Export of internal Abseil changes.

--
8b7c3bc2fb69608e9b2389b1be0b0de840a4c59d by Derek Mauro <dmauro@google.com>:

Set correct flags for clang-cl.
https://github.com/abseil/abseil-cpp/pull/278

clang-cl produce binaries with MSVC ABI and wants to be as
flag-compatible with pure MSVC as possible, so this leads to all sorts
of weird cases.

clang-cl alias /Wall as clang's -Weverything which is way too verbose,
so it needs /W3 like pure MSVC.

clang-cl only understand GCC style warning flags (-W[no]blah) and just
silent drop MSVC style warning flags (/wd[num]).

clang-cl needs MSVC define flags since it is consuming the same header
files as pure MSVC.

CMake set CMAKE_CXX_COMPILER_ID as Clang when clang-cl is detected, so
need extra if (MSVC) to differentiate it.

We are not doing clang-cl specialization in Bazel as currently there
is no reliable way to detect clang-cl in Bazel..

Other changes:
Add ABSL_ prefix to variable names to avoid name collision in CMake.

PiperOrigin-RevId: 239841297

--
add96c3fc067d5c7b6f016d2ba74725a443a185e by CJ Johnson <johnsoncj@google.com>:

Eventually Storage will need to refer to the type `absl::InlinedVector<...>*`. This can be done via a forward declaration. However, doing so would move the defaulted allocator template parameter to the forward declaration and thus inside an internal file. Instead of doing that, this change gives Storage access to the template and it's parameters so the complete type can be formed without including it.

PiperOrigin-RevId: 239811298

--
b5f5279f1b13b09cae5c745597d64ea1efab146b by CJ Johnson <johnsoncj@google.com>:

Simplify/cleanup the benchmark tests for InlinedVector

PiperOrigin-RevId: 239805767

--
f5991e51b43b13a0ae95025474071f5039a33d27 by Matt Calabrese <calabrese@google.com>:

Update the internal-only IsSwappable traits to be nested inside of namespace absl so that the script to add inline namespaces for LTS releases works with the implementation.

PiperOrigin-RevId: 239622024

--
d1cb234dc5706f033ad56f4eb16d94ac5da80d52 by Abseil Team <absl-team@google.com>:

Mutex: fix tsan annotations

This fixes 2 bugs:
1. We call cond directly in Mutex::AwaitCommon without using EvalConditionAnnotated. As the result we call into user code ignoring synchronization, miss synchronization and report false positives later. Use EvalConditionAnnotated to call cond as we should.

2. We call Mutex invariant ignoring synchronization. Result is the same: we miss synchronization and report false positive races later. Reuse EvalConditionAnnotated to call mutex invariant too.

PiperOrigin-RevId: 239583878

--
52295e4922a9b408fa0dd03d27bc91ccc6645cd7 by Abseil Team <absl-team@google.com>:

Clarify how to obtain the same behavior as std::unordered_map::erase if need be.

PiperOrigin-RevId: 239549513

--
6e76e68ed084fd1247981dbb92677ce8e563b0ec by Jon Cohen <cohenjon@google.com>:

Avoid the -S -B form of `cmake` since it's only supported starting in CMake 3.13

PiperOrigin-RevId: 239473143
GitOrigin-RevId: 8b7c3bc2fb69608e9b2389b1be0b0de840a4c59d
Change-Id: Ib6d356fa1a7435260273df991e65df4149bd5861
pull/286/head
Abseil Team 6 years ago committed by Derek Mauro
parent 253eb74164
commit eab2078b53
  1. 21
      CMake/install_test_project/test.sh
  2. 8
      absl/container/flat_hash_map.h
  3. 8
      absl/container/flat_hash_set.h
  4. 28
      absl/container/inlined_vector.h
  5. 118
      absl/container/inlined_vector_benchmark.cc
  6. 16
      absl/container/internal/inlined_vector.h
  7. 19
      absl/container/internal/raw_hash_set_test.cc
  8. 36
      absl/meta/type_traits.h
  9. 107
      absl/synchronization/mutex.cc

@ -30,27 +30,32 @@ project_dir="${absl_dir}"/CMake/install_test_project
project_build_dir=/buildfs/project-build
install_dir="${project_build_dir}"/install
mkdir -p "${absl_build_dir}"
mkdir -p "${project_build_dir}"
mkdir -p "${install_dir}"
install_absl() {
pushd "${absl_build_dir}"
if [[ "${#}" -eq 1 ]]; then
cmake -DCMAKE_INSTALL_PREFIX="${1}" -B "${absl_build_dir}" -S "${absl_dir}"
cmake -DCMAKE_INSTALL_PREFIX="${1}" "${absl_dir}"
else
cmake -B "${absl_build_dir}" -S "${absl_dir}"
cmake "${absl_dir}"
fi
cmake --build "${absl_build_dir}" --target install -- -j
cmake --build . --target install -- -j
popd
}
uninstall_absl() {
xargs rm < "${absl_build_dir}"/install_manifest.txt
rm -rf "${absl_build_dir}"
mkdir -p "${absl_build_dir}"
}
# Test build, install, and link against installed abseil
install_absl "${install_dir}"
cmake \
-H"${project_dir}" \
-B"${project_build_dir}" \
-DCMAKE_PREFIX_PATH="${install_dir}"
cmake --build "${project_build_dir}" --target simple
pushd "${project_build_dir}"
cmake "${project_dir}" -DCMAKE_PREFIX_PATH="${install_dir}"
cmake --build . --target simple
output="$(${project_build_dir}/simple "printme" 2>&1)"
if [[ "${output}" != *"Arg 1: printme"* ]]; then

@ -219,8 +219,12 @@ class flat_hash_map : public absl::container_internal::raw_hash_map<
// Erases the element at `position` of the `flat_hash_map`, returning
// `void`.
//
// NOTE: this return behavior is different than that of STL containers in
// general and `std::unordered_map` in particular.
// NOTE: returning `void` in this case is different than that of STL
// containers in general and `std::unordered_map` in particular (which
// return an iterator to the element following the erased element). If that
// iterator is needed, simply post increment the iterator:
//
// map.erase(it++);
//
// iterator erase(const_iterator first, const_iterator last):
//

@ -212,8 +212,12 @@ class flat_hash_set
// Erases the element at `position` of the `flat_hash_set`, returning
// `void`.
//
// NOTE: this return behavior is different than that of STL containers in
// general and `std::unordered_map` in particular.
// NOTE: returning `void` in this case is different than that of STL
// containers in general and `std::unordered_set` in particular (which
// return an iterator to the element following the erased element). If that
// iterator is needed, simply post increment the iterator:
//
// set.erase(it++);
//
// iterator erase(const_iterator first, const_iterator last):
//

@ -66,7 +66,10 @@ namespace absl {
// designed to cover the same API footprint as covered by `std::vector`.
template <typename T, size_t N, typename A = std::allocator<T>>
class InlinedVector {
using Storage = inlined_vector_internal::InlinedVectorStorage<T, N, A>;
static_assert(
N > 0, "InlinedVector cannot be instantiated with `0` inlined elements.");
using Storage = inlined_vector_internal::Storage<InlinedVector>;
using Tag = typename Storage::Tag;
using AllocatorAndTag = typename Storage::AllocatorAndTag;
using Allocation = typename Storage::Allocation;
@ -283,8 +286,7 @@ class InlinedVector {
// will no longer be inlined and `capacity()` will equal its capacity on the
// allocated heap.
size_type capacity() const noexcept {
return allocated() ? allocation().capacity()
: Storage::GetInlinedCapacity();
return allocated() ? allocation().capacity() : static_cast<size_type>(N);
}
// `InlinedVector::data()`
@ -802,19 +804,19 @@ class InlinedVector {
// `InlinedVector::shrink_to_fit()`
//
// Reduces memory usage by freeing unused memory. After this call, calls to
// `capacity()` will be equal to `max(Storage::GetInlinedCapacity(), size())`.
// `capacity()` will be equal to `max(N, size())`.
//
// If `size() <= Storage::GetInlinedCapacity()` and the elements are currently
// stored on the heap, they will be moved to the inlined storage and the heap
// memory will be deallocated.
// If `size() <= N` and the elements are currently stored on the heap, they
// will be moved to the inlined storage and the heap memory will be
// deallocated.
//
// If `size() > Storage::GetInlinedCapacity()` and `size() < capacity()` the
// elements will be moved to a smaller heap allocation.
// If `size() > N` and `size() < capacity()` the elements will be moved to a
// smaller heap allocation.
void shrink_to_fit() {
const auto s = size();
if (ABSL_PREDICT_FALSE(!allocated() || s == capacity())) return;
if (s <= Storage::GetInlinedCapacity()) {
if (s <= N) {
// Move the elements to the inlined storage.
// We have to do this using a temporary, because `inlined_storage` and
// `allocation_storage` are in a union field.
@ -943,7 +945,7 @@ class InlinedVector {
const size_type s = size();
assert(s <= capacity());
size_type target = (std::max)(Storage::GetInlinedCapacity(), s + delta);
size_type target = (std::max)(N, s + delta);
// Compute new capacity by repeatedly doubling current capacity
// TODO(psrc): Check and avoid overflow?
@ -1046,7 +1048,7 @@ class InlinedVector {
}
void InitAssign(size_type n) {
if (n > Storage::GetInlinedCapacity()) {
if (n > N) {
Allocation new_allocation(allocator(), n);
init_allocation(new_allocation);
UninitializedFill(allocated_space(), allocated_space() + n);
@ -1058,7 +1060,7 @@ class InlinedVector {
}
void InitAssign(size_type n, const_reference v) {
if (n > Storage::GetInlinedCapacity()) {
if (n > N) {
Allocation new_allocation(allocator(), n);
init_allocation(new_allocation);
UninitializedFill(allocated_space(), allocated_space() + n, v);

@ -23,17 +23,13 @@
namespace {
using IntVec = absl::InlinedVector<int, 8>;
void BM_InlinedVectorFill(benchmark::State& state) {
const int len = state.range(0);
absl::InlinedVector<int, 8> v;
int val = 10;
for (auto _ : state) {
IntVec v;
for (int i = 0; i < len; i++) {
v.push_back(i);
}
benchmark::DoNotOptimize(v);
v.push_back(val);
}
state.SetItemsProcessed(static_cast<int64_t>(state.iterations()) * len);
}
BENCHMARK(BM_InlinedVectorFill)->Range(0, 1024);
@ -43,23 +39,25 @@ void BM_InlinedVectorFillRange(benchmark::State& state) {
for (int i = 0; i < len; i++) {
ia[i] = i;
}
auto* from = ia.get();
auto* to = from + len;
for (auto _ : state) {
IntVec v(ia.get(), ia.get() + len);
benchmark::DoNotOptimize(from);
benchmark::DoNotOptimize(to);
absl::InlinedVector<int, 8> v(from, to);
benchmark::DoNotOptimize(v);
}
state.SetItemsProcessed(static_cast<int64_t>(state.iterations()) * len);
}
BENCHMARK(BM_InlinedVectorFillRange)->Range(0, 1024);
void BM_StdVectorFill(benchmark::State& state) {
const int len = state.range(0);
std::vector<int> v;
int val = 10;
for (auto _ : state) {
std::vector<int> v;
for (int i = 0; i < len; i++) {
v.push_back(i);
}
benchmark::DoNotOptimize(v);
benchmark::DoNotOptimize(val);
v.push_back(val);
}
state.SetItemsProcessed(static_cast<int64_t>(state.iterations()) * len);
}
BENCHMARK(BM_StdVectorFill)->Range(0, 1024);
@ -124,7 +122,7 @@ struct Buffer { // some arbitrary structure for benchmarking.
void* user_data;
};
void BM_InlinedVectorTenAssignments(benchmark::State& state) {
void BM_InlinedVectorAssignments(benchmark::State& state) {
const int len = state.range(0);
using BufferVec = absl::InlinedVector<Buffer, 2>;
@ -133,18 +131,25 @@ void BM_InlinedVectorTenAssignments(benchmark::State& state) {
BufferVec dst;
for (auto _ : state) {
for (int i = 0; i < 10; ++i) {
dst = src;
}
benchmark::DoNotOptimize(dst);
benchmark::DoNotOptimize(src);
dst = src;
}
}
BENCHMARK(BM_InlinedVectorTenAssignments)
->Arg(0)->Arg(1)->Arg(2)->Arg(3)->Arg(4)->Arg(20);
BENCHMARK(BM_InlinedVectorAssignments)
->Arg(0)
->Arg(1)
->Arg(2)
->Arg(3)
->Arg(4)
->Arg(20);
void BM_CreateFromContainer(benchmark::State& state) {
for (auto _ : state) {
absl::InlinedVector<int, 4> x(absl::InlinedVector<int, 4>{1, 2, 3});
benchmark::DoNotOptimize(x);
absl::InlinedVector<int, 4> src{1, 2, 3};
benchmark::DoNotOptimize(src);
absl::InlinedVector<int, 4> dst(std::move(src));
benchmark::DoNotOptimize(dst);
}
}
BENCHMARK(BM_CreateFromContainer);
@ -214,6 +219,8 @@ void BM_SwapElements(benchmark::State& state) {
Vec b;
for (auto _ : state) {
using std::swap;
benchmark::DoNotOptimize(a);
benchmark::DoNotOptimize(b);
swap(a, b);
}
}
@ -259,60 +266,44 @@ BENCHMARK_TEMPLATE(BM_Sizeof, absl::InlinedVector<std::string, 8>);
void BM_InlinedVectorIndexInlined(benchmark::State& state) {
absl::InlinedVector<int, 8> v = {1, 2, 3, 4, 5, 6, 7};
for (auto _ : state) {
for (int i = 0; i < 1000; ++i) {
benchmark::DoNotOptimize(v);
benchmark::DoNotOptimize(v[4]);
}
benchmark::DoNotOptimize(v);
benchmark::DoNotOptimize(v[4]);
}
state.SetItemsProcessed(1000 * static_cast<int64_t>(state.iterations()));
}
BENCHMARK(BM_InlinedVectorIndexInlined);
void BM_InlinedVectorIndexExternal(benchmark::State& state) {
absl::InlinedVector<int, 8> v = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10};
for (auto _ : state) {
for (int i = 0; i < 1000; ++i) {
benchmark::DoNotOptimize(v);
benchmark::DoNotOptimize(v[4]);
}
benchmark::DoNotOptimize(v);
benchmark::DoNotOptimize(v[4]);
}
state.SetItemsProcessed(1000 * static_cast<int64_t>(state.iterations()));
}
BENCHMARK(BM_InlinedVectorIndexExternal);
void BM_StdVectorIndex(benchmark::State& state) {
std::vector<int> v = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10};
for (auto _ : state) {
for (int i = 0; i < 1000; ++i) {
benchmark::DoNotOptimize(v);
benchmark::DoNotOptimize(v[4]);
}
benchmark::DoNotOptimize(v);
benchmark::DoNotOptimize(v[4]);
}
state.SetItemsProcessed(1000 * static_cast<int64_t>(state.iterations()));
}
BENCHMARK(BM_StdVectorIndex);
#define UNROLL_2(x) \
benchmark::DoNotOptimize(x); \
benchmark::DoNotOptimize(x);
#define UNROLL_4(x) UNROLL_2(x) UNROLL_2(x)
#define UNROLL_8(x) UNROLL_4(x) UNROLL_4(x)
#define UNROLL_16(x) UNROLL_8(x) UNROLL_8(x);
void BM_InlinedVectorDataInlined(benchmark::State& state) {
absl::InlinedVector<int, 8> v = {1, 2, 3, 4, 5, 6, 7};
for (auto _ : state) {
UNROLL_16(v.data());
benchmark::DoNotOptimize(v);
benchmark::DoNotOptimize(v.data());
}
state.SetItemsProcessed(16 * static_cast<int64_t>(state.iterations()));
}
BENCHMARK(BM_InlinedVectorDataInlined);
void BM_InlinedVectorDataExternal(benchmark::State& state) {
absl::InlinedVector<int, 8> v = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10};
for (auto _ : state) {
UNROLL_16(v.data());
benchmark::DoNotOptimize(v);
benchmark::DoNotOptimize(v.data());
}
state.SetItemsProcessed(16 * static_cast<int64_t>(state.iterations()));
}
@ -321,7 +312,8 @@ BENCHMARK(BM_InlinedVectorDataExternal);
void BM_StdVectorData(benchmark::State& state) {
std::vector<int> v = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10};
for (auto _ : state) {
UNROLL_16(v.data());
benchmark::DoNotOptimize(v);
benchmark::DoNotOptimize(v.data());
}
state.SetItemsProcessed(16 * static_cast<int64_t>(state.iterations()));
}
@ -330,54 +322,54 @@ BENCHMARK(BM_StdVectorData);
void BM_InlinedVectorSizeInlined(benchmark::State& state) {
absl::InlinedVector<int, 8> v = {1, 2, 3, 4, 5, 6, 7};
for (auto _ : state) {
UNROLL_16(v.size());
benchmark::DoNotOptimize(v);
benchmark::DoNotOptimize(v.size());
}
state.SetItemsProcessed(16 * static_cast<int64_t>(state.iterations()));
}
BENCHMARK(BM_InlinedVectorSizeInlined);
void BM_InlinedVectorSizeExternal(benchmark::State& state) {
absl::InlinedVector<int, 8> v = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10};
for (auto _ : state) {
UNROLL_16(v.size());
benchmark::DoNotOptimize(v);
benchmark::DoNotOptimize(v.size());
}
state.SetItemsProcessed(16 * static_cast<int64_t>(state.iterations()));
}
BENCHMARK(BM_InlinedVectorSizeExternal);
void BM_StdVectorSize(benchmark::State& state) {
std::vector<int> v = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10};
for (auto _ : state) {
UNROLL_16(v.size());
benchmark::DoNotOptimize(v);
benchmark::DoNotOptimize(v.size());
}
state.SetItemsProcessed(16 * static_cast<int64_t>(state.iterations()));
}
BENCHMARK(BM_StdVectorSize);
void BM_InlinedVectorEmptyInlined(benchmark::State& state) {
absl::InlinedVector<int, 8> v = {1, 2, 3, 4, 5, 6, 7};
for (auto _ : state) {
UNROLL_16(v.empty());
benchmark::DoNotOptimize(v);
benchmark::DoNotOptimize(v.empty());
}
state.SetItemsProcessed(16 * static_cast<int64_t>(state.iterations()));
}
BENCHMARK(BM_InlinedVectorEmptyInlined);
void BM_InlinedVectorEmptyExternal(benchmark::State& state) {
absl::InlinedVector<int, 8> v = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10};
for (auto _ : state) {
UNROLL_16(v.empty());
benchmark::DoNotOptimize(v);
benchmark::DoNotOptimize(v.empty());
}
state.SetItemsProcessed(16 * static_cast<int64_t>(state.iterations()));
}
BENCHMARK(BM_InlinedVectorEmptyExternal);
void BM_StdVectorEmpty(benchmark::State& state) {
std::vector<int> v = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10};
for (auto _ : state) {
UNROLL_16(v.empty());
benchmark::DoNotOptimize(v);
benchmark::DoNotOptimize(v.empty());
}
state.SetItemsProcessed(16 * static_cast<int64_t>(state.iterations()));
}
BENCHMARK(BM_StdVectorEmpty);

@ -24,11 +24,12 @@
namespace absl {
namespace inlined_vector_internal {
template <typename T, size_t N, typename A>
class InlinedVectorStorage {
static_assert(
N > 0, "InlinedVector cannot be instantiated with `0` inline elements.");
template <typename InlinedVector>
class Storage;
template <template <typename, size_t, typename> class InlinedVector, typename T,
size_t N, typename A>
class Storage<InlinedVector<T, N, A>> {
public:
using allocator_type = A;
using value_type = typename allocator_type::value_type;
@ -44,12 +45,7 @@ class InlinedVectorStorage {
using reverse_iterator = std::reverse_iterator<iterator>;
using const_reverse_iterator = std::reverse_iterator<const_iterator>;
constexpr static size_type GetInlinedCapacity() {
return static_cast<size_type>(N);
}
explicit InlinedVectorStorage(const allocator_type& a)
: allocator_and_tag_(a) {}
explicit Storage(const allocator_type& a) : allocator_and_tag_(a) {}
// TODO(johnsoncj): Make the below types and members private after migration

@ -845,6 +845,25 @@ TEST(Table, Erase) {
EXPECT_TRUE(t.find(0) == t.end());
}
TEST(Table, EraseMaintainsValidIterator) {
IntTable t;
const int kNumElements = 100;
for (int i = 0; i < kNumElements; i ++) {
EXPECT_TRUE(t.emplace(i).second);
}
EXPECT_EQ(t.size(), kNumElements);
int num_erase_calls = 0;
auto it = t.begin();
while (it != t.end()) {
t.erase(it++);
num_erase_calls++;
}
EXPECT_TRUE(t.empty());
EXPECT_EQ(num_erase_calls, kNumElements);
}
// Collect N bad keys by following algorithm:
// 1. Create an empty table and reserve it to 2 * N.
// 2. Insert N random elements.

@ -483,16 +483,17 @@ inline void AssertHashEnabled() {
} // namespace type_traits_internal
} // namespace absl
// An internal namespace that is required to implement the C++17 swap traits.
//
// NOTE: This is its own top-level namespace to avoid subtleties due to
// functions named "swap" that may appear in the absl namespace.
namespace absl_internal_swap {
// It is not further nested in type_traits_internal to avoid long symbol names.
namespace swap_internal {
// Necessary for the traits.
using std::swap;
// This declaration prevents global `swap` and `absl::swap` overloads from being
// considered unless ADL picks them up.
void swap();
template <class T>
using IsSwappableImpl = decltype(swap(std::declval<T&>(), std::declval<T&>()));
@ -520,22 +521,13 @@ struct IsNothrowSwappable
// Swap()
//
// Performs the swap idiom from a namespace with no additional `swap` overloads.
// Performs the swap idiom from a namespace where valid candidates may only be
// found in `std` or via ADL.
template <class T, absl::enable_if_t<IsSwappable<T>::value, int> = 0>
void Swap(T& lhs, T& rhs) noexcept(IsNothrowSwappable<T>::value) {
swap(lhs, rhs);
}
} // namespace absl_internal_swap
namespace absl {
namespace type_traits_internal {
// Make the swap-related traits/function accessible from this namespace.
using absl_internal_swap::IsNothrowSwappable;
using absl_internal_swap::IsSwappable;
using absl_internal_swap::Swap;
// StdSwapIsUnconstrained
//
// Some standard library implementations are broken in that they do not
@ -543,6 +535,16 @@ using absl_internal_swap::Swap;
// one of those implementations.
using StdSwapIsUnconstrained = IsSwappable<void()>;
} // namespace swap_internal
namespace type_traits_internal {
// Make the swap-related traits/function accessible from this namespace.
using swap_internal::IsNothrowSwappable;
using swap_internal::IsSwappable;
using swap_internal::Swap;
using swap_internal::StdSwapIsUnconstrained;
} // namespace type_traits_internal
} // namespace absl

@ -118,6 +118,10 @@ ABSL_CONST_INIT absl::base_internal::AtomicHook<
} // namespace
static inline bool EvalConditionAnnotated(const Condition *cond, Mutex *mu,
bool locking, bool trylock,
bool read_lock);
void RegisterMutexProfiler(void (*fn)(int64_t wait_timestamp)) {
submit_profile_data.Store(fn);
}
@ -233,15 +237,14 @@ enum { // Mutex and CondVar events passed as "ev" to PostSynchEvent
SYNCH_EV_SIGNALALL,
};
enum { // Event flags
SYNCH_F_R = 0x01, // reader event
SYNCH_F_LCK = 0x02, // PostSynchEvent called with mutex held
SYNCH_F_ACQ = 0x04, // event is an acquire
enum { // Event flags
SYNCH_F_R = 0x01, // reader event
SYNCH_F_LCK = 0x02, // PostSynchEvent called with mutex held
SYNCH_F_TRY = 0x04, // TryLock or ReaderTryLock
SYNCH_F_UNLOCK = 0x08, // Unlock or ReaderUnlock
SYNCH_F_LCK_W = SYNCH_F_LCK,
SYNCH_F_LCK_R = SYNCH_F_LCK | SYNCH_F_R,
SYNCH_F_ACQ_W = SYNCH_F_ACQ,
SYNCH_F_ACQ_R = SYNCH_F_ACQ | SYNCH_F_R,
};
} // anonymous namespace
@ -250,20 +253,20 @@ static const struct {
int flags;
const char *msg;
} event_properties[] = {
{ SYNCH_F_LCK_W|SYNCH_F_ACQ_W, "TryLock succeeded " },
{ 0, "TryLock failed " },
{ SYNCH_F_LCK_R|SYNCH_F_ACQ_R, "ReaderTryLock succeeded " },
{ 0, "ReaderTryLock failed " },
{ SYNCH_F_ACQ_W, "Lock blocking " },
{ SYNCH_F_LCK_W, "Lock returning " },
{ SYNCH_F_ACQ_R, "ReaderLock blocking " },
{ SYNCH_F_LCK_R, "ReaderLock returning " },
{ SYNCH_F_LCK_W, "Unlock " },
{ SYNCH_F_LCK_R, "ReaderUnlock " },
{ 0, "Wait on " },
{ 0, "Wait unblocked " },
{ 0, "Signal on " },
{ 0, "SignalAll on " },
{SYNCH_F_LCK_W | SYNCH_F_TRY, "TryLock succeeded "},
{0, "TryLock failed "},
{SYNCH_F_LCK_R | SYNCH_F_TRY, "ReaderTryLock succeeded "},
{0, "ReaderTryLock failed "},
{0, "Lock blocking "},
{SYNCH_F_LCK_W, "Lock returning "},
{0, "ReaderLock blocking "},
{SYNCH_F_LCK_R, "ReaderLock returning "},
{SYNCH_F_LCK_W | SYNCH_F_UNLOCK, "Unlock "},
{SYNCH_F_LCK_R | SYNCH_F_UNLOCK, "ReaderUnlock "},
{0, "Wait on "},
{0, "Wait unblocked "},
{0, "Signal on "},
{0, "SignalAll on "},
};
static absl::base_internal::SpinLock synch_event_mu(
@ -415,9 +418,26 @@ static void PostSynchEvent(void *obj, int ev) {
ABSL_RAW_LOG(INFO, "%s%p %s %s", event_properties[ev].msg, obj,
(e == nullptr ? "" : e->name), buffer);
}
if ((event_properties[ev].flags & SYNCH_F_LCK) != 0 && e != nullptr &&
e->invariant != nullptr) {
(*e->invariant)(e->arg);
const int flags = event_properties[ev].flags;
if ((flags & SYNCH_F_LCK) != 0 && e != nullptr && e->invariant != nullptr) {
// Calling the invariant as is causes problems under ThreadSanitizer.
// We are currently inside of Mutex Lock/Unlock and are ignoring all
// memory accesses and synchronization. If the invariant transitively
// synchronizes something else and we ignore the synchronization, we will
// get false positive race reports later.
// Reuse EvalConditionAnnotated to properly call into user code.
struct local {
static bool pred(SynchEvent *ev) {
(*ev->invariant)(ev->arg);
return false;
}
};
Condition cond(&local::pred, e);
Mutex *mu = static_cast<Mutex *>(obj);
const bool locking = (flags & SYNCH_F_UNLOCK) == 0;
const bool trylock = (flags & SYNCH_F_TRY) != 0;
const bool read_lock = (flags & SYNCH_F_R) != 0;
EvalConditionAnnotated(&cond, mu, locking, trylock, read_lock);
}
UnrefSynchEvent(e);
}
@ -1553,7 +1573,7 @@ bool Mutex::AwaitCommon(const Condition &cond, KernelTimeout t) {
ABSL_TSAN_MUTEX_PRE_LOCK(this, TsanFlags(how));
this->LockSlowLoop(&waitp, flags);
bool res = waitp.cond != nullptr || // => cond known true from LockSlowLoop
cond.Eval();
EvalConditionAnnotated(&cond, this, true, false, how == kShared);
ABSL_TSAN_MUTEX_POST_LOCK(this, TsanFlags(how), 0);
return res;
}
@ -1731,12 +1751,17 @@ void Mutex::LockSlow(MuHow how, const Condition *cond, int flags) {
// Compute cond->Eval() and tell race detectors that we do it under mutex mu.
static inline bool EvalConditionAnnotated(const Condition *cond, Mutex *mu,
bool locking, Mutex::MuHow how) {
bool locking, bool trylock,
bool read_lock) {
// Delicate annotation dance.
// We are currently inside of read/write lock/unlock operation.
// All memory accesses are ignored inside of mutex operations + for unlock
// operation tsan considers that we've already released the mutex.
bool res = false;
#ifdef THREAD_SANITIZER
const int flags = read_lock ? __tsan_mutex_read_lock : 0;
const int tryflags = flags | (trylock ? __tsan_mutex_try_lock : 0);
#endif
if (locking) {
// For lock we pretend that we have finished the operation,
// evaluate the predicate, then unlock the mutex and start locking it again
@ -1744,24 +1769,26 @@ static inline bool EvalConditionAnnotated(const Condition *cond, Mutex *mu,
// Note: we can't simply do POST_LOCK, Eval, PRE_LOCK, because then tsan
// will think the lock acquisition is recursive which will trigger
// deadlock detector.
ABSL_TSAN_MUTEX_POST_LOCK(mu, TsanFlags(how), 0);
ABSL_TSAN_MUTEX_POST_LOCK(mu, tryflags, 0);
res = cond->Eval();
ABSL_TSAN_MUTEX_PRE_UNLOCK(mu, TsanFlags(how));
ABSL_TSAN_MUTEX_POST_UNLOCK(mu, TsanFlags(how));
ABSL_TSAN_MUTEX_PRE_LOCK(mu, TsanFlags(how));
// There is no "try" version of Unlock, so use flags instead of tryflags.
ABSL_TSAN_MUTEX_PRE_UNLOCK(mu, flags);
ABSL_TSAN_MUTEX_POST_UNLOCK(mu, flags);
ABSL_TSAN_MUTEX_PRE_LOCK(mu, tryflags);
} else {
// Similarly, for unlock we pretend that we have unlocked the mutex,
// lock the mutex, evaluate the predicate, and start unlocking it again
// to match the annotation at the end of outer unlock operation.
ABSL_TSAN_MUTEX_POST_UNLOCK(mu, TsanFlags(how));
ABSL_TSAN_MUTEX_PRE_LOCK(mu, TsanFlags(how));
ABSL_TSAN_MUTEX_POST_LOCK(mu, TsanFlags(how), 0);
ABSL_TSAN_MUTEX_POST_UNLOCK(mu, flags);
ABSL_TSAN_MUTEX_PRE_LOCK(mu, flags);
ABSL_TSAN_MUTEX_POST_LOCK(mu, flags, 0);
res = cond->Eval();
ABSL_TSAN_MUTEX_PRE_UNLOCK(mu, TsanFlags(how));
ABSL_TSAN_MUTEX_PRE_UNLOCK(mu, flags);
}
// Prevent unused param warnings in non-TSAN builds.
static_cast<void>(mu);
static_cast<void>(how);
static_cast<void>(trylock);
static_cast<void>(read_lock);
return res;
}
@ -1807,7 +1834,8 @@ bool Mutex::LockSlowWithDeadline(MuHow how, const Condition *cond,
v, (how->fast_or | (v & zap_desig_waker[flags & kMuHasBlocked])) +
how->fast_add,
std::memory_order_acquire, std::memory_order_relaxed)) {
if (cond == nullptr || EvalConditionAnnotated(cond, this, true, how)) {
if (cond == nullptr ||
EvalConditionAnnotated(cond, this, true, false, how == kShared)) {
return true;
}
unlock = true;
@ -1825,7 +1853,8 @@ bool Mutex::LockSlowWithDeadline(MuHow how, const Condition *cond,
}
this->LockSlowLoop(&waitp, flags);
return waitp.cond != nullptr || // => cond known true from LockSlowLoop
cond == nullptr || EvalConditionAnnotated(cond, this, true, how);
cond == nullptr ||
EvalConditionAnnotated(cond, this, true, false, how == kShared);
}
// RAW_CHECK_FMT() takes a condition, a printf-style format string, and
@ -1881,7 +1910,8 @@ void Mutex::LockSlowLoop(SynchWaitParams *waitp, int flags) {
waitp->how->fast_add,
std::memory_order_acquire, std::memory_order_relaxed)) {
if (waitp->cond == nullptr ||
EvalConditionAnnotated(waitp->cond, this, true, waitp->how)) {
EvalConditionAnnotated(waitp->cond, this, true, false,
waitp->how == kShared)) {
break; // we timed out, or condition true, so return
}
this->UnlockSlow(waitp); // got lock but condition false
@ -1924,7 +1954,8 @@ void Mutex::LockSlowLoop(SynchWaitParams *waitp, int flags) {
std::memory_order_release,
std::memory_order_relaxed));
if (waitp->cond == nullptr ||
EvalConditionAnnotated(waitp->cond, this, true, waitp->how)) {
EvalConditionAnnotated(waitp->cond, this, true, false,
waitp->how == kShared)) {
break; // we timed out, or condition true, so return
}
this->UnlockSlow(waitp); // got lock but condition false

Loading…
Cancel
Save