Index: cfe/trunk/include/clang/Basic/LangOptions.def
===================================================================
--- cfe/trunk/include/clang/Basic/LangOptions.def
+++ cfe/trunk/include/clang/Basic/LangOptions.def
@@ -173,6 +173,7 @@
 LANGOPT(CUDAIsDevice      , 1, 0, "compiling for CUDA device")
 LANGOPT(CUDAAllowVariadicFunctions, 1, 0, "allowing variadic functions in CUDA device code")
 LANGOPT(CUDAHostDeviceConstexpr, 1, 1, "treating unattributed constexpr functions as __host__ __device__")
+LANGOPT(CUDADeviceFlushDenormalsToZero, 1, 0, "flushing denormals to zero")
 
 LANGOPT(AssumeSaneOperatorNew , 1, 1, "implicit __attribute__((malloc)) for C++'s new operators")
 LANGOPT(SizedDeallocation , 1, 0, "enable sized deallocation functions")
Index: cfe/trunk/include/clang/Driver/Options.td
===================================================================
--- cfe/trunk/include/clang/Driver/Options.td
+++ cfe/trunk/include/clang/Driver/Options.td
@@ -382,6 +382,9 @@
   HelpText<"Enable device-side debug info generation. Disables ptxas optimizations.">;
 def cuda_path_EQ : Joined<["--"], "cuda-path=">, Group<i_Group>,
   HelpText<"CUDA installation path">;
+def fcuda_flush_denormals_to_zero : Flag<["-"], "fcuda-flush-denormals-to-zero">,
+  Flags<[CC1Option]>, HelpText<"Flush denormal floating point values to zero in CUDA device mode.">;
+def fno_cuda_flush_denormals_to_zero : Flag<["-"], "fno-cuda-flush-denormals-to-zero">;
 def dA : Flag<["-"], "dA">, Group<d_Group>;
 def dD : Flag<["-"], "dD">, Group<d_Group>, Flags<[CC1Option]>,
   HelpText<"Print macro definitions in -E mode in addition to normal output">;
Index: cfe/trunk/lib/CodeGen/CGCall.cpp
===================================================================
--- cfe/trunk/lib/CodeGen/CGCall.cpp
+++ cfe/trunk/lib/CodeGen/CGCall.cpp
@@ -1768,6 +1768,10 @@
     // __syncthreads(), and so can't have certain optimizations applied around
     // them).  LLVM will remove this attribute where it safely can.
     FuncAttrs.addAttribute(llvm::Attribute::Convergent);
+
+    // Respect -fcuda-flush-denormals-to-zero.
+    if (getLangOpts().CUDADeviceFlushDenormalsToZero)
+      FuncAttrs.addAttribute("nvptx-f32ftz", "true");
   }
 
   ClangToLLVMArgMapping IRFunctionArgs(getContext(), FI);
Index: cfe/trunk/lib/CodeGen/CodeGenModule.cpp
===================================================================
--- cfe/trunk/lib/CodeGen/CodeGenModule.cpp
+++ cfe/trunk/lib/CodeGen/CodeGenModule.cpp
@@ -472,6 +472,14 @@
     getModule().addModuleFlag(llvm::Module::Override, "Cross-DSO CFI", 1);
   }
 
+  if (LangOpts.CUDAIsDevice && getTarget().getTriple().isNVPTX()) {
+    // Indicate whether __nvvm_reflect should be configured to flush denormal
+    // floating point values to 0.  (This corresponds to its "__CUDA_FTZ"
+    // property.)
+    getModule().addModuleFlag(llvm::Module::Override, "nvvm-reflect-ftz",
+                              LangOpts.CUDADeviceFlushDenormalsToZero ? 1 : 0);
+  }
+
   if (uint32_t PLevel = Context.getLangOpts().PICLevel) {
     llvm::PICLevel::Level PL = llvm::PICLevel::Default;
     switch (PLevel) {
Index: cfe/trunk/lib/Driver/ToolChains.cpp
===================================================================
--- cfe/trunk/lib/Driver/ToolChains.cpp
+++ cfe/trunk/lib/Driver/ToolChains.cpp
@@ -4208,6 +4208,10 @@
   Linux::addClangTargetOptions(DriverArgs, CC1Args);
   CC1Args.push_back("-fcuda-is-device");
 
+  if (DriverArgs.hasFlag(options::OPT_fcuda_flush_denormals_to_zero,
+                         options::OPT_fno_cuda_flush_denormals_to_zero, false))
+    CC1Args.push_back("-fcuda-flush-denormals-to-zero");
+
   if (DriverArgs.hasArg(options::OPT_nocudalib))
     return;
 
Index: cfe/trunk/lib/Frontend/CompilerInvocation.cpp
===================================================================
--- cfe/trunk/lib/Frontend/CompilerInvocation.cpp
+++ cfe/trunk/lib/Frontend/CompilerInvocation.cpp
@@ -1571,6 +1571,9 @@
   if (Args.hasArg(OPT_fno_cuda_host_device_constexpr))
     Opts.CUDAHostDeviceConstexpr = 0;
 
+  if (Opts.CUDAIsDevice && Args.hasArg(OPT_fcuda_flush_denormals_to_zero))
+    Opts.CUDADeviceFlushDenormalsToZero = 1;
+
   if (Opts.ObjC1) {
     if (Arg *arg = Args.getLastArg(OPT_fobjc_runtime_EQ)) {
       StringRef value = arg->getValue();
Index: cfe/trunk/test/CodeGenCUDA/flush-denormals.cu
===================================================================
--- cfe/trunk/test/CodeGenCUDA/flush-denormals.cu
+++ cfe/trunk/test/CodeGenCUDA/flush-denormals.cu
@@ -0,0 +1,23 @@
+// RUN: %clang_cc1 -fcuda-is-device \
+// RUN:   -triple nvptx-nvidia-cuda -emit-llvm -o - %s | FileCheck %s -check-prefix NOFTZ
+// RUN: %clang_cc1 -fcuda-is-device -fcuda-flush-denormals-to-zero \
+// RUN:   -triple nvptx-nvidia-cuda -emit-llvm -o - %s | FileCheck %s -check-prefix FTZ
+
+#include "Inputs/cuda.h"
+
+// Checks that device function calls get emitted with the "ntpvx-f32ftz"
+// attribute set to "true" when we compile CUDA device code with
+// -fcuda-flush-denormals-to-zero.  Further, check that we reflect the presence
+// or absence of -fcuda-flush-denormals-to-zero in a module flag.
+
+// CHECK: define void @foo() #0
+extern "C" __device__ void foo() {}
+
+// FTZ: attributes #0 = {{.*}} "nvptx-f32ftz"="true"
+// NOFTZ-NOT: attributes #0 = {{.*}} "nvptx-f32ftz"
+
+// FTZ:!llvm.module.flags = !{[[MODFLAG:![0-9]+]]}
+// FTZ:[[MODFLAG]] = !{i32 4, !"nvvm-reflect-ftz", i32 1}
+
+// NOFTZ:!llvm.module.flags = !{[[MODFLAG:![0-9]+]]}
+// NOFTZ:[[MODFLAG]] = !{i32 4, !"nvvm-reflect-ftz", i32 0}