clang: Guess at some platform FTZ/DAZ default settings

This is to avoid performance regressions when the default attribute
behavior is fixed to assume ieee.

I tested the default on x86_64 ubuntu, which seems to default to
FTZ/DAZ, but am guessing for x86 and PS4.

Author: arsenm

clang/include/clang/Driver/ToolChain.h

@@ -573,12 +573,19 @@ class ToolChain {
   virtual void AddCCKextLibArgs(const llvm::opt::ArgList &Args,
                                 llvm::opt::ArgStringList &CmdArgs) const;
 
+  /// If a runtime library exists that sets global flags for unsafe floating
+  /// point math, return true.
+  ///
+  /// This checks for presence of the -Ofast, -ffast-math or -funsafe-math flags.
+  virtual bool isFastMathRuntimeAvailable(
+    const llvm::opt::ArgList &Args, std::string &Path) const;
+
   /// AddFastMathRuntimeIfAvailable - If a runtime library exists that sets
   /// global flags for unsafe floating point math, add it and return true.
   ///
   /// This checks for presence of the -Ofast, -ffast-math or -funsafe-math flags.
-  virtual bool AddFastMathRuntimeIfAvailable(
-      const llvm::opt::ArgList &Args, llvm::opt::ArgStringList &CmdArgs) const;
+  bool addFastMathRuntimeIfAvailable(
+    const llvm::opt::ArgList &Args, llvm::opt::ArgStringList &CmdArgs) const;
 
   /// addProfileRTLibs - When -fprofile-instr-profile is specified, try to pass
   /// a suitable profile runtime library to the linker.

clang/lib/Driver/ToolChain.cpp

@@ -917,28 +917,35 @@ void ToolChain::AddCCKextLibArgs(const ArgList &Args,
   CmdArgs.push_back("-lcc_kext");
 }
 
-bool ToolChain::AddFastMathRuntimeIfAvailable(const ArgList &Args,
-                                              ArgStringList &CmdArgs) const {
+bool ToolChain::isFastMathRuntimeAvailable(const ArgList &Args,
+                                           std::string &Path) const {
   // Do not check for -fno-fast-math or -fno-unsafe-math when -Ofast passed
   // (to keep the linker options consistent with gcc and clang itself).
   if (!isOptimizationLevelFast(Args)) {
     // Check if -ffast-math or -funsafe-math.
     Arg *A =
-        Args.getLastArg(options::OPT_ffast_math, options::OPT_fno_fast_math,
-                        options::OPT_funsafe_math_optimizations,
-                        options::OPT_fno_unsafe_math_optimizations);
+      Args.getLastArg(options::OPT_ffast_math, options::OPT_fno_fast_math,
+                      options::OPT_funsafe_math_optimizations,
+                      options::OPT_fno_unsafe_math_optimizations);
 
     if (!A || A->getOption().getID() == options::OPT_fno_fast_math ||
         A->getOption().getID() == options::OPT_fno_unsafe_math_optimizations)
       return false;
   }
   // If crtfastmath.o exists add it to the arguments.
-  std::string Path = GetFilePath("crtfastmath.o");
-  if (Path == "crtfastmath.o") // Not found.
-    return false;
+  Path = GetFilePath("crtfastmath.o");
+  return (Path != "crtfastmath.o"); // Not found.
+}
+
+bool ToolChain::addFastMathRuntimeIfAvailable(const ArgList &Args,
+                                              ArgStringList &CmdArgs) const {
+  std::string Path;
+  if (isFastMathRuntimeAvailable(Args, Path)) {
+    CmdArgs.push_back(Args.MakeArgString(Path));
+    return true;
+  }
 
-  CmdArgs.push_back(Args.MakeArgString(Path));
-  return true;
+  return false;
 }
 
 SanitizerMask ToolChain::getSupportedSanitizers() const {

clang/lib/Driver/ToolChains/Gnu.cpp

@@ -502,7 +502,7 @@ void tools::gnutools::Linker::ConstructJob(Compilation &C, const JobAction &JA,
     }
 
     // Add crtfastmath.o if available and fast math is enabled.
-    ToolChain.AddFastMathRuntimeIfAvailable(Args, CmdArgs);
+    ToolChain.addFastMathRuntimeIfAvailable(Args, CmdArgs);
   }
 
   Args.AddAllArgs(CmdArgs, options::OPT_L);

clang/lib/Driver/ToolChains/Linux.cpp

@@ -988,3 +988,22 @@ void Linux::addProfileRTLibs(const llvm::opt::ArgList &Args,
         Twine("-u", llvm::getInstrProfRuntimeHookVarName())));
   ToolChain::addProfileRTLibs(Args, CmdArgs);
 }
+
+llvm::DenormalMode Linux::getDefaultDenormalModeForType(
+  const llvm::opt::ArgList &DriverArgs,
+  Action::OffloadKind DeviceOffloadKind,
+  const llvm::fltSemantics *FPType) const {
+  switch (getTriple().getArch()) {
+  case llvm::Triple::x86:
+  case llvm::Triple::x86_64: {
+    std::string Unused;
+    // DAZ and FTZ are turned on in crtfastmath.o
+    if (!DriverArgs.hasArg(options::OPT_nostdlib, options::OPT_nostartfiles) &&
+        isFastMathRuntimeAvailable(DriverArgs, Unused))
+      return llvm::DenormalMode::getPreserveSign();
+    return llvm::DenormalMode::getIEEE();
+  }
+  default:
+    return llvm::DenormalMode::getIEEE();
+  }
+}

clang/lib/Driver/ToolChains/Linux.h

@@ -46,6 +46,11 @@ class LLVM_LIBRARY_VISIBILITY Linux : public Generic_ELF {
 
   std::vector<std::string> ExtraOpts;
 
+  llvm::DenormalMode getDefaultDenormalModeForType(
+    const llvm::opt::ArgList &DriverArgs,
+    Action::OffloadKind DeviceOffloadKind,
+    const llvm::fltSemantics *FPType = nullptr) const override;
+
 protected:
   Tool *buildAssembler() const override;
   Tool *buildLinker() const override;

clang/lib/Driver/ToolChains/MinGW.cpp

@@ -300,7 +300,7 @@ void tools::MinGW::Linker::ConstructJob(Compilation &C, const JobAction &JA,
 
     if (!Args.hasArg(options::OPT_nostartfiles)) {
       // Add crtfastmath.o if available and fast math is enabled.
-      TC.AddFastMathRuntimeIfAvailable(Args, CmdArgs);
+      TC.addFastMathRuntimeIfAvailable(Args, CmdArgs);
 
       CmdArgs.push_back(Args.MakeArgString(TC.GetFilePath("crtend.o")));
     }

clang/lib/Driver/ToolChains/PS4CPU.h

@@ -88,6 +88,14 @@ class LLVM_LIBRARY_VISIBILITY PS4CPU : public Generic_ELF {
   // capable of unit splitting.
   bool canSplitThinLTOUnit() const override { return false; }
 
+  llvm::DenormalMode getDefaultDenormalModeForType(
+    const llvm::opt::ArgList &DriverArgs,
+    Action::OffloadKind DeviceOffloadKind,
+    const llvm::fltSemantics *FPType) const override {
+    // DAZ and FTZ are on by default.
+    return llvm::DenormalMode::getPreserveSign();
+  }
+
 protected:
   Tool *buildAssembler() const override;
   Tool *buildLinker() const override;

clang/test/Driver/default-denormal-fp-math.c

@@ -0,0 +1,19 @@
+// RUN: %clang -### -target arm-unknown-linux-gnu -c %s -v 2>&1 | FileCheck -check-prefix=CHECK-IEEE %s
+// RUN: %clang -### -target i386-unknown-linux-gnu -c %s -v 2>&1 | FileCheck -check-prefix=CHECK-IEEE %s
+
+// RUN: %clang -### -target x86_64-unknown-linux-gnu --sysroot=%S/Inputs/basic_linux_tree -c %s -v 2>&1 | FileCheck -check-prefix=CHECK-IEEE %s
+
+// crtfastmath enables ftz and daz
+// RUN: %clang -### -target x86_64-unknown-linux-gnu -ffast-math --sysroot=%S/Inputs/basic_linux_tree -c %s -v 2>&1 | FileCheck -check-prefix=CHECK-PRESERVESIGN %s
+
+// crt not linked in with nostartfiles
+// RUN: %clang -### -target x86_64-unknown-linux-gnu -ffast-math -nostartfiles --sysroot=%S/Inputs/basic_linux_tree -c %s -v 2>&1 | FileCheck -check-prefix=CHECK-IEEE %s
+
+// If there's no crtfastmath, don't assume ftz/daz
+// RUN: %clang -### -target x86_64-unknown-linux-gnu -ffast-math --sysroot=/dev/null -c %s -v 2>&1 | FileCheck -check-prefix=CHECK-IEEE %s
+
+// RUN: %clang -### -target x86_64-scei-ps4 -c %s -v 2>&1 | FileCheck -check-prefix=CHECK-PRESERVESIGN %s
+
+
+// CHECK-IEEE: -fdenormal-fp-math=ieee,ieee
+// CHECK-PRESERVESIGN: -fdenormal-fp-math=preserve-sign,preserve-sign