[HIP] Add missing options for lto

[yaxunl]

Summary

Add -mcpu, -mattr, -mllvm, and -save-temps options for lto when necessary.

[tra]

clang/lib/Driver/ToolChains/HIP.cpp

76

I think StringRef will result in comparison by value. If we happen to have duplicate features, we may end up skipping a comma.
You need to either iterate ofer the pointers to elements of the Features, or use a different method of figuring out if a comma is necessary.
E.g. us a bool set at the end of the body.

[yaxunl]

Fix issue about -mattr. Also use generic LTO option translation and use -plugin-opt.

[yaxunl]

clang/lib/Driver/ToolChains/HIP.cpp

76

Fixed by using unifyTargetFeatures. This makes sure the features are unique.

[tra]

LGTM.

clang/lib/Driver/ToolChains/AMDGPU.h

public:
 
   static bool isWave64(const llvm::opt::ArgList &DriverArgs,
                        llvm::AMDGPU::GPUKind Kind);
   /// Needed for using lto.
   bool HasNativeLLVMSupport() const override {
     return true;
   }
 
+  /// Needed for translating LTO options.
+  const char *getDefaultLinker() const override { return "ld.lld"; }
 };
 
 class LLVM_LIBRARY_VISIBILITY ROCMToolChain : public AMDGPUToolChain {
 protected:
   RocmInstallationDetector RocmInstallation;
 
 public:
   ROCMToolChain(const Driver &D, const llvm::Triple &Triple,

clang/lib/Driver/ToolChains/HIP.cpp

 } // namespace
 
 void AMDGCN::Linker::constructLldCommand(Compilation &C, const JobAction &JA,
                                           const InputInfoList &Inputs,
                                           const InputInfo &Output,
                                           const llvm::opt::ArgList &Args) const {
   // Construct lld command.
   // The output from ld.lld is an HSA code object file.
-  ArgStringList LldArgs{"-flavor",
-                        "gnu",
-                        "--no-undefined",
-                        "-shared",
-                        "-mllvm",
-                        "-amdgpu-internalize-symbols",
-                        "-o",
-                        Output.getFilename()};
+  ArgStringList LldArgs{"-flavor", "gnu", "--no-undefined", "-shared",
+                        "-plugin-opt=-amdgpu-internalize-symbols"};
+
+  auto &TC = getToolChain();
+  auto &D = TC.getDriver();
+  assert(!Inputs.empty() && "Must have at least one input.");
+  addLTOOptions(TC, Args, LldArgs, Output, Inputs[0],
+                D.getLTOMode() == LTOK_Thin);
+
+  // Extract all the -m options
+  std::vector<llvm::StringRef> Features;
+  amdgpu::getAMDGPUTargetFeatures(D, Args, Features);
+  unifyTargetFeatures(Features);
+
+  // Add features to mattr such as cumode
+  std::string MAttrString = "-plugin-opt=-mattr=";
+  for (auto OneFeature : Features) {
+    MAttrString.append(Args.MakeArgString(OneFeature));
+    if (OneFeature != Features.back())
+      MAttrString.append(",");
+  }
+  if (!Features.empty())
+    LldArgs.push_back(Args.MakeArgString(MAttrString));
+
+  for (const Arg *A : Args.filtered(options::OPT_mllvm)) {
+    LldArgs.push_back(
+        Args.MakeArgString(Twine("-plugin-opt=") + A->getValue(0)));
+  }
+
+  if (C.getDriver().isSaveTempsEnabled())
+    LldArgs.push_back("-save-temps");
+
+  LldArgs.append({"-o", Output.getFilename()});
   for (auto Input : Inputs)
     LldArgs.push_back(Input.getFilename());
   const char *Lld = Args.MakeArgString(getToolChain().GetProgramPath("lld"));
   C.addCommand(std::make_unique<Command>(JA, *this, Lld, LldArgs, Inputs));
 }
 
 // Construct a clang-offload-bundler command to bundle code objects for
 // different GPU's into a HIP fat binary.

clang/test/Driver/hip-link-save-temps.hip

 // RUN:   --offload-arch=gfx906  %T/obj1.o %T/obj2.o 2>&1 | \
 // RUN:   FileCheck -check-prefixes=CHECK,SLNO %s
 
 // CHECK: "{{.*clang-offload-bundler.*}}" {{.*}} "-outputs=obj1-host-x86_64-unknown-linux-gnu.o,obj1-hip-amdgcn-amd-amdhsa-gfx900.o,obj1-hip-amdgcn-amd-amdhsa-gfx906.o" "-unbundle"
 // CHECK: "{{.*clang-offload-bundler.*}}" {{.*}} "-outputs=obj2-host-x86_64-unknown-linux-gnu.o,obj2-hip-amdgcn-amd-amdhsa-gfx900.o,obj2-hip-amdgcn-amd-amdhsa-gfx906.o" "-unbundle"
 // CHECK-NOT: llvm-link
 // CHECK-NOT: opt
 // CHECK-NOT: llc
-// CHECK: "{{.*lld.*}}" {{.*}} "-mllvm" "-amdgpu-internalize-symbols"
+// CHECK: "{{.*lld.*}}" {{.*}} "-plugin-opt=-amdgpu-internalize-symbols"
 // CHECK-SAME: "-o" "a.out-hip-amdgcn-amd-amdhsa-gfx900" "obj1-hip-amdgcn-amd-amdhsa-gfx900.o" "obj2-hip-amdgcn-amd-amdhsa-gfx900.o"
-// CHECK: "{{.*lld.*}}" {{.*}} "-mllvm" "-amdgpu-internalize-symbols"
+// CHECK: "{{.*lld.*}}" {{.*}} "-plugin-opt=-amdgpu-internalize-symbols"
 // CHECK-SAME: "-o" "a.out-hip-amdgcn-amd-amdhsa-gfx906" "obj1-hip-amdgcn-amd-amdhsa-gfx906.o" "obj2-hip-amdgcn-amd-amdhsa-gfx906.o"
 // CHECK: {{".*llvm-mc.*"}} "-triple" "amdgcn-amd-amdhsa" "-o"
 // CHECK-SAME: "[[OBJBUNDLE:.*.o]]" "{{.*}}.mcin" "--filetype=obj"
 // OUT: "{{.*ld.*}}" {{.*}} "-o" "executable" {{.*}} "[[OBJBUNDLE]]"
 // NOUT: "{{.*ld.*}}" {{.*}} "-o" "a.out" {{.*}} "[[OBJBUNDLE]]"
 // SLO: "{{.*llvm-ar.*}}" "rcsD" "libTest.a" {{.*}} "[[OBJBUNDLE]]"
 // SLNO: "{{.*llvm-ar.*}}" "rcsD" "a.out" {{.*}} "[[OBJBUNDLE]]"

clang/test/Driver/hip-save-temps.hip

 // NORDC-SAME: "-o" "hip-save-temps-hip-amdgcn-amd-amdhsa-gfx900.s"
 
 // RDC: {{".*clang.*"}} "-cc1" {{.*}} "-emit-llvm-bc" {{.*}} [[CPU]] {{.*}} "-o" "hip-save-temps-hip-amdgcn-amd-amdhsa-gfx900.bc"
 // NORDC: {{".*clang.*"}} "-cc1as" {{.*}} "-filetype" "obj" {{.*}} [[CPU]] {{.*}} "-o" "hip-save-temps-hip-amdgcn-amd-amdhsa-gfx900.o"
 // CHECK-NOT: "{{.*}}llvm-link"
 // CHECK-NOT: "{{.*}}opt"
 // CHECK-NOT: "{{.*}}llc"
 // NORDC: {{.*lld.*}}"-o" "hip-save-temps-hip-amdgcn-amd-amdhsa-gfx900.out"
-// RDCL: "{{.*lld.*}}" {{.*}} "-mllvm" "-amdgpu-internalize-symbols"
+// RDCL: "{{.*lld.*}}" {{.*}} "-plugin-opt=-amdgpu-internalize-symbols" {{.*}}"-save-temps"
 // RDCL-SAME: "-o" "a.out-hip-amdgcn-amd-amdhsa-gfx900"
 // RDCC: "{{.*clang.*}}" "-cc1" {{.*}} "-E" {{.*}} "-o" "hip-save-temps-host-x86_64-unknown-linux-gnu.cui"
 // RDCC: "{{.*clang.*}}" "-cc1" {{.*}} "-emit-llvm-bc" {{.*}} "-o" "hip-save-temps-host-x86_64-unknown-linux-gnu.bc"
 // RDCC: "{{.*clang.*}}" "-cc1" {{.*}} "-S" {{.*}} "-o" "hip-save-temps-host-x86_64-unknown-linux-gnu.s"
 // RDCC: "{{.*clang.*}}" "-cc1as" {{.*}} "-o" "hip-save-temps-host-x86_64-unknown-linux-gnu.o"
 // RDCC: "{{.*clang-offload-bundler.*}}" {{.*}} "-outputs=hip-save-temps.o"
 // RDCL: "{{.*clang-offload-bundler.*}}" {{.*}} "-outputs=hip-save-temps-hip-amdgcn-amd-amdhsa.hipfb"
 // RDCL: {{.*}}llvm-mc{{.*}}"-o" "hip-save-temps-hip-amdgcn-amd-amdhsa.o" "hip-save-temps-hip-amdgcn-amd-amdhsa.mcin" "--filetype=obj"

clang/test/Driver/hip-toolchain-features.hip

 // REQUIRES: clang-driver
 // REQUIRES: x86-registered-target
 // REQUIRES: amdgpu-registered-target
 
-// RUN: %clang -### -c -target x86_64-linux-gnu -fgpu-rdc \
+// RUN: %clang -### -target x86_64-linux-gnu -fgpu-rdc -nogpulib \
 // RUN:   --cuda-gpu-arch=gfx803 --cuda-gpu-arch=gfx900 %s \
 // RUN:   -mxnack 2>&1 | FileCheck %s -check-prefix=XNACK
-// RUN: %clang -### -c -target x86_64-linux-gnu -fgpu-rdc \
+// RUN: %clang -### -target x86_64-linux-gnu -fgpu-rdc -nogpulib \
 // RUN:   --cuda-gpu-arch=gfx803 --cuda-gpu-arch=gfx900 %s \
 // RUN:   -mno-xnack 2>&1 | FileCheck %s -check-prefix=NOXNACK
 
 // XNACK: {{.*}}clang{{.*}}"-target-feature" "+xnack"
+// XNACK: {{.*}}lld{{.*}}"-plugin-opt=-mattr=+xnack"
 // NOXNACK: {{.*}}clang{{.*}}"-target-feature" "-xnack"
+// NOXNACK: {{.*}}lld{{.*}}"-plugin-opt=-mattr=-xnack"
 
 
-// RUN: %clang -### -c -target x86_64-linux-gnu -fgpu-rdc \
+// RUN: %clang -### -target x86_64-linux-gnu -fgpu-rdc -nogpulib \
 // RUN:   --cuda-gpu-arch=gfx803 --cuda-gpu-arch=gfx900 %s \
 // RUN:   -msram-ecc 2>&1 | FileCheck %s -check-prefix=SRAM
-// RUN: %clang -### -c -target x86_64-linux-gnu -fgpu-rdc \
+// RUN: %clang -### -target x86_64-linux-gnu -fgpu-rdc -nogpulib \
 // RUN:   --cuda-gpu-arch=gfx803 --cuda-gpu-arch=gfx900 %s \
 // RUN:   -mno-sram-ecc 2>&1 | FileCheck %s -check-prefix=NOSRAM
 
 // SRAM: {{.*}}clang{{.*}}"-target-feature" "+sram-ecc"
+// SRAM: {{.*}}lld{{.*}}"-plugin-opt=-mattr=+sram-ecc"
 // NOSRAM: {{.*}}clang{{.*}}"-target-feature" "-sram-ecc"
+// NOSRAM: {{.*}}lld{{.*}}"-plugin-opt=-mattr=-sram-ecc"
 
 
-// RUN: %clang -### -c -target x86_64-linux-gnu -fgpu-rdc \
+// RUN: %clang -### -target x86_64-linux-gnu -fgpu-rdc -nogpulib \
 // RUN:   --cuda-gpu-arch=gfx803 --cuda-gpu-arch=gfx900 %s \
 // RUN:   -mxnack -msram-ecc \
 // RUN:   2>&1 | FileCheck %s -check-prefix=ALL3
-// RUN: %clang -### -c -target x86_64-linux-gnu -fgpu-rdc \
+// RUN: %clang -### -target x86_64-linux-gnu -fgpu-rdc -nogpulib \
 // RUN:   --cuda-gpu-arch=gfx803 --cuda-gpu-arch=gfx900 %s \
 // RUN:   -mno-xnack -mno-sram-ecc \
 // RUN:   2>&1 | FileCheck %s -check-prefix=NOALL3
 
 // ALL3: {{.*}}clang{{.*}}"-target-feature" "+xnack" "-target-feature" "+sram-ecc"
+// ALL3: {{.*}}lld{{.*}}"-plugin-opt=-mattr=+xnack,+sram-ecc"
+// NOALL3: {{.*}}clang{{.*}}"-target-feature" "-xnack" "-target-feature" "-sram-ecc"
+// NOALL3: {{.*}}lld{{.*}}"-plugin-opt=-mattr=-xnack,-sram-ecc"
 // NOALL3: {{.*}}clang{{.*}}"-target-feature" "-xnack" "-target-feature" "-sram-ecc"
 
 // RUN: %clang -### -target x86_64-linux-gnu -fgpu-rdc -nogpulib \
 // RUN:   --cuda-gpu-arch=gfx1010 %s \
 // RUN:   -mcumode -mcumode -mno-cumode -mwavefrontsize64 -mcumode \
 // RUN:   -mwavefrontsize64 -mno-wavefrontsize64 2>&1 \
 // RUN:   | FileCheck %s -check-prefix=DUP
 // DUP: {{.*}}clang{{.*}} "-target-feature" "-wavefrontsize16"
 // DUP-SAME: "-target-feature" "+wavefrontsize32"
 // DUP-SAME: "-target-feature" "-wavefrontsize64"
 // DUP-SAME: "-target-feature" "+cumode"
+// DUP: {{.*}}lld{{.*}} "-plugin-opt=-mattr=-wavefrontsize16,+wavefrontsize32,-wavefrontsize64,+cumode"

clang/test/Driver/hip-toolchain-mllvm.hip

 // REQUIRES: clang-driver
 // REQUIRES: x86-registered-target
 // REQUIRES: amdgpu-registered-target
 
 // RUN: %clang -### -target x86_64-linux-gnu \
 // RUN:   --cuda-gpu-arch=gfx803 --cuda-gpu-arch=gfx900 \
 // RUN:   -mllvm -amdgpu-function-calls=0 \
 // RUN:   %s 2>&1 | FileCheck %s
 
+// RUN: %clang -### -target x86_64-linux-gnu \
+// RUN:   --cuda-gpu-arch=gfx803 --cuda-gpu-arch=gfx900 \
+// RUN:   -fgpu-rdc -mllvm -amdgpu-function-calls=0 \
+// RUN:   %s 2>&1 | FileCheck -check-prefixes=CHECK,RDC %s
+
 // CHECK: [[CLANG:".*clang.*"]] "-cc1" "-triple" "amdgcn-amd-amdhsa"
 // CHECK-SAME: "-aux-triple" "x86_64-unknown-linux-gnu"
-// CHECK-SAME: "-emit-obj"
 // CHECK-SAME: {{.*}} "-target-cpu" "gfx803"
 // CHECK-SAME: {{.*}} "-mllvm" "-amdgpu-function-calls=0" {{.*}}
 
 // CHECK-NOT: {{".*opt"}}
 // CHECK-NOT: {{".*llc"}}
+// RDC: [[LLD:".*lld.*"]] {{.*}} "-plugin-opt=-amdgpu-function-calls=0"
 
 // CHECK: [[CLANG]] "-cc1" "-triple" "amdgcn-amd-amdhsa"
 // CHECK-SAME: "-aux-triple" "x86_64-unknown-linux-gnu"
-// CHECK-SAME: "-emit-obj"
 // CHECK-SAME: {{.*}} "-target-cpu" "gfx900"
 // CHECK-SAME: {{.*}} "-mllvm" "-amdgpu-function-calls=0" {{.*}}
 
 // CHECK-NOT: {{".*opt"}}
 // CHECK-NOT: {{".*llc"}}
+// RDC: [[LLD:".*lld.*"]] {{.*}} "-plugin-opt=-amdgpu-function-calls=0"

clang/test/Driver/hip-toolchain-opt.hip

 // Os-SAME: "-Os"
 // Oz-SAME: "-Oz"
 // Og-SAME: "-Og"
 
 // ALL-NOT: "{{.*}}opt"
 
 // ALL-NOT: "{{.*}}llc"
 
+// ALL: "{{.*}}lld{{.*}}" {{.*}} "-plugin-opt=mcpu=gfx900"
+// DEFAULT-NOT: "-plugin-opt=O{{.*}}"
+// O0-SAME: "-plugin-opt=O0"
+// O1-SAME: "-plugin-opt=O1"
+// O2-SAME: "-plugin-opt=O2"
+// O3-SAME: "-plugin-opt=O3"
+// Os-SAME: "-plugin-opt=O2"
+// Oz-SAME: "-plugin-opt=O2"
+// Og-SAME: "-plugin-opt=O1"
+
 // ALL: "{{.*}}clang{{.*}}" "-cc1" "-triple" "x86_64-unknown-linux-gnu"
 // DEFAULT-NOT: "-O{{.}}"
 // O0-SAME: "-O0"
 // O1-SAME: "-O1"
 // O2-SAME: "-O2"
 // O3-SAME: "-O3"
 // Os-SAME: "-Os"
 // Oz-SAME: "-Oz"
 // Og-SAME: "-Og"

clang/test/Driver/hip-toolchain-rdc-separate.hip

 // LINK: [[BUNDLER:".*clang-offload-bundler"]] "-type=o"
 // LINK-SAME: "-targets=host-x86_64-unknown-linux-gnu,hip-amdgcn-amd-amdhsa-gfx803,hip-amdgcn-amd-amdhsa-gfx900"
 // LINK-SAME: "-inputs=[[B_O]]" "-outputs={{.*o}},[[B_BC1:.*o]],[[B_BC2:.*o]]"
 // LINK: "-unbundle"
 
 // LINK-NOT: "*.llvm-link"
 // LINK-NOT: ".*opt"
 // LINK-NOT: ".*llc"
-// LINK: {{".*lld.*"}} {{.*}} "-mllvm" "-amdgpu-internalize-symbols"
+// LINK: {{".*lld.*"}} {{.*}} "-plugin-opt=-amdgpu-internalize-symbols"
+// LINK: "-plugin-opt=mcpu=gfx803"
 // LINK-SAME: "-o" "[[IMG_DEV1:.*.out]]" "[[A_BC1]]" "[[B_BC1]]"
 
 // LINK-NOT: "*.llvm-link"
 // LINK-NOT: ".*opt"
 // LINK-NOT: ".*llc"
-// LINK: {{".*lld.*"}} {{.*}} "-mllvm" "-amdgpu-internalize-symbols"
+// LINK: {{".*lld.*"}} {{.*}} "-plugin-opt=-amdgpu-internalize-symbols"
+// LINK: "-plugin-opt=mcpu=gfx900"
 // LINK-SAME: "-o" "[[IMG_DEV2:.*.out]]" "[[A_BC2]]" "[[B_BC2]]"
 
 // LINK: [[BUNDLER:".*clang-offload-bundler"]] "-type=o"
 // LINK-SAME: "-targets={{.*}},hip-amdgcn-amd-amdhsa-gfx803,hip-amdgcn-amd-amdhsa-gfx900"
 // LINK-SAME: "-inputs={{.*}},[[IMG_DEV1]],[[IMG_DEV2]]" "-outputs=[[BUNDLE:.*hipfb]]"
 
 // LINK: {{".*llvm-mc.*"}} "-triple" "amdgcn-amd-amdhsa" "-o"
 // LINK-SAME: "[[OBJBUNDLE:.*o]]" "{{.*}}.mcin" "--filetype=obj"
 
 // LINK: [[LD:".*ld.*"]] {{.*}} "-o" "a.out" {{.*}} "[[A_OBJ_HOST]]"
 // LINK-SAME: "[[B_OBJ_HOST]]" "[[OBJBUNDLE]]"

clang/test/Driver/hip-toolchain-rdc.hip

 // CHECK-SAME: "{{.*}}lib1.bc" "{{.*}}lib2.bc"
 // CHECK-SAME: "-target-cpu" "gfx803"
 // CHECK-SAME: {{.*}} "-o" [[B_BC1:".*bc"]] "-x" "hip"
 // CHECK-SAME: {{.*}} [[B_SRC]]
 
 // CHECK-NOT: "*.llvm-link"
 // CHECK-NOT: ".*opt"
 // CHECK-NOT: ".*llc"
-// CHECK: {{".*lld.*"}} {{.*}} "-mllvm" "-amdgpu-internalize-symbols"
+// CHECK: {{".*lld.*"}} {{.*}} "-plugin-opt=-amdgpu-internalize-symbols"
+// CHECK-SAME: "-plugin-opt=mcpu=gfx803"
 // CHECK-SAME: "-o" "[[IMG_DEV1:.*.out]]" [[A_BC1]] [[B_BC1]]
 
 // generate image for device side path on gfx900
 // CHECK: [[CLANG]] "-cc1" "-triple" "amdgcn-amd-amdhsa"
 // CHECK-SAME: "-aux-triple" "x86_64-unknown-linux-gnu"
 // CHECK-SAME: "-emit-llvm-bc"
 // CHECK-SAME: {{.*}} "-main-file-name" "a.cu"
 // CHECK-SAME: "-fcuda-is-device" "-fgpu-rdc"

 // CHECK-SAME: "{{.*}}lib1.bc" "{{.*}}lib2.bc"
 // CHECK-SAME: "-target-cpu" "gfx900"
 // CHECK-SAME: {{.*}} "-o" [[B_BC2:".*bc"]] "-x" "hip"
 // CHECK-SAME: {{.*}} [[B_SRC]]
 
 // CHECK-NOT: "*.llvm-link"
 // CHECK-NOT: ".*opt"
 // CHECK-NOT: ".*llc"
-// CHECK: {{".*lld.*"}} {{.*}} "-mllvm" "-amdgpu-internalize-symbols"
+// CHECK: {{".*lld.*"}} {{.*}} "-plugin-opt=-amdgpu-internalize-symbols"
+// CHECK-SAME: "-plugin-opt=mcpu=gfx900"
 // CHECK-SAME: "-o" "[[IMG_DEV2:.*.out]]" [[A_BC2]] [[B_BC2]]
 
 // combine images generated into hip fat binary object
 // CHECK: [[BUNDLER:".*clang-offload-bundler"]] "-type=o"
 // CHECK-SAME: "-targets={{.*}},hip-amdgcn-amd-amdhsa-gfx803,hip-amdgcn-amd-amdhsa-gfx900"
 // CHECK-SAME: "-inputs={{.*}},[[IMG_DEV1]],[[IMG_DEV2]]" "-outputs=[[BUNDLE:.*hipfb]]"
 
 // CHECK: [[MC:".*llvm-mc"]] "-triple" "amdgcn-amd-amdhsa"
 // CHECK-SAME: "-o" [[OBJBUNDLE:".*o"]] "{{.*}}.mcin" "--filetype=obj"
 
 // output the executable
 // CHECK: [[LD:".*ld.*"]] {{.*}}"-o" "a.out" {{.*}} [[A_OBJ_HOST]] [[B_OBJ_HOST]] [[OBJBUNDLE]]