Initial analyzer/summarizer commit

.gitignore

@@ -34,3 +34,143 @@
 # Build folders
 [Bb]uild/
 cmake-*/
+
+## Python begins here.
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+.pybuilder/
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+#   For a library or package, you might want to ignore these files since the code is
+#   intended to run in multiple environments; otherwise, check them in:
+# .python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static type analyzer
+.pytype/
+
+# Cython debug symbols
+cython_debug/

op-finder/OperationFinder.cpp

@@ -30,6 +30,7 @@ void OperationFinder::processArithmetic(const clang::BinaryOperator* op, const c
   log.line = line_number;
   log.operation = getOpcode(op).str();
   log.current_for_loops = _for_loop_stack;
+  log.is_fallthrough = _in_fallthrough;
 
   _processExpressionTypes(log, op, op->getLHS(), op->getRHS());
 
@@ -56,6 +57,7 @@ void OperationFinder::processUnaryArithmetic(const clang::UnaryOperator* op, con
   log.line = line_number;
   log.operation = getOpcode(op).str();
   log.current_for_loops = _for_loop_stack;
+  log.is_fallthrough = _in_fallthrough;
 
   _processExpressionTypes(log, op, lhs, nullptr);
 
@@ -64,6 +66,16 @@ void OperationFinder::processUnaryArithmetic(const clang::UnaryOperator* op, con
   llvm::outs() << "\n";
 }
 
+void OperationFinder::fallthroughBranchEntered()
+{
+  _in_fallthrough = true;
+}
+
+void OperationFinder::fallthroughBranchExited()
+{
+  _in_fallthrough = false;
+}
+
 void OperationFinder::forLoopEntered()
 {
   _for_loop_stack.push_back(_next_for_loop_id);

op-finder/OperationFinder.hpp

@@ -17,6 +17,9 @@ public:
   void processArithmetic(const clang::BinaryOperator* op, const clang::SourceManager& source_manager);
   void processUnaryArithmetic(const clang::UnaryOperator* op, const clang::SourceManager& source_manager);
 
+  void fallthroughBranchEntered();
+  void fallthroughBranchExited();
+
   void forLoopEntered();
   void forLoopExited();
 
@@ -24,6 +27,7 @@ private:
   void _processExpressionTypes(OperationLog& log, const clang::Expr* source, const clang::Expr* op1, const clang::Expr* op2);
 
   int _next_for_loop_id = 0;
+  bool _in_fallthrough = false;
   std::vector<int> _for_loop_stack;
   IOperationOutput* _storage;
 };

op-finder/OperationFinderAstVisitor.cpp

@@ -52,6 +52,7 @@ bool OperationFinderAstVisitor::dataTraverseStmtPre(clang::Stmt* stmt)
     if (loop->getInit())
     {
       _loop_init = loop->getInit();
+      _op_finder->fallthroughBranchEntered();
     }
     else
     {
@@ -69,6 +70,7 @@ bool OperationFinderAstVisitor::dataTraverseStmtPost(clang::Stmt* stmt)
   if (_loop_init && _loop_init == stmt)
   {
     _op_finder->forLoopEntered();
+    _op_finder->fallthroughBranchExited();
     _loop_init = nullptr;
   }
   else if (clang::dyn_cast<clang::ForStmt>(stmt))

op-finder/OperationLog.hpp

@@ -14,6 +14,7 @@ struct OperationLog
   std::string operand_lhs;
   std::string operand_rhs;
   std::string operand_result;
+  bool is_fallthrough;
   std::vector<int> current_for_loops;
 };
 
@@ -35,6 +36,7 @@ inline void to_json(nlohmann::json& j, const OperationLog& l)
           {"operand_lhs", l.operand_lhs},
           {"operand_rhs", l.operand_rhs},
           {"operand_result", l.operand_result},
+          {"is_fallthrough", l.is_fallthrough},
           {"current_for_loops", l.current_for_loops}
   };
 }
@@ -46,6 +48,7 @@ inline void from_json(const nlohmann::json& j, OperationLog& l)
   j.at("operand_lhs").get_to(l.operand_lhs);
   j.at("operand_rhs").get_to(l.operand_rhs);
   j.at("operand_result").get_to(l.operand_result);
+  j.at("is_fallthrough").get_to(l.is_fallthrough);
   j.at("current_for_loops").get_to(l.current_for_loops);
 }
 

op-summarizer/gcovreader.py

@@ -0,0 +1,36 @@
+import json
+
+from dataclasses import dataclass
+from typing import Dict, List
+
+
+@dataclass
+class GCovLine:
+    line_number: int
+    count: int
+    is_fallthrough: bool = False
+
+
+class GCovFile:
+    def __init__(self, path: str) -> None:
+        self._path = path
+        self._data: dict = {}
+
+        self.files: Dict[str, List[GCovLine]] = {}
+
+    def read(self) -> None:
+        with open(self._path, "r") as infile:
+            self._data = json.load(infile)
+
+        for file in self._data["files"]:
+            name: str = file["file"]
+            lines: List[GCovLine] = []
+
+            for line in file["lines"]:
+                if len(line["branches"]):
+                    for branch in line["branches"]:
+                        lines.append(GCovLine(line["line_number"], branch["count"], branch["fallthrough"]))
+                else:
+                    lines.append(GCovLine(line["line_number"], line["count"]))
+
+            self.files[name] = lines

op-summarizer/op-summarizer.py

@@ -0,0 +1,58 @@
+from dataclasses import dataclass
+from typing import Dict, List, Optional
+
+from gcovreader import GCovFile, GCovLine
+from opfinderreader import OperationLogReader, OperationLog
+
+
+@dataclass(frozen=True)
+class UniqueOperation:
+    name: str
+    type_lhs: Optional[str]
+    type_rhs: Optional[str]
+    type_expr: Optional[str]
+
+    @staticmethod
+    def from_operation_log(op_log: OperationLog) -> "UniqueOperation":
+        return UniqueOperation(
+            op_log.operation,
+            op_log.operand_lhs,
+            op_log.operand_rhs,
+            op_log.operand_result
+        )
+
+
+if __name__ == "__main__":
+    gcov = GCovFile("./data/gcov.json")
+    ops = OperationLogReader("./data/opfinder.json")
+
+    gcov.read()
+    ops.read()
+
+    for file_name in gcov.files:
+        op_counter: Dict[UniqueOperation, int] = {}
+
+        if file_name not in ops.files:
+            print(f"Couldn't find {file_name} in op-finder output. Skipping.")
+            continue
+
+        loop_stack: List[int] = []
+
+        for gcov_line in gcov.files[file_name]:
+            op_lines = ops.get_lines(file_name, gcov_line.line_number)
+            for opfinder_line in op_lines:
+                # TODO: revise this. Need a special case for for-loop clauses
+                # or branching in general.
+                if opfinder_line.is_fallthrough != gcov_line.is_fallthrough:
+                    continue
+
+                unique_op = UniqueOperation.from_operation_log(opfinder_line)
+
+                if unique_op in op_counter:
+                    op_counter[unique_op] += gcov_line.count
+                else:
+                    op_counter[unique_op] = gcov_line.count
+
+        print(f"Unique operations for file {file_name}:")
+        for uop, count in op_counter.items():
+            print(f"\t{count}: {uop}")

op-summarizer/opfinderreader.py

@@ -0,0 +1,43 @@
+import json
+
+from dataclasses import dataclass, field
+from typing import Dict, List
+
+
+@dataclass
+class OperationLog:
+    operation: str
+    line: int
+    operand_lhs: str
+    operand_rhs: str
+    operand_result: str
+    is_fallthrough: bool
+    current_for_loops: list[int] = field(default_factory=list)
+
+
+class OperationLogReader:
+    def __init__(self, path: str) -> None:
+        self._path = path
+        self._data: dict = {}
+
+        self.files: Dict[str, List[OperationLog]] = {}
+
+    def read(self) -> None:
+        with open(self._path, "r") as infile:
+            self._data = json.load(infile)
+
+        for name, ops_list in self._data.items():
+            ops: List[OperationLog] = []
+
+            for op_json in ops_list:
+                ops.append(OperationLog(**op_json))
+
+            self.files[name] = ops
+
+    def get_lines(self, file: str, line_number: int) -> List[OperationLog]:
+        res: List[OperationLog] = []
+        for line in self.files[file]:
+            if line.line == line_number:
+                res.append(line)
+
+        return res