallingConv::C;
 
-  // First, check that this is a known library functions.
-  if (!TLI->getLibFunc(FuncName, Func))
+  // First, check that this is a known library functions and that the prototype
+  // is correct.
+  if (!TLI->getLibFunc(*Callee, Func))
     return nullptr;
 
   // We never change the calling convention.