[GSYM] Callsites: Add data format support and loading from YAML

Author: Alex B

llvm/include/llvm/DebugInfo/GSYM/CallSiteInfo.h

@@ -0,0 +1,224 @@
+//===- CallSiteInfo.h -------------------------------------------*- 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
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEBUGINFO_GSYM_CALLSITEINFO_H
+#define LLVM_DEBUGINFO_GSYM_CALLSITEINFO_H
+
+#include "llvm/ADT/DenseMap.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/StringSet.h"
+#include "llvm/DebugInfo/GSYM/ExtractRanges.h"
+#include "llvm/Support/YAMLParser.h"
+#include <string>
+#include <unordered_map>
+#include <vector>
+
+namespace llvm {
+class DataExtractor;
+class raw_ostream;
+class StringTableBuilder;
+class CachedHashStringRef;
+
+namespace yaml {
+struct CallSiteYAML;
+struct FunctionYAML;
+struct FunctionsYAML;
+} // namespace yaml
+
+namespace gsym {
+class FileWriter;
+struct FunctionInfo;
+struct CallSiteInfo {
+public:
+  enum Flags : uint8_t {
+    None = 0,
+    // This flag specifies that the call site can only call a function within
+    // the same link unit as the call site.
+    InternalCall = 1 << 0,
+    // This flag specifies that the call site can only call a function outside
+    // the link unit that the call site is in.
+    ExternalCall = 1 << 1,
+  };
+
+  /// The return address of the call site.
+  uint64_t ReturnAddress;
+
+  /// Offsets into the string table for function names regex patterns.
+  std::vector<uint32_t> MatchRegex;
+
+  /// Bitwise OR of CallSiteInfo::Flags values
+  uint8_t Flags;
+
+  /// Decode a CallSiteInfo object from a binary data stream.
+  ///
+  /// \param Data The binary stream to read the data from.
+  /// \param Offset The current offset within the data stream.
+  /// \param BaseAddr The base address for decoding (unused here but included
+  /// for consistency).
+  ///
+  /// \returns A CallSiteInfo or an error describing the issue.
+  static llvm::Expected<CallSiteInfo>
+  decode(DataExtractor &Data, uint64_t &Offset, uint64_t BaseAddr);
+
+  /// Encode this CallSiteInfo object into a FileWriter stream.
+  ///
+  /// \param O The binary stream to write the data to.
+  /// \returns An error object that indicates success or failure.
+  llvm::Error encode(FileWriter &O) const;
+};
+
+struct CallSiteInfoCollection {
+public:
+  std::vector<CallSiteInfo> CallSites;
+
+  void clear() { CallSites.clear(); }
+
+  /// Query if a CallSiteInfoCollection object is valid.
+  ///
+  /// \returns True if the collection is not empty.
+  bool isValid() const { return !CallSites.empty(); }
+
+  /// Decode a CallSiteInfoCollection object from a binary data stream.
+  ///
+  /// \param Data The binary stream to read the data from.
+  /// \param BaseAddr The base address for decoding (unused here but included
+  /// for consistency).
+  ///
+  /// \returns A CallSiteInfoCollection or an error describing the issue.
+  static llvm::Expected<CallSiteInfoCollection> decode(DataExtractor &Data,
+                                                       uint64_t BaseAddr);
+
+  /// Encode this CallSiteInfoCollection object into a FileWriter stream.
+  ///
+  /// \param O The binary stream to write the data to.
+  /// \returns An error object that indicates success or failure.
+  llvm::Error encode(FileWriter &O) const;
+};
+
+bool operator==(const CallSiteInfoCollection &LHS,
+                const CallSiteInfoCollection &RHS);
+
+bool operator==(const CallSiteInfo &LHS, const CallSiteInfo &RHS);
+
+class CallSiteInfoLoader {
+public:
+  /// Constructor that initializes the CallSiteInfoLoader with necessary data
+  /// structures.
+  ///
+  /// \param StringOffsetMap A reference to a DenseMap that maps existing string
+  /// offsets to CachedHashStringRef. \param StrTab A reference to a
+  /// StringTableBuilder used for managing looking up and creating new strings.
+  /// \param StringStorage A reference to a StringSet for storing the data for
+  /// generated strings.
+  CallSiteInfoLoader(DenseMap<uint64_t, CachedHashStringRef> &StringOffsetMap,
+                     StringTableBuilder &StrTab, StringSet<> &StringStorage)
+      : StringOffsetMap(StringOffsetMap), StrTab(StrTab),
+        StringStorage(StringStorage) {}
+
+  /// Loads call site information from a YAML file and populates the provided
+  /// FunctionInfo vector.
+  ///
+  /// This method reads the specified YAML file, parses its content, and updates
+  /// the `Funcs` vector with call site information based on the YAML data.
+  ///
+  /// \param Funcs A reference to a vector of FunctionInfo objects to be
+  /// populated.
+  /// \param YAMLFile A StringRef representing the path to the YAML
+  /// file to be loaded.
+  ///
+  /// \returns An `llvm::Error` indicating success or describing any issues
+  /// encountered during the loading process.
+  llvm::Error loadYAML(std::vector<FunctionInfo> &Funcs, StringRef YAMLFile);
+
+private:
+  /// Retrieves an existing string from the StringOffsetMap using the provided
+  /// offset.
+  ///
+  /// \param offset A 32-bit unsigned integer representing the offset of the
+  /// string.
+  ///
+  /// \returns A StringRef corresponding to the string for the given offset.
+  ///
+  /// \note This method asserts that the offset exists in the StringOffsetMap.
+  StringRef stringFromOffset(uint32_t offset) const;
+
+  /// Obtains the offset corresponding to a given string in the StrTab. If the
+  /// string does not already exist, it is created.
+  ///
+  /// \param str A StringRef representing the string for which the offset is
+  /// requested.
+  ///
+  /// \returns A 32-bit unsigned integer representing the offset of the string.
+  uint32_t offsetFromString(StringRef str);
+
+  /// Reads the content of the YAML file specified by `YAMLFile` into
+  /// `yamlContent`.
+  ///
+  /// \param YAMLFile A StringRef representing the path to the YAML file.
+  /// \param Buffer The memory buffer containing the YAML content.
+  ///
+  /// \returns An `llvm::Error` indicating success or describing any issues
+  /// encountered while reading the file.
+  llvm::Error readYAMLFile(StringRef YAMLFile,
+                           std::unique_ptr<llvm::MemoryBuffer> &Buffer);
+
+  /// Parses the YAML content and populates `functionsYAML` with the parsed
+  /// data.
+  ///
+  /// \param Buffer The memory buffer containing the YAML content.
+  /// \param functionsYAML A reference to an llvm::yaml::FunctionsYAML object to
+  /// be populated.
+  ///
+  /// \returns An `llvm::Error` indicating success or describing any issues
+  /// encountered during parsing.
+  llvm::Error parseYAML(llvm::MemoryBuffer &Buffer,
+                        llvm::yaml::FunctionsYAML &functionsYAML);
+
+  /// Builds a map from function names to FunctionInfo pointers based on the
+  /// provided `Funcs` vector.
+  ///
+  /// \param Funcs A reference to a vector of FunctionInfo objects.
+  ///
+  /// \returns An unordered_map mapping function names (std::string) to their
+  /// corresponding FunctionInfo pointers.
+  std::unordered_map<std::string, FunctionInfo *>
+  buildFunctionMap(std::vector<FunctionInfo> &Funcs);
+
+  /// Processes the parsed YAML functions and updates the `FuncMap` accordingly.
+  ///
+  /// \param functionsYAML A constant reference to an llvm::yaml::FunctionsYAML
+  /// object containing parsed YAML data.
+  /// \param FuncMap A reference to an unordered_map mapping function names to
+  /// FunctionInfo pointers.
+  /// \param YAMLFile A StringRef representing the name of the YAML file (used
+  /// for error messages).
+  ///
+  /// \returns An `llvm::Error` indicating success or describing any issues
+  /// encountered during processing.
+  llvm::Error
+  processYAMLFunctions(const llvm::yaml::FunctionsYAML &functionsYAML,
+                       std::unordered_map<std::string, FunctionInfo *> &FuncMap,
+                       StringRef YAMLFile);
+
+  /// Map of existing string offsets to CachedHashStringRef.
+  DenseMap<uint64_t, CachedHashStringRef> &StringOffsetMap;
+
+  /// The gSYM string table builder.
+  StringTableBuilder &StrTab;
+
+  /// The gSYM string storage - we store generated strings here.
+  StringSet<> &StringStorage;
+};
+
+raw_ostream &operator<<(raw_ostream &OS, const CallSiteInfo &CSI);
+raw_ostream &operator<<(raw_ostream &OS, const CallSiteInfoCollection &CSIC);
+
+} // namespace gsym
+} // namespace llvm
+
+#endif // LLVM_DEBUGINFO_GSYM_CALLSITEINFO_H

llvm/include/llvm/DebugInfo/GSYM/FunctionInfo.h

@@ -10,6 +10,7 @@
 #define LLVM_DEBUGINFO_GSYM_FUNCTIONINFO_H
 
 #include "llvm/ADT/SmallString.h"
+#include "llvm/DebugInfo/GSYM/CallSiteInfo.h"
 #include "llvm/DebugInfo/GSYM/ExtractRanges.h"
 #include "llvm/DebugInfo/GSYM/InlineInfo.h"
 #include "llvm/DebugInfo/GSYM/LineTable.h"
@@ -63,7 +64,9 @@ class GsymReader;
 ///   enum InfoType {
 ///     EndOfList = 0u,
 ///     LineTableInfo = 1u,
-///     InlineInfo = 2u
+///     InlineInfo = 2u,
+///     MergedFunctionsInfo = 3u,
+///     CallSiteInfo = 4u
 ///   };
 ///
 /// This stream of tuples is terminated by a "InfoType" whose value is
@@ -73,7 +76,7 @@ class GsymReader;
 /// clients to still parse the format and skip over any data that they don't
 /// understand or want to parse.
 ///
-/// So the function information encoding essientially looks like:
+/// So the function information encoding essentially looks like:
 ///
 /// struct {
 ///   uint32_t Size;
@@ -92,6 +95,7 @@ struct FunctionInfo {
   std::optional<LineTable> OptLineTable;
   std::optional<InlineInfo> Inline;
   std::optional<MergedFunctionsInfo> MergedFunctions;
+  std::optional<CallSiteInfoCollection> CallSites;
   /// If we encode a FunctionInfo during segmenting so we know its size, we can
   /// cache that encoding here so we don't need to re-encode it when saving the
   /// GSYM file.
@@ -107,7 +111,7 @@ struct FunctionInfo {
   /// debug info, we might end up with multiple FunctionInfo objects for the
   /// same range and we need to be able to tell which one is the better object
   /// to use.
-  bool hasRichInfo() const { return OptLineTable || Inline; }
+  bool hasRichInfo() const { return OptLineTable || Inline || CallSites; }
 
   /// Query if a FunctionInfo object is valid.
   ///

llvm/include/llvm/DebugInfo/GSYM/GsymCreator.h

@@ -352,13 +352,22 @@ class GsymCreator {
   /// \param   FI The function info object to emplace into our functions list.
   void addFunctionInfo(FunctionInfo &&FI);
 
+  /// Load call site information from a YAML file.
+  ///
+  /// This function reads call site information from a specified YAML file and
+  /// adds it to the GSYM data.
+  ///
+  /// \param YAMLFile The path to the YAML file containing call site
+  /// information.
+  llvm::Error loadCallSitesFromYAML(StringRef YAMLFile);
+
   /// Organize merged FunctionInfo's
   ///
   /// This method processes the list of function infos (Funcs) to identify and
   /// group functions with overlapping address ranges.
   ///
   /// \param  Out Output stream to report information about how merged
-  /// FunctionInfo's were handeled.
+  /// FunctionInfo's were handled.
   void prepareMergedFunctions(OutputAggregator &Out);
 
   /// Finalize the data in the GSYM creator prior to saving the data out.

llvm/include/llvm/DebugInfo/GSYM/GsymReader.h

@@ -181,6 +181,26 @@ class GsymReader {
   /// \param MFI The object to dump.
   void dump(raw_ostream &OS, const MergedFunctionsInfo &MFI);
 
+  /// Dump a CallSiteInfo object.
+  ///
+  /// This function will output the details of a CallSiteInfo object in a
+  /// human-readable format.
+  ///
+  /// \param OS The output stream to dump to.
+  ///
+  /// \param CSI The CallSiteInfo object to dump.
+  void dump(raw_ostream &OS, const CallSiteInfo &CSI);
+
+  /// Dump a CallSiteInfoCollection object.
+  ///
+  /// This function will iterate over a collection of CallSiteInfo objects and
+  /// dump each one.
+  ///
+  /// \param OS The output stream to dump to.
+  ///
+  /// \param CSIC The CallSiteInfoCollection object to dump.
+  void dump(raw_ostream &OS, const CallSiteInfoCollection &CSIC);
+
   /// Dump a LineTable object.
   ///
   /// This function will convert any string table indexes and file indexes

llvm/lib/DebugInfo/GSYM/CMakeLists.txt

@@ -8,6 +8,7 @@ add_llvm_component_library(LLVMDebugInfoGSYM
   InlineInfo.cpp
   LineTable.cpp
   LookupResult.cpp
+  CallSiteInfo.cpp
   MergedFunctionsInfo.cpp
   ObjectFileTransformer.cpp
   ExtractRanges.cpp