Merge pull request llvm#171 from AMD-Lightning-Internal/amd/dev/rlieberm/FortranDrop002

Initial Flang-new drop 001

Author: ronlieb

.github/workflows/release-binaries.yml

@@ -328,7 +328,7 @@ jobs:
       run: |
         # Build some of the mlir tools that take a long time to link
         if [ "${{ needs.prepare.outputs.build-flang }}" = "true" ]; then
-          ninja -C ${{ steps.setup-stage.outputs.build-prefix }}/build/tools/clang/stage2-bins/ -j2 flang bbc
+          ninja -C ${{ steps.setup-stage.outputs.build-prefix }}/build/tools/clang/stage2-bins/ -j2 flang-new bbc
         fi
         ninja -C ${{ steps.setup-stage.outputs.build-prefix }}/build/tools/clang/stage2-bins/ \
             mlir-bytecode-parser-fuzzer \

clang/include/clang/Basic/DiagnosticGroups.td

@@ -1593,6 +1593,3 @@ def ExtractAPIMisuse : DiagGroup<"extractapi-misuse">;
 // Warnings about using the non-standard extension having an explicit specialization
 // with a storage class specifier.
 def ExplicitSpecializationStorageClass : DiagGroup<"explicit-specialization-storage-class">;
-
-// A warning for options that enable a feature that is not yet complete
-def ExperimentalOption : DiagGroup<"experimental-option">;

clang/include/clang/Driver/Options.td

@@ -7102,6 +7102,7 @@ defm xor_operator : OptInFC1FFlag<"xor-operator", "Enable .XOR. as a synonym of
 defm logical_abbreviations : OptInFC1FFlag<"logical-abbreviations", "Enable logical abbreviations">;
 defm implicit_none : OptInFC1FFlag<"implicit-none", "No implicit typing allowed unless overridden by IMPLICIT statements">;
 defm underscoring : OptInFC1FFlag<"underscoring", "Appends one trailing underscore to external names">;
+defm offload_global_filtering : OptInFC1FFlag<"offload-global-filtering", "Enable/disable OpenMP global filtering pass">;
 defm ppc_native_vec_elem_order: BoolOptionWithoutMarshalling<"f", "ppc-native-vector-element-order",
   PosFlag<SetTrue, [], [ClangOption], "Specifies PowerPC native vector element order (default)">,
   NegFlag<SetFalse, [], [ClangOption], "Specifies PowerPC non-native vector element order">>;
@@ -7122,6 +7123,10 @@ defm loop_versioning : BoolOptionWithoutMarshalling<"f", "version-loops-for-stri
 
 def fhermetic_module_files : Flag<["-"], "fhermetic-module-files">, Group<f_Group>,
   HelpText<"Emit hermetic module files (no nested USE association)">;
+
+def do_concurrent_parallel_EQ : Joined<["-"], "fdo-concurrent-parallel=">,
+  HelpText<"Try to map `do concurrent` loops to OpenMP (on host or device)">,
+      Values<"none,host,device">;
 } // let Visibility = [FC1Option, FlangOption]
 
 def J : JoinedOrSeparate<["-"], "J">,
@@ -8503,7 +8508,7 @@ def source_date_epoch : Separate<["-"], "source-date-epoch">,
 // CUDA Options
 //===----------------------------------------------------------------------===//
 
-let Visibility = [CC1Option] in {
+let Visibility = [CC1Option, FC1Option] in {
 
 def fcuda_is_device : Flag<["-"], "fcuda-is-device">,
   HelpText<"Generate code for CUDA device">,

clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp

@@ -1893,7 +1893,6 @@ void CGOpenMPRuntimeGPU::emitReduction(
     return;
 
   bool ParallelReduction = isOpenMPParallelDirective(Options.ReductionKind);
-  bool DistributeReduction = isOpenMPDistributeDirective(Options.ReductionKind);
   bool TeamsReduction = isOpenMPTeamsDirective(Options.ReductionKind);
 
   ASTContext &C = CGM.getContext();
@@ -1990,7 +1989,7 @@ void CGOpenMPRuntimeGPU::emitReduction(
   llvm::OpenMPIRBuilder::InsertPointTy AfterIP =
       cantFail(OMPBuilder.createReductionsGPU(
           OmpLoc, AllocaIP, CodeGenIP, ReductionInfos, false, TeamsReduction,
-          DistributeReduction, llvm::OpenMPIRBuilder::ReductionGenCBKind::Clang,
+          llvm::OpenMPIRBuilder::ReductionGenCBKind::Clang,
           CGF.getTarget().getGridValue(),
           C.getLangOpts().OpenMPCUDAReductionBufNum, RTLoc));
   CGF.Builder.restoreIP(AfterIP);

clang/lib/Driver/Driver.cpp

@@ -2256,7 +2256,7 @@ void Driver::PrintHelp(bool ShowHidden) const {
 
 void Driver::PrintVersion(const Compilation &C, raw_ostream &OS) const {
   if (IsFlangMode()) {
-    OS << getClangToolFullVersion("alpha flang-new") << '\n';
+    OS << getClangToolFullVersion("flang-new") << '\n';
   } else {
     // FIXME: The following handlers should use a callback mechanism, we don't
     // know what the client would like to do.

clang/lib/Driver/ToolChains/Clang.cpp

@@ -9139,7 +9139,9 @@ void ClangAs::ConstructJob(Compilation &C, const JobAction &JA,
   assert(Input.isFilename() && "Invalid input.");
   CmdArgs.push_back(Input.getFilename());
 
-  const char *Exec = getToolChain().getDriver().getClangProgramPath();
+  // TODO This is a workaround to enable using -save-temps with flang-new
+  // const char *Exec = getToolChain().getDriver().getClangProgramPath();
+  const char *Exec = Args.MakeArgString(getToolChain().GetProgramPath("clang"));
   if (D.CC1Main && !D.CCGenDiagnostics) {
     // Invoke cc1as directly in this process.
     C.addCommand(std::make_unique<CC1Command>(

clang/lib/Driver/ToolChains/Flang.cpp

@@ -124,6 +124,8 @@ void Flang::addOtherOptions(const ArgList &Args, ArgStringList &CmdArgs) const {
                    options::OPT_std_EQ, options::OPT_W_Joined,
                    options::OPT_fconvert_EQ, options::OPT_fpass_plugin_EQ,
                    options::OPT_funderscoring, options::OPT_fno_underscoring,
+                   options::OPT_foffload_global_filtering,
+                   options::OPT_fno_offload_global_filtering,
                    options::OPT_funsigned, options::OPT_fno_unsigned});
 
   llvm::codegenoptions::DebugInfoKind DebugInfoKind;
@@ -151,7 +153,8 @@ void Flang::addCodegenOptions(const ArgList &Args,
     CmdArgs.push_back("-fversion-loops-for-stride");
 
   Args.addAllArgs(CmdArgs,
-                  {options::OPT_flang_experimental_hlfir,
+                  {options::OPT_do_concurrent_parallel_EQ,
+                   options::OPT_flang_experimental_hlfir,
                    options::OPT_flang_deprecated_no_hlfir,
                    options::OPT_fno_ppc_native_vec_elem_order,
                    options::OPT_fppc_native_vec_elem_order,

clang/test/ClangScanDeps/multiple-commands.c

@@ -134,7 +134,7 @@
 // CHECK-NEXT:           "{{.*}}tu_save_temps_module.o"
 // CHECK:                "{{.*}}tu_save_temps_module.s"
 // CHECK:              ]
-// CHECK-NEXT:         "executable": "clang_tool"
+// CHECK-NEXT:         "executable": [[CLANG:"[^"]*clang"]]
 // CHECK:              "input-file": "[[PREFIX]]{{.}}tu_save_temps_module.c"
 // CHECK-NEXT:       }
 // CHECK-NEXT:     ]

clang/test/Driver/hip-target-id.hip

@@ -26,7 +26,7 @@
 // CHECK-SAME: "-target-feature" "+sramecc"
 // CHECK-SAME: "-target-feature" "+xnack"
 
-// TMP: [[CLANG:"[^"]*clang[^"]*"]] "-cc1as" "-triple" "amdgcn-amd-amdhsa"
+// TMP: [[CLANG_TMP:"[^"]*clang"]] "-cc1as" "-triple" "amdgcn-amd-amdhsa"
 // TMP-SAME: "-target-cpu" "gfx908"
 // TMP-SAME: "-target-feature" "+sramecc"
 // TMP-SAME: "-target-feature" "+xnack"