[2/N] Enable clang-tidy checks in torch/csrc/profiler (#113439)

Fixes #ISSUE_NUMBER

Pull Request resolved: https://github.com/pytorch/pytorch/pull/113439
Approved by: https://github.com/Skylion007

.lintrunner.toml

@@ -266,7 +266,6 @@ exclude_patterns = [
     'torch/csrc/jit/serialization/export.cpp',
     'torch/csrc/lazy/**/*',
     'torch/csrc/onnx/init.cpp',
-    'torch/csrc/profiler/**/*',
     'torch/csrc/quantized/**/*',
     'torch/csrc/mps/**/*',
 ]

torch/csrc/profiler/collection.h

@@ -47,9 +47,9 @@ struct TORCH_API RawTensorMetadataBase {
   explicit RawTensorMetadataBase(const at::Tensor& t);
 
   StorageImplData data_;
-  c10::ScalarType dtype_;
+  c10::ScalarType dtype_{c10::ScalarType::Undefined};
-  c10::Layout layout_;
+  c10::Layout layout_{c10::Layout::Strided};
-  uint32_t dim_;
+  uint32_t dim_{0};
 };
 
 // Collected during profiling.
@@ -64,8 +64,8 @@ struct TORCH_API RawTensorMetadata : RawTensorMetadataBase {
   // Wrap `weak_self_` in `c10::optional` and split device into components to
   // keep struct default constructable. (which the std::array initializer needs)
   c10::optional<WeakTensor> weak_self_;
-  c10::DeviceType device_type_;
+  c10::DeviceType device_type_{c10::DeviceType::CPU};
-  c10::DeviceIndex device_index_;
+  c10::DeviceIndex device_index_{-1};
 };
 
 // Used during post processing.
@@ -367,8 +367,8 @@ struct TORCH_API Result : public std::enable_shared_from_this<Result> {
   template <typename T, typename Fn>
   void visit_if_base(Fn&& fn) const {
     visit([&](const auto& extra_fields) {
-      using extra_fields_t = typename std::remove_cv<
+      using extra_fields_t = typename std::remove_cv_t<
-          typename std::remove_reference<decltype(extra_fields)>::type>::type;
+          typename std::remove_reference_t<decltype(extra_fields)>>;
 
       if constexpr (std::is_base_of_v<T, extra_fields_t>) {
         fn(extra_fields);

torch/csrc/profiler/combined_traceback.cpp

@@ -90,6 +90,7 @@ SymbolizedTracebacks symbolize(
   for (const auto& e : to_symbolize) {
     if (e->python_) {
       if (cur_python != e->python_ && !cur_py_frames.empty()) {
+        // NOLINTNEXTLINE(clang-analyzer-core.CallAndMessage)
         cur_python->appendSymbolized(cur_py_frames, r);
         cur_py_frames.clear();
       }
@@ -103,11 +104,13 @@ SymbolizedTracebacks symbolize(
     }
   }
   if (!cur_py_frames.empty()) {
+    // NOLINTNEXTLINE(clang-analyzer-core.CallAndMessage)
     cur_python->appendSymbolized(cur_py_frames, r);
     cur_py_frames.clear();
   }
   std::vector<std::vector<uint64_t>> python_frame_fragments =
       std::move(r.tracebacks);
+  r.tracebacks = {};
 
   for (const auto& sc : to_symbolize) {
     r.tracebacks.emplace_back();

torch/csrc/profiler/combined_traceback.h

@@ -29,7 +29,7 @@ struct TORCH_API CapturedTraceback : public c10::GatheredContext {
   CapturedTraceback() = default;
   CapturedTraceback(const CapturedTraceback&) = delete;
   CapturedTraceback& operator=(const CapturedTraceback&) = delete;
-  ~CapturedTraceback();
+  ~CapturedTraceback() override;
 
   using visitproc = int (*)(void* self, void* arg);
 

torch/csrc/profiler/containers.h

@@ -7,7 +7,6 @@
 #include <forward_list>
 #include <new>
 #include <utility>
-#include <vector>
 
 #include <c10/macros/Macros.h>
 #include <c10/util/ArrayRef.h>
@@ -49,7 +48,7 @@ class AppendOnlyList {
  public:
   using array_t = block_t<T, ChunkSize>;
   static_assert(
-      std::is_base_of<std::array<T, ChunkSize>, array_t>::value,
+      std::is_base_of_v<std::array<T, ChunkSize>, array_t>,
       "AppendOnlyList expects raw low level pointer storage.");
   static_assert(ChunkSize > 0, "Block cannot be empty.");
 

torch/csrc/profiler/data_flow.cpp

@@ -10,6 +10,7 @@ namespace impl {
 namespace {
 static constexpr TensorImplAddress NoTensorImpl{nullptr};
 
+// NOLINTNEXTLINE(cppcoreguidelines-pro-type-member-init)
 struct RawTensorInfo {
   TensorImplAddress impl_;
   StorageImplData storage_;

torch/csrc/profiler/events.h

@@ -1,6 +1,7 @@
 #pragma once
 
 #include <array>
+#include <cstdint>
 #include <cstring>
 #include <vector>
 

torch/csrc/profiler/kineto_client_interface.cpp

@@ -1,7 +1,6 @@
 #ifdef USE_KINETO
 #include <libkineto.h>
 #include <torch/csrc/autograd/profiler_kineto.h>
-#include <cstdlib>
 
 // Ondemand tracing is not supported on Apple or edge platform
 #if defined(__APPLE__) || defined(EDGE_PROFILER_USE_KINETO)

torch/csrc/profiler/kineto_shim.cpp

@@ -1,8 +1,6 @@
 #include <torch/csrc/profiler/collection.h>
 #include <torch/csrc/profiler/kineto_shim.h>
 
-#include <type_traits>
-
 #ifdef USE_KINETO
 #include <libkineto.h>
 #endif

torch/csrc/profiler/orchestration/observer.cpp

@@ -94,8 +94,8 @@ ProfilerConfig ProfilerConfig::fromIValue(
 // ----------------------------------------------------------------------------
 // -- Profiler base class -----------------------------------------------------
 // ----------------------------------------------------------------------------
-/*explicit*/ ProfilerStateBase::ProfilerStateBase(const ProfilerConfig& config)
+/*explicit*/ ProfilerStateBase::ProfilerStateBase(ProfilerConfig config)
-    : c10::MemoryReportingInfoBase(), config_(config) {}
+    : c10::MemoryReportingInfoBase(), config_(std::move(config)) {}
 
 ProfilerStateBase::~ProfilerStateBase() {
   if (handle_) {

torch/csrc/profiler/orchestration/observer.h

@@ -108,7 +108,7 @@ struct TORCH_API ProfilerConfig {
 // -- Profiler base class -----------------------------------------------------
 // ----------------------------------------------------------------------------
 struct TORCH_API ProfilerStateBase : public c10::MemoryReportingInfoBase {
-  explicit ProfilerStateBase(const ProfilerConfig& config);
+  explicit ProfilerStateBase(ProfilerConfig config);
   ~ProfilerStateBase() override;
 
   static ProfilerStateBase* get(bool global);

torch/csrc/profiler/python/combined_traceback.cpp

@@ -2,8 +2,6 @@
 #include <torch/csrc/python_headers.h>
 #include <torch/csrc/utils/pybind.h>
 #include <torch/csrc/utils/pythoncapi_compat.h>
-#include <iostream>
-
 namespace py = pybind11;
 
 namespace torch {

torch/csrc/profiler/python/init.cpp

@@ -56,6 +56,7 @@ PyTypeObject THPCapturedTracebackType = {
     nullptr, /* tp_getattro */
     nullptr, /* tp_setattro */
     nullptr, /* tp_as_buffer */
+    // NOLINTNEXTLINE(misc-redundant-expression)
     Py_TPFLAGS_DEFAULT | Py_TPFLAGS_HAVE_GC, /* tp_flags */
     nullptr, /* tp_doc */
     (traverseproc)THPCapturedTraceback_traverse, /* tp_traverse */

torch/csrc/profiler/standalone/execution_trace_observer.cpp

@@ -10,7 +10,6 @@
 #endif // _WIN32
 
 #include <fmt/format.h>
-#include <algorithm>
 #include <chrono>
 #include <cmath>
 #include <fstream>
@@ -19,7 +18,6 @@
 #include <mutex>
 #include <sstream>
 #include <stack>
-#include <stdexcept>
 #include <vector>
 
 #include <ATen/core/TensorBody.h>

torch/csrc/profiler/unwind/unwind.cpp

@@ -59,6 +59,7 @@ struct UpgradeExclusive {
   }
 
  private:
+  // NOLINTNEXTLINE(cppcoreguidelines-avoid-const-or-ref-data-members)
   std::shared_lock<std::shared_timed_mutex>& rdlock_;
 };
 
@@ -97,6 +98,7 @@ struct LibraryInfo {
 };
 
 static const char* process_name() {
+  // NOLINTNEXTLINE(*-c-arrays*)
   static char name[PATH_MAX + 1] = "";
   if (*name == '\0') {
     ssize_t len = readlink("/proc/self/exe", name, PATH_MAX);
@@ -145,6 +147,7 @@ struct UnwindCache {
                 library_name = process_name();
               }
               auto eh_frame_hdr =
+                  // NOLINTNEXTLINE(performance-no-int-to-ptr)
                   (void*)(segments[i].p_vaddr + info->dlpi_addr);
               self->all_libraries_.emplace_back(
                   std::move(library_name),
@@ -272,6 +275,7 @@ extern "C" void unwind_c(std::vector<void*>* result, int64_t rsp, int64_t rbp);
 extern "C" void unwind_c(std::vector<void*>* result, int64_t rsp, int64_t rbp) {
   std::shared_lock lock(cache_mutex_);
   UnwindState state{};
+  // NOLINTNEXTLINE(performance-no-int-to-ptr)
   state.rip = *(int64_t*)(rsp);
   // +8 because we saved rsp after the return address was already pushed
   // to the stack
@@ -279,6 +283,7 @@ extern "C" void unwind_c(std::vector<void*>* result, int64_t rsp, int64_t rbp) {
   state.rbp = rbp;
   unwind_cache.checkRefresh(lock);
   while (true) { // unwind for _start sets rip as being undefined
+    // NOLINTNEXTLINE(performance-no-int-to-ptr)
     result->push_back((void*)state.rip);
     const Unwinder& uw = unwind_cache.unwinderFor(state.rip, lock);
     if (uw.terminator()) {
@@ -355,6 +360,7 @@ struct Symbolizer {
     auto& entry = getOrCreate(maybe_library->first);
     entry.queried.push_back(addr);
     auto libaddress = maybe_library->second - 1;
+    // NOLINTNEXTLINE(performance-no-int-to-ptr)
     entry.comm->out() << (void*)libaddress << "\n";
     // we need to make sure we don't write more than 64k bytes to
     // a pipe before reading the results. Otherwise the buffer may
@@ -393,6 +399,7 @@ struct Symbolizer {
   Entry& getOrCreate(const std::string& name) {
     auto it = entries_.find(name);
     if (it == entries_.end()) {
+      // NOLINTNEXTLINE(*-c-arrays*)
       const char* args[] = {
           "addr2line", "-C", "-f", "-e", name.c_str(), nullptr};
       it = entries_

torch/csrc/profiler/unwind/unwinder.h

@@ -1,7 +1,7 @@
 #pragma once
-#include <stdint.h>
 #include <torch/csrc/profiler/unwind/action.h>
 #include <torch/csrc/profiler/unwind/unwind_error.h>
+#include <cstdint>
 #include <limits>
 
 struct UnwindState {
@@ -54,10 +54,13 @@ struct Unwinder {
     r.rsp = (reg_ == D_RSP ? cur.rsp : cur.rbp) + off_;
     r.rbp = rbp_off_ == std::numeric_limits<int64_t>::max()
         ? cur.rbp
+        // NOLINTNEXTLINE(performance-no-int-to-ptr)
         : *(int64_t*)(r.rsp + rbp_off_);
     if (deref_) {
+      // NOLINTNEXTLINE(performance-no-int-to-ptr)
       r.rsp = *(int64_t*)r.rsp;
     }
+    // NOLINTNEXTLINE(performance-no-int-to-ptr)
     r.rip = *(int64_t*)(r.rsp + rip_off_);
 
     return r;
@@ -70,5 +73,5 @@ struct Unwinder {
   int64_t off_;
   int64_t rip_off_;
   int64_t rbp_off_;
-  bool deref_;
+  bool deref_{false};
 };