Merge remote-tracking branch 'upstream/main' into delayed_privatization_15

Author: ergawy

clang/docs/UsersManual.rst

@@ -2319,6 +2319,8 @@ are listed below.
    on ELF targets when using the integrated assembler. This flag currently
    only has an effect on ELF targets.
 
+.. _funique_internal_linkage_names:
+
 .. option:: -f[no]-unique-internal-linkage-names
 
    Controls whether Clang emits a unique (best-effort) symbol name for internal
@@ -2448,27 +2450,41 @@ usual build cycle when using sample profilers for optimization:
    usual build flags that you always build your application with. The only
    requirement is that DWARF debug info including source line information is
    generated. This DWARF information is important for the profiler to be able
-   to map instructions back to source line locations.
+   to map instructions back to source line locations. The usefulness of this
+   DWARF information can be improved with the ``-fdebug-info-for-profiling``
+   and ``-funique-internal-linkage-names`` options.
 
-   On Linux, ``-g`` or just ``-gline-tables-only`` is sufficient:
+   On Linux:
 
    .. code-block:: console
 
-     $ clang++ -O2 -gline-tables-only code.cc -o code
+     $ clang++ -O2 -gline-tables-only \
+       -fdebug-info-for-profiling -funique-internal-linkage-names \
+       code.cc -o code
 
    While MSVC-style targets default to CodeView debug information, DWARF debug
    information is required to generate source-level LLVM profiles. Use
    ``-gdwarf`` to include DWARF debug information:
 
-   .. code-block:: console
+   .. code-block:: winbatch
+
+     > clang-cl /O2 -gdwarf -gline-tables-only ^
+       /clang:-fdebug-info-for-profiling /clang:-funique-internal-linkage-names ^
+       code.cc /Fe:code /fuse-ld=lld /link /debug:dwarf
+
+.. note::
 
-     $ clang-cl -O2 -gdwarf -gline-tables-only coff-profile.cpp -fuse-ld=lld -link -debug:dwarf
+   :ref:`-funique-internal-linkage-names <funique_internal_linkage_names>`
+   generates unique names based on given command-line source file paths. If
+   your build system uses absolute source paths and these paths may change
+   between steps 1 and 4, then the uniqued function names may change and result
+   in unused profile data. Consider omitting this option in such cases.
 
 2. Run the executable under a sampling profiler. The specific profiler
    you use does not really matter, as long as its output can be converted
    into the format that the LLVM optimizer understands.
 
-   Two such profilers are the the Linux Perf profiler
+   Two such profilers are the Linux Perf profiler
    (https://perf.wiki.kernel.org/) and Intel's Sampling Enabling Product (SEP),
    available as part of `Intel VTune
    <https://software.intel.com/content/www/us/en/develop/tools/oneapi/components/vtune-profiler.html>`_.
@@ -2482,7 +2498,9 @@ usual build cycle when using sample profilers for optimization:
 
    .. code-block:: console
 
-     $ perf record -b ./code
+     $ perf record -b -e BR_INST_RETIRED.NEAR_TAKEN:uppp ./code
+
+   If the event above is unavailable, ``branches:u`` is probably next-best.
 
    Note the use of the ``-b`` flag. This tells Perf to use the Last Branch
    Record (LBR) to record call chains. While this is not strictly required,
@@ -2532,21 +2550,42 @@ usual build cycle when using sample profilers for optimization:
    that executes faster than the original one. Note that you are not
    required to build the code with the exact same arguments that you
    used in the first step. The only requirement is that you build the code
-   with ``-gline-tables-only`` and ``-fprofile-sample-use``.
+   with the same debug info options and ``-fprofile-sample-use``.
+
+   On Linux:
 
    .. code-block:: console
 
-     $ clang++ -O2 -gline-tables-only -fprofile-sample-use=code.prof code.cc -o code
+     $ clang++ -O2 -gline-tables-only \
+       -fdebug-info-for-profiling -funique-internal-linkage-names \
+       -fprofile-sample-use=code.prof code.cc -o code
 
-  [OPTIONAL] Sampling-based profiles can have inaccuracies or missing block/
-  edge counters. The profile inference algorithm (profi) can be used to infer
-  missing blocks and edge counts, and improve the quality of profile data.
-  Enable it with ``-fsample-profile-use-profi``.
+   On Windows:
 
-  .. code-block:: console
+   .. code-block:: winbatch
+
+     > clang-cl /O2 -gdwarf -gline-tables-only ^
+       /clang:-fdebug-info-for-profiling /clang:-funique-internal-linkage-names ^
+       /fprofile-sample-use=code.prof code.cc /Fe:code /fuse-ld=lld /link /debug:dwarf
+
+   [OPTIONAL] Sampling-based profiles can have inaccuracies or missing block/
+   edge counters. The profile inference algorithm (profi) can be used to infer
+   missing blocks and edge counts, and improve the quality of profile data.
+   Enable it with ``-fsample-profile-use-profi``. For example, on Linux:
+
+   .. code-block:: console
+
+     $ clang++ -fsample-profile-use-profi -O2 -gline-tables-only \
+       -fdebug-info-for-profiling -funique-internal-linkage-names \
+       -fprofile-sample-use=code.prof code.cc -o code
+
+   On Windows:
+
+   .. code-block:: winbatch
 
-    $ clang++ -O2 -gline-tables-only -fprofile-sample-use=code.prof \
-      -fsample-profile-use-profi code.cc -o code
+     > clang-cl /clang:-fsample-profile-use-profi /O2 -gdwarf -gline-tables-only ^
+       /clang:-fdebug-info-for-profiling /clang:-funique-internal-linkage-names ^
+       /fprofile-sample-use=code.prof code.cc /Fe:code /fuse-ld=lld /link /debug:dwarf
 
 Sample Profile Formats
 """"""""""""""""""""""

flang/lib/Lower/ConvertVariable.cpp

@@ -496,8 +496,8 @@ static fir::GlobalOp defineGlobal(Fortran::lower::AbstractConverter &converter,
   if (mlir::isa<fir::SequenceType>(symTy) &&
       !Fortran::semantics::IsAllocatableOrPointer(sym)) {
     mlir::Type eleTy = mlir::cast<fir::SequenceType>(symTy).getEleTy();
-    if (eleTy.isa<mlir::IntegerType, mlir::FloatType, fir::ComplexType,
-                  fir::LogicalType>()) {
+    if (mlir::isa<mlir::IntegerType, mlir::FloatType, fir::ComplexType,
+                  fir::LogicalType>(eleTy)) {
       const auto *details =
           sym.detailsIf<Fortran::semantics::ObjectEntityDetails>();
       if (details->init()) {

flang/lib/Optimizer/CodeGen/CodeGen.cpp

@@ -1964,12 +1964,12 @@ struct ValueOpCommon {
                  mlir::ArrayAttr arrAttr) {
     llvm::SmallVector<int64_t> indices;
     for (auto i = arrAttr.begin(), e = arrAttr.end(); i != e; ++i) {
-      if (auto intAttr = i->dyn_cast<mlir::IntegerAttr>()) {
+      if (auto intAttr = mlir::dyn_cast<mlir::IntegerAttr>(*i)) {
         indices.push_back(intAttr.getInt());
       } else {
-        auto fieldName = i->cast<mlir::StringAttr>().getValue();
+        auto fieldName = mlir::cast<mlir::StringAttr>(*i).getValue();
         ++i;
-        auto ty = i->cast<mlir::TypeAttr>().getValue();
+        auto ty = mlir::cast<mlir::TypeAttr>(*i).getValue();
         auto index = mlir::cast<fir::RecordType>(ty).getFieldIndex(fieldName);
         indices.push_back(index);
       }
@@ -3014,7 +3014,7 @@ static void selectMatchAndRewrite(const fir::LLVMTypeConverter &lowering,
       caseValues.push_back(intAttr.getInt());
       continue;
     }
-    assert(attr.template dyn_cast_or_null<mlir::UnitAttr>());
+    assert(mlir::dyn_cast_or_null<mlir::UnitAttr>(attr));
     assert((t + 1 == conds) && "unit must be last");
     defaultDestination = dest;
     defaultOperands = destOps ? *destOps : mlir::ValueRange{};

flang/lib/Optimizer/Dialect/FIROps.cpp

@@ -2498,10 +2498,8 @@ static constexpr llvm::StringRef getTargetOffsetAttr() {
 
 template <typename OpT>
 static mlir::LogicalResult verifyIntegralSwitchTerminator(OpT op) {
-  if (!op.getSelector()
-           .getType()
-           .template isa<mlir::IntegerType, mlir::IndexType,
-                         fir::IntegerType>())
+  if (!mlir::isa<mlir::IntegerType, mlir::IndexType, fir::IntegerType>(
+          op.getSelector().getType()))
     return op.emitOpError("must be an integer");
   auto cases =
       op->template getAttrOfType<mlir::ArrayAttr>(op.getCasesAttr()).getValue();
@@ -2576,7 +2574,7 @@ static void printIntegralSwitchTerminator(OpT op, mlir::OpAsmPrinter &p) {
     if (i)
       p << ", ";
     auto &attr = cases[i];
-    if (auto intAttr = attr.template dyn_cast_or_null<mlir::IntegerAttr>())
+    if (auto intAttr = mlir::dyn_cast_or_null<mlir::IntegerAttr>(attr))
       p << intAttr.getValue();
     else
       p.printAttribute(attr);

flang/lib/Optimizer/Dialect/FIRType.cpp

@@ -695,9 +695,9 @@ BoxProcType::verify(llvm::function_ref<mlir::InFlightDiagnostic()> emitError,
 }
 
 static bool cannotBePointerOrHeapElementType(mlir::Type eleTy) {
-  return eleTy.isa<BoxType, BoxCharType, BoxProcType, ShapeType, ShapeShiftType,
+  return mlir::isa<BoxType, BoxCharType, BoxProcType, ShapeType, ShapeShiftType,
                    SliceType, FieldType, LenType, HeapType, PointerType,
-                   ReferenceType, TypeDescType>();
+                   ReferenceType, TypeDescType>(eleTy);
 }
 
 //===----------------------------------------------------------------------===//
@@ -776,10 +776,10 @@ void fir::CharacterType::print(mlir::AsmPrinter &printer) const {
 mlir::LogicalResult
 fir::ClassType::verify(llvm::function_ref<mlir::InFlightDiagnostic()> emitError,
                        mlir::Type eleTy) {
-  if (eleTy.isa<fir::RecordType, fir::SequenceType, fir::HeapType,
+  if (mlir::isa<fir::RecordType, fir::SequenceType, fir::HeapType,
                 fir::PointerType, mlir::NoneType, mlir::IntegerType,
                 mlir::FloatType, fir::CharacterType, fir::LogicalType,
-                fir::ComplexType, mlir::ComplexType>())
+                fir::ComplexType, mlir::ComplexType>(eleTy))
     return mlir::success();
   return emitError() << "invalid element type\n";
 }
@@ -1050,8 +1050,8 @@ void fir::ReferenceType::print(mlir::AsmPrinter &printer) const {
 mlir::LogicalResult fir::ReferenceType::verify(
     llvm::function_ref<mlir::InFlightDiagnostic()> emitError,
     mlir::Type eleTy) {
-  if (eleTy.isa<ShapeType, ShapeShiftType, SliceType, FieldType, LenType,
-                ReferenceType, TypeDescType>())
+  if (mlir::isa<ShapeType, ShapeShiftType, SliceType, FieldType, LenType,
+                ReferenceType, TypeDescType>(eleTy))
     return emitError() << "cannot build a reference to type: " << eleTy << '\n';
   return mlir::success();
 }
@@ -1126,9 +1126,9 @@ mlir::LogicalResult fir::SequenceType::verify(
     llvm::ArrayRef<int64_t> shape, mlir::Type eleTy,
     mlir::AffineMapAttr layoutMap) {
   // DIMENSION attribute can only be applied to an intrinsic or record type
-  if (eleTy.isa<BoxType, BoxCharType, BoxProcType, ShapeType, ShapeShiftType,
+  if (mlir::isa<BoxType, BoxCharType, BoxProcType, ShapeType, ShapeShiftType,
                 ShiftType, SliceType, FieldType, LenType, HeapType, PointerType,
-                ReferenceType, TypeDescType, SequenceType>())
+                ReferenceType, TypeDescType, SequenceType>(eleTy))
     return emitError() << "cannot build an array of this element type: "
                        << eleTy << '\n';
   return mlir::success();
@@ -1199,9 +1199,9 @@ void fir::TypeDescType::print(mlir::AsmPrinter &printer) const {
 mlir::LogicalResult fir::TypeDescType::verify(
     llvm::function_ref<mlir::InFlightDiagnostic()> emitError,
     mlir::Type eleTy) {
-  if (eleTy.isa<BoxType, BoxCharType, BoxProcType, ShapeType, ShapeShiftType,
+  if (mlir::isa<BoxType, BoxCharType, BoxProcType, ShapeType, ShapeShiftType,
                 ShiftType, SliceType, FieldType, LenType, ReferenceType,
-                TypeDescType>())
+                TypeDescType>(eleTy))
     return emitError() << "cannot build a type descriptor of type: " << eleTy
                        << '\n';
   return mlir::success();

flang/lib/Optimizer/Transforms/AddDebugInfo.cpp

@@ -11,6 +11,7 @@
 /// This pass populates some debug information for the module and functions.
 //===----------------------------------------------------------------------===//
 
+#include "DebugTypeGenerator.h"
 #include "flang/Common/Version.h"
 #include "flang/Optimizer/Builder/FIRBuilder.h"
 #include "flang/Optimizer/Builder/Todo.h"
@@ -106,14 +107,27 @@ void AddDebugInfoPass::runOnOperation() {
       filePath = llvm::sys::path::parent_path(funcLoc.getFilename().getValue());
     }
 
-    mlir::StringAttr funcName =
+    mlir::StringAttr fullName =
         mlir::StringAttr::get(context, funcOp.getName());
-    mlir::LLVM::DIBasicTypeAttr bT = mlir::LLVM::DIBasicTypeAttr::get(
-        context, llvm::dwarf::DW_TAG_base_type, "void", /*sizeInBits=*/0,
-        /*encoding=*/1);
-    // FIXME: Provide proper type for subroutine
+    auto result = fir::NameUniquer::deconstruct(funcOp.getName());
+    mlir::StringAttr funcName =
+        mlir::StringAttr::get(context, result.second.name);
+
+    llvm::SmallVector<mlir::LLVM::DITypeAttr> types;
+    fir::DebugTypeGenerator typeGen(module);
+    for (auto resTy : funcOp.getResultTypes()) {
+      auto tyAttr =
+          typeGen.convertType(resTy, fileAttr, cuAttr, funcOp.getLoc());
+      types.push_back(tyAttr);
+    }
+    for (auto inTy : funcOp.getArgumentTypes()) {
+      auto tyAttr = typeGen.convertType(fir::unwrapRefType(inTy), fileAttr,
+                                        cuAttr, funcOp.getLoc());
+      types.push_back(tyAttr);
+    }
+
     mlir::LLVM::DISubroutineTypeAttr subTypeAttr =
-        mlir::LLVM::DISubroutineTypeAttr::get(context, CC, {bT, bT});
+        mlir::LLVM::DISubroutineTypeAttr::get(context, CC, types);
     mlir::LLVM::DIFileAttr funcFileAttr =
         mlir::LLVM::DIFileAttr::get(context, fileName, filePath);
 
@@ -130,11 +144,13 @@ void AddDebugInfoPass::runOnOperation() {
       subprogramFlags =
           subprogramFlags | mlir::LLVM::DISubprogramFlags::Definition;
     }
-    // FIXME: Provide proper line and scopeline.
+    unsigned line = 1;
+    if (auto funcLoc = l.dyn_cast<mlir::FileLineColLoc>())
+      line = funcLoc.getLine();
+
     auto spAttr = mlir::LLVM::DISubprogramAttr::get(
-        context, id, compilationUnit, fileAttr, funcName, funcName,
-        funcFileAttr, /*line=*/1, /*scopeline=*/1, subprogramFlags,
-        subTypeAttr);
+        context, id, compilationUnit, fileAttr, funcName, fullName,
+        funcFileAttr, line, line, subprogramFlags, subTypeAttr);
     funcOp->setLoc(builder.getFusedLoc({funcOp->getLoc()}, spAttr));
   });
 }

flang/lib/Optimizer/Transforms/CMakeLists.txt

@@ -22,6 +22,7 @@ add_flang_library(FIRTransforms
   OMPMarkDeclareTarget.cpp
   VScaleAttr.cpp
   FunctionAttr.cpp
+  DebugTypeGenerator.cpp
 
   DEPENDS
   FIRDialect

flang/lib/Optimizer/Transforms/DebugTypeGenerator.cpp

@@ -0,0 +1,63 @@
+//===-- DebugTypeGenerator.cpp -- type conversion ---------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// Coding style: https://mlir.llvm.org/getting_started/DeveloperGuide/
+//
+//===----------------------------------------------------------------------===//
+
+#define DEBUG_TYPE "flang-debug-type-generator"
+
+#include "DebugTypeGenerator.h"
+#include "llvm/ADT/ScopeExit.h"
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/Support/Debug.h"
+
+namespace fir {
+
+DebugTypeGenerator::DebugTypeGenerator(mlir::ModuleOp m)
+    : module(m), kindMapping(getKindMapping(m)) {
+  LLVM_DEBUG(llvm::dbgs() << "DITypeAttr generator\n");
+}
+
+static mlir::LLVM::DITypeAttr genPlaceholderType(mlir::MLIRContext *context) {
+  return mlir::LLVM::DIBasicTypeAttr::get(
+      context, llvm::dwarf::DW_TAG_base_type, "void", 32, 1);
+}
+
+static mlir::LLVM::DITypeAttr genBasicType(mlir::MLIRContext *context,
+                                           mlir::StringAttr name,
+                                           unsigned bitSize,
+                                           unsigned decoding) {
+  return mlir::LLVM::DIBasicTypeAttr::get(
+      context, llvm::dwarf::DW_TAG_base_type, name, bitSize, decoding);
+}
+
+mlir::LLVM::DITypeAttr
+DebugTypeGenerator::convertType(mlir::Type Ty, mlir::LLVM::DIFileAttr fileAttr,
+                                mlir::LLVM::DIScopeAttr scope,
+                                mlir::Location loc) {
+  mlir::MLIRContext *context = module.getContext();
+  if (Ty.isIntOrIndex()) {
+    return genBasicType(context, mlir::StringAttr::get(context, "integer"),
+                        Ty.getIntOrFloatBitWidth(), llvm::dwarf::DW_ATE_signed);
+  } else if (Ty.isa<mlir::FloatType>() || Ty.isa<fir::RealType>()) {
+    return genBasicType(context, mlir::StringAttr::get(context, "real"),
+                        Ty.getIntOrFloatBitWidth(), llvm::dwarf::DW_ATE_float);
+  } else if (auto logTy = Ty.dyn_cast_or_null<fir::LogicalType>()) {
+    return genBasicType(context,
+                        mlir::StringAttr::get(context, logTy.getMnemonic()),
+                        kindMapping.getLogicalBitsize(logTy.getFKind()),
+                        llvm::dwarf::DW_ATE_boolean);
+  } else {
+    // FIXME: These types are currently unhandled. We are generating a
+    // placeholder type to allow us to test supported bits.
+    return genPlaceholderType(context);
+  }
+}
+
+} // namespace fir