[LifetimeSafety] Add missing origins stats for lifetime analysis

clang/include/clang/Analysis/Analyses/LifetimeSafety/LifetimeSafety.h

@@ -21,9 +21,13 @@
 #define LLVM_CLANG_ANALYSIS_ANALYSES_LIFETIMESAFETY_H
 
 #include "clang/Analysis/Analyses/LifetimeSafety/Facts.h"
+#include "clang/Analysis/Analyses/LifetimeSafety/LifetimeStats.h"
 #include "clang/Analysis/Analyses/LifetimeSafety/LiveOrigins.h"
 #include "clang/Analysis/Analyses/LifetimeSafety/LoanPropagation.h"
+#include "clang/Analysis/Analyses/LifetimeSafety/Origins.h"
 #include "clang/Analysis/AnalysisDeclContext.h"
+#include "llvm/Support/raw_ostream.h"
+#include <string>
 
 namespace clang::lifetimes {
 
@@ -51,9 +55,14 @@ class LifetimeSafetyReporter {
 
 /// The main entry point for the analysis.
 void runLifetimeSafetyAnalysis(AnalysisDeclContext &AC,
-                               LifetimeSafetyReporter *Reporter);
+                               LifetimeSafetyReporter *Reporter,
+                               LifetimeSafetyStats &Stats, bool CollectStats);
 
 namespace internal {
+
+void collectLifetimeStats(AnalysisDeclContext &AC, OriginManager &OM,
+                          LifetimeSafetyStats &Stats);
+
 /// An object to hold the factories for immutable collections, ensuring
 /// that all created states share the same underlying memory management.
 struct LifetimeFactory {
@@ -65,6 +74,7 @@ struct LifetimeFactory {
 /// Running the lifetime safety analysis and querying its results. It
 /// encapsulates the various dataflow analyses.
 class LifetimeSafetyAnalysis {
+
 public:
   LifetimeSafetyAnalysis(AnalysisDeclContext &AC,
                          LifetimeSafetyReporter *Reporter);

clang/include/clang/Analysis/Analyses/LifetimeSafety/LifetimeStats.h

@@ -0,0 +1,33 @@
+//===- LifetimeStats.h - Miscellaneous statistics related to lifetime safety
+//analysis ---*- 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
+//
+//===----------------------------------------------------------------------------------------------===//
+//
+// This file declares the data structures and utility function for collection of
+// staticstics related to Lifetimesafety analysis.
+//
+//===----------------------------------------------------------------------------------------------===//
+
+#ifndef LLVM_CLANG_ANALYSIS_ANALYSES_LIFETIMESAFETY_LIFETIMESTATS_H
+#define LLVM_CLANG_ANALYSIS_ANALYSES_LIFETIMESAFETY_LIFETIMESTATS_H
+
+#include "llvm/ADT/StringMap.h"
+
+namespace clang::lifetimes {
+/// A structure to hold the statistics related to LifetimeAnalysis.
+/// Currently it holds only the missing origin details.
+struct LifetimeSafetyStats {
+  /// A map from `ExpressionClassName<QualType>` to their missing origin
+  /// counts.
+  llvm::StringMap<unsigned> MissingOriginCount;
+};
+
+// utility function to print missing origin stats.
+void printStats(const LifetimeSafetyStats &Stats);
+} // namespace clang::lifetimes
+
+#endif // LLVM_CLANG_ANALYSIS_ANALYSES_LIFETIMESAFETY_LIFETIMESTATS_H

clang/include/clang/Analysis/Analyses/LifetimeSafety/Origins.h

@@ -16,7 +16,10 @@
 
 #include "clang/AST/Decl.h"
 #include "clang/AST/Expr.h"
+#include "clang/AST/TypeBase.h"
+#include "clang/Analysis/Analyses/LifetimeSafety/LifetimeStats.h"
 #include "clang/Analysis/Analyses/LifetimeSafety/Utils.h"
+#include "llvm/Support/raw_ostream.h"
 
 namespace clang::lifetimes::internal {
 
@@ -78,6 +81,9 @@ class OriginManager {
 
   void dump(OriginID OID, llvm::raw_ostream &OS) const;
 
+  /// Collects statistics about expressions that lack associated origins.
+  void collectMissingOrigins(Stmt *FunctionBody, LifetimeSafetyStats &LSStats);
+
 private:
   OriginID getNextOriginID() { return NextOriginID++; }
 

clang/include/clang/Sema/AnalysisBasedWarnings.h

@@ -14,9 +14,11 @@
 #define LLVM_CLANG_SEMA_ANALYSISBASEDWARNINGS_H
 
 #include "clang/AST/Decl.h"
+#include "clang/Analysis/Analyses/LifetimeSafety/Facts.h"
+#include "clang/Analysis/Analyses/LifetimeSafety/LifetimeSafety.h"
+#include "clang/Analysis/Analyses/LifetimeSafety/LifetimeStats.h"
+#include "clang/Analysis/AnalysisDeclContext.h"
 #include "clang/Sema/ScopeInfo.h"
-#include "llvm/ADT/DenseMap.h"
-#include "llvm/ADT/MapVector.h"
 #include <memory>
 
 namespace clang {
@@ -101,6 +103,11 @@ class AnalysisBasedWarnings {
   /// a single function.
   unsigned MaxUninitAnalysisBlockVisitsPerFunction;
 
+  /// Statistics collected during lifetime safety analysis.
+  /// These are accumulated across all analyzed functions and printed
+  /// when -print-stats is enabled.
+  clang::lifetimes::LifetimeSafetyStats LSStats;
+
   /// @}
 
 public:

clang/lib/Analysis/LifetimeSafety/CMakeLists.txt

@@ -5,6 +5,7 @@ add_clang_library(clangAnalysisLifetimeSafety
   LifetimeAnnotations.cpp
   LifetimeSafety.cpp
   LiveOrigins.cpp
+  LifetimeStats.cpp
   Loans.cpp
   LoanPropagation.cpp
   Origins.cpp

clang/lib/Analysis/LifetimeSafety/LifetimeSafety.cpp

@@ -18,8 +18,10 @@
 #include "clang/Analysis/Analyses/LifetimeSafety/Checker.h"
 #include "clang/Analysis/Analyses/LifetimeSafety/Facts.h"
 #include "clang/Analysis/Analyses/LifetimeSafety/FactsGenerator.h"
+#include "clang/Analysis/Analyses/LifetimeSafety/LifetimeStats.h"
 #include "clang/Analysis/Analyses/LifetimeSafety/LiveOrigins.h"
 #include "clang/Analysis/Analyses/LifetimeSafety/LoanPropagation.h"
+#include "clang/Analysis/Analyses/LifetimeSafety/Origins.h"
 #include "clang/Analysis/AnalysisDeclContext.h"
 #include "clang/Analysis/CFG.h"
 #include "llvm/ADT/FoldingSet.h"
@@ -68,11 +70,22 @@ void LifetimeSafetyAnalysis::run() {
 
   runLifetimeChecker(*LoanPropagation, *LiveOrigins, FactMgr, AC, Reporter);
 }
+
+void collectLifetimeStats(AnalysisDeclContext &AC, OriginManager &OM,
+                          LifetimeSafetyStats &Stats) {
+  if (!AC.getBody())
+    return;
+  Stmt *FunctionBody = AC.getBody();
+  OM.collectMissingOrigins(FunctionBody, Stats);
+}
 } // namespace internal
 
 void runLifetimeSafetyAnalysis(AnalysisDeclContext &AC,
-                               LifetimeSafetyReporter *Reporter) {
+                               LifetimeSafetyReporter *Reporter,
+                               LifetimeSafetyStats &Stats, bool CollectStats) {
   internal::LifetimeSafetyAnalysis Analysis(AC, Reporter);
   Analysis.run();
+  if (CollectStats)
+    collectLifetimeStats(AC, Analysis.getFactManager().getOriginMgr(), Stats);
 }
 } // namespace clang::lifetimes

clang/lib/Analysis/LifetimeSafety/LifetimeStats.cpp

@@ -0,0 +1,30 @@
+//===- LifetimeStats.cpp - Miscellaneous statistics related to C++ Lifetime
+//Safety analysis -*--------- 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
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines the data structures and utility function for collection of
+// staticstics related to Lifetimesafety analysis.
+//
+//===----------------------------------------------------------------------===//
+
+#include "clang/Analysis/Analyses/LifetimeSafety/LifetimeStats.h"
+#include "llvm/Support/raw_ostream.h"
+
+namespace clang::lifetimes {
+void printStats(const LifetimeSafetyStats &Stats) {
+  llvm::errs() << "\n*** LifetimeSafety Missing Origin Stats "
+                  "(expression_type : count) :\n\n";
+  unsigned TotalMissingOrigins = 0;
+  for (const auto &[expr, count] : Stats.MissingOriginCount) {
+    llvm::errs() << expr << " : " << count << '\n';
+    TotalMissingOrigins += count;
+  }
+  llvm::errs() << "Total missing origins: " << TotalMissingOrigins << "\n";
+  llvm::errs() << "\n****************************************\n";
+}
+} // namespace clang::lifetimes

clang/lib/Analysis/LifetimeSafety/Origins.cpp

@@ -7,8 +7,51 @@
 //===----------------------------------------------------------------------===//
 
 #include "clang/Analysis/Analyses/LifetimeSafety/Origins.h"
+#include "clang/AST/Expr.h"
+#include "clang/AST/RecursiveASTVisitor.h"
+#include "clang/AST/TypeBase.h"
+#include "clang/Analysis/Analyses/LifetimeSafety/LifetimeAnnotations.h"
+#include "clang/Analysis/Analyses/LifetimeSafety/LifetimeStats.h"
+#include "llvm/ADT/StringMap.h"
 
 namespace clang::lifetimes::internal {
+namespace {
+bool isPointerType(QualType QT) {
+  return QT->isPointerOrReferenceType() || isGslPointerType(QT);
+}
+// Check if a type has an origin.
+bool hasOrigin(const Expr *E) {
+  return E->isGLValue() || isPointerType(E->getType());
+}
+/// An utility class to traverse the function body in the analysis
+/// context and collect the count of expressions with missing origins.
+class MissingOriginCollector
+    : public RecursiveASTVisitor<MissingOriginCollector> {
+public:
+  MissingOriginCollector(
+      const llvm::DenseMap<const clang::Expr *, OriginID> &ExprToOriginId,
+      llvm::StringMap<unsigned> &MissingOriginCount)
+      : ExprToOriginId(ExprToOriginId), MissingOriginCount(MissingOriginCount) {
+  }
+  bool VisitExpr(Expr *E) {
+    if (!hasOrigin(E))
+      return true;
+    // Check if we have an origin for this expression.
+    auto It = this->ExprToOriginId.find(E);
+    if (It == this->ExprToOriginId.end()) {
+      // No origin found: count this as missing origin.
+      std::string ExprStr = std::string(E->getStmtClassName()) +
+                            " (Type: " + E->getType().getAsString() + ")";
+      MissingOriginCount[ExprStr]++;
+    }
+    return true;
+  }
+
+private:
+  const llvm::DenseMap<const clang::Expr *, OriginID> &ExprToOriginId;
+  llvm::StringMap<unsigned> &MissingOriginCount;
+};
+} // namespace
 
 void OriginManager::dump(OriginID OID, llvm::raw_ostream &OS) const {
   OS << OID << " (";
@@ -88,4 +131,14 @@ OriginID OriginManager::getOrCreate(const ValueDecl &D) {
   return NewID;
 }
 
+void OriginManager::collectMissingOrigins(Stmt *FunctionBody,
+                                          LifetimeSafetyStats &LSStats) {
+  if (!FunctionBody)
+    return;
+
+  MissingOriginCollector Collector(this->ExprToOriginID,
+                                   LSStats.MissingOriginCount);
+  Collector.TraverseStmt(const_cast<Stmt *>(FunctionBody));
+}
+
 } // namespace clang::lifetimes::internal

clang/lib/Sema/AnalysisBasedWarnings.cpp

@@ -47,12 +47,12 @@
 #include "clang/Sema/SemaInternal.h"
 #include "llvm/ADT/ArrayRef.h"
 #include "llvm/ADT/BitVector.h"
-#include "llvm/ADT/DenseMap.h"
 #include "llvm/ADT/MapVector.h"
 #include "llvm/ADT/STLFunctionalExtras.h"
 #include "llvm/ADT/SmallVector.h"
 #include "llvm/ADT/StringRef.h"
 #include "llvm/Support/Debug.h"
+#include "llvm/Support/raw_ostream.h"
 #include <algorithm>
 #include <deque>
 #include <iterator>
@@ -3107,7 +3107,8 @@ void clang::sema::AnalysisBasedWarnings::IssueWarnings(
   if (EnableLifetimeSafetyAnalysis && S.getLangOpts().CPlusPlus) {
     if (AC.getCFG()) {
       lifetimes::LifetimeSafetyReporterImpl LifetimeSafetyReporter(S);
-      lifetimes::runLifetimeSafetyAnalysis(AC, &LifetimeSafetyReporter);
+      lifetimes::runLifetimeSafetyAnalysis(AC, &LifetimeSafetyReporter, LSStats,
+                                           S.CollectStats);
     }
   }
   // Check for violations of "called once" parameter properties.
@@ -3203,4 +3204,5 @@ void clang::sema::AnalysisBasedWarnings::PrintStats() const {
                << " average block visits per function.\n"
                << "  " << MaxUninitAnalysisBlockVisitsPerFunction
                << " max block visits per function.\n";
+  clang::lifetimes::printStats(LSStats);
 }