[CUDA][HIP] Rename and fix -fcuda-approx-transcendentals

Rename -fcuda-approx-transcendentals as -fgpu-approx-transcendentals and pass it to both device and host clang -cc1. Fix its interaction with -ffast-math to allow -fno-gpu-approx-transcendentals to override the implicit -fcuda-approx-transcendentals due to -ffast-math. Rename the predefined macro to be __CLANG_GPU_APPROX_TRANSCENDENTALS__. Emit the macro for both device and host compilation. Reviewed by: Artem Belevich, Fangrui Song Differential Revision: https://reviews.llvm.org/D154797
llvm · Jul 25, 2023 · e178824 · e178824
1 parent 12832c1
commit e178824
Show file tree

Hide file tree

Showing 16 changed files with 86 additions and 39 deletions.
diff --git a/clang/include/clang/Basic/LangOptions.def b/clang/include/clang/Basic/LangOptions.def
@@ -265,7 +265,7 @@ ENUM_LANGOPT(HLSLVersion, HLSLLangStd, 16, HLSL_Unset, "HLSL Version")
 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(CUDADeviceApproxTranscendentals, 1, 0, "using approximate transcendental functions")
+LANGOPT(GPUDeviceApproxTranscendentals, 1, 0, "using approximate transcendental functions")
 LANGOPT(GPURelocatableDeviceCode, 1, 0, "generate relocatable device code")
 LANGOPT(GPUAllowDeviceInit, 1, 0, "allowing device side global init functions for HIP")
 LANGOPT(GPUMaxThreadsPerBlock, 32, 1024, "default max threads per block for kernel launch bounds for HIP")

diff --git a/clang/include/clang/Driver/Options.td b/clang/include/clang/Driver/Options.td
@@ -6801,11 +6801,12 @@ def sycl_std_EQ : Joined<["-"], "sycl-std=">, Group<sycl_Group>,
   MarshallingInfoEnum<LangOpts<"SYCLVersion">, "SYCL_None">,
   ShouldParseIf<!strconcat(fsycl_is_device.KeyPath, "||", fsycl_is_host.KeyPath)>;
 
-defm cuda_approx_transcendentals : BoolFOption<"cuda-approx-transcendentals",
-  LangOpts<"CUDADeviceApproxTranscendentals">, DefaultFalse,
+defm gpu_approx_transcendentals : BoolFOption<"gpu-approx-transcendentals",
+  LangOpts<"GPUDeviceApproxTranscendentals">, DefaultFalse,
   PosFlag<SetTrue, [CC1Option], "Use">, NegFlag<SetFalse, [], "Don't use">,
-  BothFlags<[], " approximate transcendental functions">>,
-  ShouldParseIf<fcuda_is_device.KeyPath>;
+  BothFlags<[], " approximate transcendental functions">>;
+def : Flag<["-"], "fcuda-approx-transcendentals">, Alias<fgpu_approx_transcendentals>;
+def : Flag<["-"], "fno-cuda-approx-transcendentals">, Alias<fno_gpu_approx_transcendentals>;
 
 //===----------------------------------------------------------------------===//
 // Frontend Options - cc1 + fc1

diff --git a/clang/lib/Driver/ToolChains/Clang.cpp b/clang/lib/Driver/ToolChains/Clang.cpp
@@ -7245,6 +7245,18 @@ void Clang::ConstructJob(Compilation &C, const JobAction &JA,
     auto CUID = cast<InputAction>(SourceAction)->getId();
     if (!CUID.empty())
       CmdArgs.push_back(Args.MakeArgString(Twine("-cuid=") + Twine(CUID)));
+
+    // -ffast-math turns on -fgpu-approx-transcendentals implicitly, but will
+    // be overriden by -fno-gpu-approx-transcendentals.
+    bool UseApproxTranscendentals = Args.hasFlag(
+        options::OPT_ffast_math, options::OPT_fno_fast_math, false);
+    if (Args.hasFlag(options::OPT_fgpu_approx_transcendentals,
+                     options::OPT_fno_gpu_approx_transcendentals,
+                     UseApproxTranscendentals))
+      CmdArgs.push_back("-fgpu-approx-transcendentals");
+  } else {
+    Args.claimAllArgs(options::OPT_fgpu_approx_transcendentals,
+                      options::OPT_fno_gpu_approx_transcendentals);
   }
 
   if (IsHIP) {

diff --git a/clang/lib/Driver/ToolChains/Cuda.cpp b/clang/lib/Driver/ToolChains/Cuda.cpp
@@ -801,10 +801,6 @@ void CudaToolChain::addClangTargetOptions(
     CC1Args.append(
         {"-fcuda-is-device", "-mllvm", "-enable-memcpyopt-without-libcalls"});
 
-    if (DriverArgs.hasFlag(options::OPT_fcuda_approx_transcendentals,
-                           options::OPT_fno_cuda_approx_transcendentals, false))
-      CC1Args.push_back("-fcuda-approx-transcendentals");
-
     // Unsized function arguments used for variadics were introduced in CUDA-9.0
     // We still do not support generating code that actually uses variadic
     // arguments yet, but we do need to allow parsing them as recent CUDA

diff --git a/clang/lib/Driver/ToolChains/HIPAMD.cpp b/clang/lib/Driver/ToolChains/HIPAMD.cpp
@@ -243,10 +243,6 @@ void HIPAMDToolChain::addClangTargetOptions(
 
   CC1Args.push_back("-fcuda-is-device");
 
-  if (DriverArgs.hasFlag(options::OPT_fcuda_approx_transcendentals,
-                         options::OPT_fno_cuda_approx_transcendentals, false))
-    CC1Args.push_back("-fcuda-approx-transcendentals");
-
   if (!DriverArgs.hasFlag(options::OPT_fgpu_rdc, options::OPT_fno_gpu_rdc,
                           false))
     CC1Args.append({"-mllvm", "-amdgpu-internalize-symbols"});

diff --git a/clang/lib/Driver/ToolChains/HIPSPV.cpp b/clang/lib/Driver/ToolChains/HIPSPV.cpp
@@ -143,10 +143,6 @@ void HIPSPVToolChain::addClangTargetOptions(
        // TODO: Allow autovectorization when SPIR-V backend arrives.
        "-mllvm", "-vectorize-loops=false", "-mllvm", "-vectorize-slp=false"});
 
-  if (DriverArgs.hasFlag(options::OPT_fcuda_approx_transcendentals,
-                         options::OPT_fno_cuda_approx_transcendentals, false))
-    CC1Args.push_back("-fcuda-approx-transcendentals");
-
   // Default to "hidden" visibility, as object level linking will not be
   // supported for the foreseeable future.
   if (!DriverArgs.hasArg(options::OPT_fvisibility_EQ,

diff --git a/clang/lib/Frontend/InitPreprocessor.cpp b/clang/lib/Frontend/InitPreprocessor.cpp
@@ -1297,11 +1297,10 @@ static void InitializePredefinedMacros(const TargetInfo &TI,
     Builder.defineMacro("__CUDA_ARCH__");
   }
 
-  // We need to communicate this to our CUDA header wrapper, which in turn
-  // informs the proper CUDA headers of this choice.
-  if (LangOpts.CUDADeviceApproxTranscendentals || LangOpts.FastMath) {
-    Builder.defineMacro("__CLANG_CUDA_APPROX_TRANSCENDENTALS__");
-  }
+  // We need to communicate this to our CUDA/HIP header wrapper, which in turn
+  // informs the proper CUDA/HIP headers of this choice.
+  if (LangOpts.GPUDeviceApproxTranscendentals)
+    Builder.defineMacro("__CLANG_GPU_APPROX_TRANSCENDENTALS__");
 
   // Define a macro indicating that the source file is being compiled with a
   // SYCL device compiler which doesn't produce host binary.

diff --git a/clang/lib/Headers/__clang_cuda_math.h b/clang/lib/Headers/__clang_cuda_math.h
@@ -45,9 +45,9 @@
 // libdevice provides fast low precision and slow full-recision implementations
 // for some functions. Which one gets selected depends on
 // __CLANG_CUDA_APPROX_TRANSCENDENTALS__ which gets defined by clang if
-// -ffast-math or -fcuda-approx-transcendentals are in effect.
+// -ffast-math or -fgpu-approx-transcendentals are in effect.
 #pragma push_macro("__FAST_OR_SLOW")
-#if defined(__CLANG_CUDA_APPROX_TRANSCENDENTALS__)
+#if defined(__CLANG_GPU_APPROX_TRANSCENDENTALS__)
 #define __FAST_OR_SLOW(fast, slow) fast
 #else
 #define __FAST_OR_SLOW(fast, slow) slow

diff --git a/clang/lib/Headers/__clang_cuda_runtime_wrapper.h b/clang/lib/Headers/__clang_cuda_runtime_wrapper.h
@@ -196,12 +196,12 @@ inline __host__ double __signbitd(double x) {
 
 // math_function.hpp uses the __USE_FAST_MATH__ macro to determine whether we
 // get the slow-but-accurate or fast-but-inaccurate versions of functions like
-// sin and exp.  This is controlled in clang by -fcuda-approx-transcendentals.
+// sin and exp.  This is controlled in clang by -fgpu-approx-transcendentals.
 //
 // device_functions.hpp uses __USE_FAST_MATH__ for a different purpose (fast vs.
 // slow divides), so we need to scope our define carefully here.
 #pragma push_macro("__USE_FAST_MATH__")
-#if defined(__CLANG_CUDA_APPROX_TRANSCENDENTALS__)
+#if defined(__CLANG_GPU_APPROX_TRANSCENDENTALS__)
 #define __USE_FAST_MATH__ 1
 #endif
 

diff --git a/clang/lib/Headers/__clang_hip_math.h b/clang/lib/Headers/__clang_hip_math.h
@@ -34,10 +34,10 @@
 
 // Device library provides fast low precision and slow full-recision
 // implementations for some functions. Which one gets selected depends on
-// __CLANG_CUDA_APPROX_TRANSCENDENTALS__ which gets defined by clang if
-// -ffast-math or -fcuda-approx-transcendentals are in effect.
+// __CLANG_GPU_APPROX_TRANSCENDENTALS__ which gets defined by clang if
+// -ffast-math or -fgpu-approx-transcendentals are in effect.
 #pragma push_macro("__FAST_OR_SLOW")
-#if defined(__CLANG_CUDA_APPROX_TRANSCENDENTALS__)
+#if defined(__CLANG_GPU_APPROX_TRANSCENDENTALS__)
 #define __FAST_OR_SLOW(fast, slow) fast
 #else
 #define __FAST_OR_SLOW(fast, slow) slow

diff --git a/clang/test/Driver/hip-macros.hip b/clang/test/Driver/hip-macros.hip
@@ -78,3 +78,11 @@
 // PTS-DAG: #define HIP_API_PER_THREAD_DEFAULT_STREAM 1
 // NOPTS-NOT: #define __HIP_API_PER_THREAD_DEFAULT_STREAM__
 // NOPTS-NOT: #define HIP_API_PER_THREAD_DEFAULT_STREAM
+
+// RUN: %clang -E -dM --offload-arch=gfx906 -nogpuinc -nogpulib \
+// RUN:   %s 2>&1 | FileCheck --check-prefix=NOAPPROX %s
+// RUN: %clang -E -dM --offload-arch=gfx906 -nogpuinc -nogpulib -fgpu-approx-transcendentals \
+// RUN:   %s 2>&1 | FileCheck --check-prefix=APPROX %s
+// NOAPPROX-NOT: #define __CLANG_GPU_APPROX_TRANSCENDENTALS__
+// APPROX: #define __CLANG_GPU_APPROX_TRANSCENDENTALS__ 1
+// APPROX: #define __CLANG_GPU_APPROX_TRANSCENDENTALS__ 1
diff --git a/clang/test/Driver/hip-options.hip b/clang/test/Driver/hip-options.hip
@@ -169,3 +169,39 @@
 // RUN: %clang -### -nogpuinc -nogpulib -fhip-fp32-correctly-rounded-divide-sqrt \
 // RUN:   --cuda-gpu-arch=gfx906  %s 2>&1 | FileCheck -check-prefixes=CRDS %s
 // CRDS-NOT: "-f{{(no-)?}}hip-fp32-correctly-rounded-divide-sqrt"
+
+// Check -fgpu-approx-transcendentals is passed to clang -cc1 but
+// (default) -fno-gpu-approx-transcendentals is not.
+// -ffast-math implies -fgpu-approx-transcendentals, which can be overridden
+// by -fno-gpu-approx-transcendentals.
+
+// RUN: %clang -### --target=x86_64-unknown-linux-gnu -nogpuinc -nogpulib -fgpu-approx-transcendentals \
+// RUN:   --cuda-gpu-arch=gfx906 %s 2>&1 | FileCheck -check-prefix=APPROX %s
+
+// RUN: %clang -### --target=x86_64-unknown-linux-gnu -nogpuinc -nogpulib -ffast-math \
+// RUN:   --cuda-gpu-arch=gfx906 %s 2>&1 | FileCheck -check-prefix=APPROX %s
+
+// APPROX: "-cc1"{{.*}} "-triple" "amdgcn-amd-amdhsa" {{.*}} "-fgpu-approx-transcendentals"
+// APPROX: "-cc1"{{.*}} "-triple" "x86_64-unknown-linux-gnu" {{.*}} "-fgpu-approx-transcendentals"
+
+// RUN: %clang -### -nogpuinc -nogpulib -fno-gpu-approx-transcendentals \
+// RUN:   --cuda-gpu-arch=gfx906 %s 2>&1 | FileCheck -check-prefix=NOAPPROX %s
+
+// RUN: %clang -### -nogpuinc -nogpulib \
+// RUN:   --cuda-gpu-arch=gfx906 %s 2>&1 | FileCheck -check-prefix=NOAPPROX %s
+
+// RUN: %clang -### -nogpuinc -nogpulib -ffast-math -fno-fast-math \
+// RUN:   --cuda-gpu-arch=gfx906 %s 2>&1 | FileCheck -check-prefix=NOAPPROX %s
+
+// RUN: %clang -### -nogpuinc -nogpulib -ffast-math -fno-gpu-approx-transcendentals \
+// RUN:   --cuda-gpu-arch=gfx906 %s 2>&1 | FileCheck -check-prefix=NOAPPROX %s
+
+// NOAPPROX-NOT: "-f{{(no-)?}}gpu-approx-transcendentals"
+
+// Check no warnings for -fgpu-approx-transcendentals.
+
+// RUN: %clang -fdriver-only -Werror --target=x86_64-unknown-linux-gnu -nogpuinc -nogpulib -fgpu-approx-transcendentals \
+// RUN:   --cuda-gpu-arch=gfx906 %s 2>&1 | count 0
+
+// RUN: %clang -fdriver-only -Werror --target=x86_64-unknown-linux-gnu -nostdinc -nostdlib -fgpu-approx-transcendentals \
+// RUN:   -x c++ %s 2>&1 | count 0
diff --git a/clang/test/Headers/__clang_hip_math.hip b/clang/test/Headers/__clang_hip_math.hip
@@ -22,7 +22,7 @@
 // RUN:   -internal-isystem %S/../../lib/Headers/cuda_wrappers \
 // RUN:   -internal-isystem %S/Inputs/include \
 // RUN:   -triple amdgcn-amd-amdhsa -aux-triple x86_64-unknown-unknown \
-// RUN:   -target-cpu gfx906 -emit-llvm %s -fcuda-is-device -O1 -fcuda-approx-transcendentals -o - \
+// RUN:   -target-cpu gfx906 -emit-llvm %s -fcuda-is-device -O1 -fgpu-approx-transcendentals -o - \
 // RUN:   -D__HIPCC_RTC__ | FileCheck -check-prefixes=CHECK,APPROX %s
 
 #define BOOL_TYPE int
@@ -3694,10 +3694,13 @@ extern "C" __device__ BOOL_TYPE test___signbit(double x) {
 //
 // APPROX-LABEL: @test_sincosf(
 // APPROX-NEXT:  entry:
-// APPROX-NEXT:    [[CALL_I_I:%.*]] = tail call contract float @__ocml_native_sin_f32(float noundef [[X:%.*]]) #[[ATTR16]]
-// APPROX-NEXT:    store float [[CALL_I_I]], ptr [[Y:%.*]], align 4, !tbaa [[TBAA16]]
-// APPROX-NEXT:    [[CALL1_I_I:%.*]] = tail call contract float @__ocml_native_cos_f32(float noundef [[X]]) #[[ATTR16]]
-// APPROX-NEXT:    store float [[CALL1_I_I]], ptr [[Z:%.*]], align 4, !tbaa [[TBAA16]]
+// APPROX-NEXT:    [[__TMP_I:%.*]] = alloca float, align 4, addrspace(5)
+// APPROX-NEXT:    call void @llvm.lifetime.start.p5(i64 4, ptr addrspace(5) [[__TMP_I]]) #[[ATTR17]]
+// APPROX-NEXT:    [[CALL_I:%.*]] = call contract float @__ocml_sincos_f32(float noundef [[X:%.*]], ptr addrspace(5) noundef [[__TMP_I]]) #[[ATTR16]]
+// APPROX-NEXT:    store float [[CALL_I]], ptr [[Y:%.*]], align 4, !tbaa [[TBAA16]]
+// APPROX-NEXT:    [[TMP0:%.*]] = load float, ptr addrspace(5) [[__TMP_I]], align 4, !tbaa [[TBAA16]]
+// APPROX-NEXT:    store float [[TMP0]], ptr [[Z:%.*]], align 4, !tbaa [[TBAA16]]
+// APPROX-NEXT:    call void @llvm.lifetime.end.p5(i64 4, ptr addrspace(5) [[__TMP_I]]) #[[ATTR17]]
 // APPROX-NEXT:    ret void
 //
 extern "C" __device__ void test_sincosf(float x, float *y, float *z) {

diff --git a/clang/test/Headers/nvptx_device_math_sin.c b/clang/test/Headers/nvptx_device_math_sin.c
@@ -2,7 +2,7 @@
 // RUN: %clang_cc1 -x c -internal-isystem %S/Inputs/include -fopenmp -triple powerpc64le-unknown-unknown -fopenmp-targets=nvptx64-nvidia-cuda -emit-llvm-bc %s -o %t-ppc-host.bc
 // RUN: %clang_cc1 -x c -include __clang_openmp_device_functions.h -internal-isystem %S/../../lib/Headers/openmp_wrappers -internal-isystem %S/Inputs/include -fopenmp -triple nvptx64-nvidia-cuda -aux-triple powerpc64le-unknown-unknown -fopenmp-targets=nvptx64-nvidia-cuda -emit-llvm %s -fopenmp-is-target-device -fopenmp-host-ir-file-path %t-ppc-host.bc -o - | FileCheck %s --check-prefix=SLOW
 // RUN: %clang_cc1 -x c -internal-isystem %S/Inputs/include -fopenmp -triple powerpc64le-unknown-unknown -fopenmp-targets=nvptx64-nvidia-cuda -emit-llvm-bc %s -o %t-ppc-host.bc -ffast-math -ffp-contract=fast
-// RUN: %clang_cc1 -x c -include __clang_openmp_device_functions.h -internal-isystem %S/../../lib/Headers/openmp_wrappers -internal-isystem %S/Inputs/include -fopenmp -triple nvptx64-nvidia-cuda -aux-triple powerpc64le-unknown-unknown -fopenmp-targets=nvptx64-nvidia-cuda -emit-llvm %s -fopenmp-is-target-device -fopenmp-host-ir-file-path %t-ppc-host.bc -o - -ffast-math -ffp-contract=fast | FileCheck %s --check-prefix=FAST
+// RUN: %clang_cc1 -x c -include __clang_openmp_device_functions.h -internal-isystem %S/../../lib/Headers/openmp_wrappers -internal-isystem %S/Inputs/include -fopenmp -triple nvptx64-nvidia-cuda -aux-triple powerpc64le-unknown-unknown -fopenmp-targets=nvptx64-nvidia-cuda -emit-llvm %s -fopenmp-is-target-device -fopenmp-host-ir-file-path %t-ppc-host.bc -o - -fgpu-approx-transcendentals -ffast-math -ffp-contract=fast | FileCheck %s --check-prefix=FAST
 // expected-no-diagnostics
 
 #include <math.h>

diff --git a/clang/test/Headers/nvptx_device_math_sin.cpp b/clang/test/Headers/nvptx_device_math_sin.cpp
@@ -2,7 +2,7 @@
 // RUN: %clang_cc1 -x c++ -internal-isystem %S/Inputs/include -fopenmp -triple powerpc64le-unknown-unknown -fopenmp-targets=nvptx64-nvidia-cuda -emit-llvm-bc %s -o %t-ppc-host.bc
 // RUN: %clang_cc1 -x c++ -include __clang_openmp_device_functions.h -internal-isystem %S/../../lib/Headers/openmp_wrappers -internal-isystem %S/Inputs/include -fopenmp -triple nvptx64-nvidia-cuda -aux-triple powerpc64le-unknown-unknown -fopenmp-targets=nvptx64-nvidia-cuda -emit-llvm %s -fopenmp-is-target-device -fopenmp-host-ir-file-path %t-ppc-host.bc -o - | FileCheck %s --check-prefix=SLOW
 // RUN: %clang_cc1 -x c++ -internal-isystem %S/Inputs/include -fopenmp -triple powerpc64le-unknown-unknown -fopenmp-targets=nvptx64-nvidia-cuda -emit-llvm-bc %s -o %t-ppc-host.bc -ffast-math -ffp-contract=fast
-// RUN: %clang_cc1 -x c++ -include __clang_openmp_device_functions.h -internal-isystem %S/../../lib/Headers/openmp_wrappers -internal-isystem %S/Inputs/include -fopenmp -triple nvptx64-nvidia-cuda -aux-triple powerpc64le-unknown-unknown -fopenmp-targets=nvptx64-nvidia-cuda -emit-llvm %s -fopenmp-is-target-device -fopenmp-host-ir-file-path %t-ppc-host.bc -o - -ffast-math -ffp-contract=fast | FileCheck %s --check-prefix=FAST
+// RUN: %clang_cc1 -x c++ -include __clang_openmp_device_functions.h -internal-isystem %S/../../lib/Headers/openmp_wrappers -internal-isystem %S/Inputs/include -fopenmp -triple nvptx64-nvidia-cuda -aux-triple powerpc64le-unknown-unknown -fopenmp-targets=nvptx64-nvidia-cuda -emit-llvm %s -fopenmp-is-target-device -fopenmp-host-ir-file-path %t-ppc-host.bc -o - -fgpu-approx-transcendentals -ffast-math -ffp-contract=fast | FileCheck %s --check-prefix=FAST
 // expected-no-diagnostics
 
 #include <cmath>

diff --git a/clang/test/Preprocessor/cuda-approx-transcendentals.cu b/clang/test/Preprocessor/cuda-approx-transcendentals.cu
@@ -3,6 +3,6 @@
 // RUN: %clang -fcuda-approx-transcendentals --cuda-device-only -nocudainc -nocudalib -target i386-unknown-linux-gnu -x cuda -E -dM -o - /dev/null | FileCheck --check-prefix DEVICE-FAST %s
 // RUN: %clang -ffast-math --cuda-device-only -nocudainc -nocudalib -target i386-unknown-linux-gnu -x cuda -E -dM -o - /dev/null | FileCheck --check-prefix DEVICE-FAST %s
 
-// HOST-NOT: __CLANG_CUDA_APPROX_TRANSCENDENTALS__
-// DEVICE-NOFAST-NOT: __CLANG_CUDA_APPROX_TRANSCENDENTALS__
-// DEVICE-FAST: __CLANG_CUDA_APPROX_TRANSCENDENTALS__
+// HOST-NOT: __CLANG_GPU_APPROX_TRANSCENDENTALS__
+// DEVICE-NOFAST-NOT: __GPU_CUDA_APPROX_TRANSCENDENTALS__
+// DEVICE-FAST: __CLANG_GPU_APPROX_TRANSCENDENTALS__