Abseil Common Libraries (C++) (grcp 依赖) https://abseil.io/
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

1815 lines
55 KiB

7 years ago
// Copyright 2017 The Abseil Authors.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
#include "absl/container/inlined_vector.h"
#include <algorithm>
7 years ago
#include <forward_list>
#include <list>
#include <memory>
#include <scoped_allocator>
#include <sstream>
#include <stdexcept>
#include <string>
#include <vector>
#include "gmock/gmock.h"
#include "gtest/gtest.h"
#include "absl/base/attributes.h"
#include "absl/base/internal/exception_testing.h"
#include "absl/base/internal/raw_logging.h"
#include "absl/base/macros.h"
#include "absl/container/internal/test_instance_tracker.h"
Export of internal Abseil changes. -- 1c1d6e2404dfc6caa022b335df5acdac6da50fe1 by Derek Mauro <dmauro@google.com>: Fix the internal namespacing in unaligned_access.h PiperOrigin-RevId: 215434506 -- 17d4400aebf025a230690fc1c7a968ef8d85bbba by Eric Fiselier <ericwf@google.com>: gtest depends on the GCC extension allowing variadic macros to be passed a empty parameter pack for ..., but LLVM diagnoses this as a GNU extension. This patch suppresses the warning when building the absl tests. PiperOrigin-RevId: 215426161 -- f2c49dde23a9f445b9de963f1bbe840ebb568b30 by Eric Fiselier <ericwf@google.com>: Use EXPECT_DEATH_IF_SUPPORTED instead of EXPECT_DEATH. This avoids breaking the test when gtest doesn't support death tests. PiperOrigin-RevId: 215423849 -- cd687c1e121709603f4fc3726b534f6a9c52cc89 by Eric Fiselier <ericwf@google.com>: Disable LLVM's -Wmissing-variable-declarations in tests. GCC's configuration already disables this via -Wno-missing-declarations, this change makes LLVM do the same. The warning would otherwise flag most tests which use ABSL_FLAG. PiperOrigin-RevId: 215407429 -- d14098824c84e3a8c8f6fb920e0335fb48fe2010 by Eric Fiselier <ericwf@google.com>: Fix local variable shadowing in city hash implementation. PiperOrigin-RevId: 215407249 -- 4b5e140ba743f0d231790a26c49083abb4329e2c by Abseil Team <absl-team@google.com>: Make raw_hash_set::reserve 2X fast when reserve doesn't do any allocation. Make raw_hash_set::reserve ~1% faster when reserve does some (128~4k) allocation. PiperOrigin-RevId: 215348727 -- 461161e65e04b801480aa117af2534c594654ccf by Eric Fiselier <ericwf@google.com>: Internal change PiperOrigin-RevId: 215272283 -- 50413ae31ad3d3a177257416acd8ede47a17bff2 by Eric Fiselier <ericwf@google.com>: Internal Change PiperOrigin-RevId: 215233183 -- 477be54c43d61019a8fe4e190e340eb52737d383 by Abseil Team <absl-team@google.com>: Clarify misleading comment on ABSL_ATTRIBUTE_UNUSED PiperOrigin-RevId: 215185496 -- 2cafa2b5287507d3a946682aee9ab13af6d471c9 by Matt Kulukundis <kfm@google.com>: Add support for absl::Hash to various absl in types. PiperOrigin-RevId: 215039569 -- 082248901991aa3d29be0ea3689c7f213cf0fd83 by Derek Mauro <dmauro@google.com>: Remove an instance of HAS_GLOBAL_STRING from hash_function_defaults.h PiperOrigin-RevId: 214989094 -- b929f61907f0786a6133e3a9d7287e339c0a0acb by Derek Mauro <dmauro@google.com>: Internal import of Github #174 Fix code snippet in comment https://github.com/abseil/abseil-cpp/pull/174 PiperOrigin-RevId: 214958849 -- f2c5e829eca11c352e121f56eefbf87083305023 by Derek Mauro <dmauro@google.com>: Internal import of GitHub #173 Fix CMake build for absl::container. https://github.com/abseil/abseil-cpp/pull/173 PiperOrigin-RevId: 214957796 -- d704f860f9fddafb99e34e6c5032e49f73874e10 by Abseil Team <absl-team@google.com>: Internal change PiperOrigin-RevId: 214828181 GitOrigin-RevId: 1c1d6e2404dfc6caa022b335df5acdac6da50fe1 Change-Id: I551de2b1ba0cbaf6856cd5959358cf6651179dea
6 years ago
#include "absl/hash/hash_testing.h"
7 years ago
#include "absl/memory/memory.h"
#include "absl/strings/str_cat.h"
namespace {
using absl::test_internal::CopyableMovableInstance;
using absl::test_internal::CopyableOnlyInstance;
using absl::test_internal::InstanceTracker;
using testing::AllOf;
using testing::Each;
using testing::ElementsAre;
using testing::ElementsAreArray;
using testing::Eq;
using testing::Gt;
using testing::PrintToString;
using IntVec = absl::InlinedVector<int, 8>;
MATCHER_P(SizeIs, n, "") {
return testing::ExplainMatchResult(n, arg.size(), result_listener);
}
MATCHER_P(CapacityIs, n, "") {
return testing::ExplainMatchResult(n, arg.capacity(), result_listener);
}
MATCHER_P(ValueIs, e, "") {
return testing::ExplainMatchResult(e, arg.value(), result_listener);
}
// TODO(bsamwel): Add support for movable-only types.
// Test fixture for typed tests on BaseCountedInstance derived classes, see
// test_instance_tracker.h.
template <typename T>
class InstanceTest : public ::testing::Test {};
TYPED_TEST_CASE_P(InstanceTest);
// A simple reference counted class to make sure that the proper elements are
// destroyed in the erase(begin, end) test.
class RefCounted {
public:
RefCounted(int value, int* count) : value_(value), count_(count) {
Ref();
}
RefCounted(const RefCounted& v)
: value_(v.value_), count_(v.count_) {
Ref();
}
~RefCounted() {
Unref();
count_ = nullptr;
}
friend void swap(RefCounted& a, RefCounted& b) {
using std::swap;
swap(a.value_, b.value_);
swap(a.count_, b.count_);
}
RefCounted& operator=(RefCounted v) {
using std::swap;
swap(*this, v);
return *this;
}
void Ref() const {
ABSL_RAW_CHECK(count_ != nullptr, "");
++(*count_);
}
void Unref() const {
--(*count_);
ABSL_RAW_CHECK(*count_ >= 0, "");
}
int value_;
int* count_;
};
using RefCountedVec = absl::InlinedVector<RefCounted, 8>;
// A class with a vtable pointer
class Dynamic {
public:
virtual ~Dynamic() {}
};
using DynamicVec = absl::InlinedVector<Dynamic, 8>;
// Append 0..len-1 to *v
template <typename Container>
static void Fill(Container* v, int len, int offset = 0) {
for (int i = 0; i < len; i++) {
v->push_back(i + offset);
}
}
static IntVec Fill(int len, int offset = 0) {
IntVec v;
Fill(&v, len, offset);
return v;
}
// This is a stateful allocator, but the state lives outside of the
// allocator (in whatever test is using the allocator). This is odd
// but helps in tests where the allocator is propagated into nested
// containers - that chain of allocators uses the same state and is
// thus easier to query for aggregate allocation information.
template <typename T>
class CountingAllocator : public std::allocator<T> {
public:
using Alloc = std::allocator<T>;
using pointer = typename Alloc::pointer;
using size_type = typename Alloc::size_type;
CountingAllocator() : bytes_used_(nullptr) {}
explicit CountingAllocator(int64_t* b) : bytes_used_(b) {}
template <typename U>
CountingAllocator(const CountingAllocator<U>& x)
: Alloc(x), bytes_used_(x.bytes_used_) {}
pointer allocate(size_type n,
std::allocator<void>::const_pointer hint = nullptr) {
assert(bytes_used_ != nullptr);
*bytes_used_ += n * sizeof(T);
return Alloc::allocate(n, hint);
}
void deallocate(pointer p, size_type n) {
Alloc::deallocate(p, n);
assert(bytes_used_ != nullptr);
*bytes_used_ -= n * sizeof(T);
}
template<typename U>
class rebind {
public:
using other = CountingAllocator<U>;
};
friend bool operator==(const CountingAllocator& a,
const CountingAllocator& b) {
return a.bytes_used_ == b.bytes_used_;
}
friend bool operator!=(const CountingAllocator& a,
const CountingAllocator& b) {
return !(a == b);
}
int64_t* bytes_used_;
};
TEST(IntVec, SimpleOps) {
for (int len = 0; len < 20; len++) {
IntVec v;
const IntVec& cv = v; // const alias
Fill(&v, len);
EXPECT_EQ(len, v.size());
EXPECT_LE(len, v.capacity());
for (int i = 0; i < len; i++) {
EXPECT_EQ(i, v[i]);
EXPECT_EQ(i, v.at(i));
}
EXPECT_EQ(v.begin(), v.data());
EXPECT_EQ(cv.begin(), cv.data());
int counter = 0;
for (IntVec::iterator iter = v.begin(); iter != v.end(); ++iter) {
EXPECT_EQ(counter, *iter);
counter++;
}
EXPECT_EQ(counter, len);
counter = 0;
for (IntVec::const_iterator iter = v.begin(); iter != v.end(); ++iter) {
EXPECT_EQ(counter, *iter);
counter++;
}
EXPECT_EQ(counter, len);
counter = 0;
for (IntVec::const_iterator iter = v.cbegin(); iter != v.cend(); ++iter) {
EXPECT_EQ(counter, *iter);
counter++;
}
EXPECT_EQ(counter, len);
if (len > 0) {
EXPECT_EQ(0, v.front());
EXPECT_EQ(len - 1, v.back());
v.pop_back();
EXPECT_EQ(len - 1, v.size());
for (int i = 0; i < v.size(); ++i) {
EXPECT_EQ(i, v[i]);
EXPECT_EQ(i, v.at(i));
}
}
}
}
TEST(IntVec, AtThrows) {
IntVec v = {1, 2, 3};
EXPECT_EQ(v.at(2), 3);
ABSL_BASE_INTERNAL_EXPECT_FAIL(v.at(3), std::out_of_range,
"failed bounds check");
}
TEST(IntVec, ReverseIterator) {
for (int len = 0; len < 20; len++) {
IntVec v;
Fill(&v, len);
int counter = len;
for (IntVec::reverse_iterator iter = v.rbegin(); iter != v.rend(); ++iter) {
counter--;
EXPECT_EQ(counter, *iter);
}
EXPECT_EQ(counter, 0);
counter = len;
for (IntVec::const_reverse_iterator iter = v.rbegin(); iter != v.rend();
++iter) {
counter--;
EXPECT_EQ(counter, *iter);
}
EXPECT_EQ(counter, 0);
counter = len;
for (IntVec::const_reverse_iterator iter = v.crbegin(); iter != v.crend();
++iter) {
counter--;
EXPECT_EQ(counter, *iter);
}
EXPECT_EQ(counter, 0);
}
}
TEST(IntVec, Erase) {
for (int len = 1; len < 20; len++) {
for (int i = 0; i < len; ++i) {
IntVec v;
Fill(&v, len);
v.erase(v.begin() + i);
EXPECT_EQ(len - 1, v.size());
for (int j = 0; j < i; ++j) {
EXPECT_EQ(j, v[j]);
}
for (int j = i; j < len - 1; ++j) {
EXPECT_EQ(j + 1, v[j]);
}
}
}
}
// At the end of this test loop, the elements between [erase_begin, erase_end)
// should have reference counts == 0, and all others elements should have
// reference counts == 1.
TEST(RefCountedVec, EraseBeginEnd) {
for (int len = 1; len < 20; ++len) {
for (int erase_begin = 0; erase_begin < len; ++erase_begin) {
for (int erase_end = erase_begin; erase_end <= len; ++erase_end) {
std::vector<int> counts(len, 0);
RefCountedVec v;
for (int i = 0; i < len; ++i) {
v.push_back(RefCounted(i, &counts[i]));
}
int erase_len = erase_end - erase_begin;
v.erase(v.begin() + erase_begin, v.begin() + erase_end);
EXPECT_EQ(len - erase_len, v.size());
// Check the elements before the first element erased.
for (int i = 0; i < erase_begin; ++i) {
EXPECT_EQ(i, v[i].value_);
}
// Check the elements after the first element erased.
for (int i = erase_begin; i < v.size(); ++i) {
EXPECT_EQ(i + erase_len, v[i].value_);
}
// Check that the elements at the beginning are preserved.
for (int i = 0; i < erase_begin; ++i) {
EXPECT_EQ(1, counts[i]);
}
// Check that the erased elements are destroyed
for (int i = erase_begin; i < erase_end; ++i) {
EXPECT_EQ(0, counts[i]);
}
// Check that the elements at the end are preserved.
for (int i = erase_end; i< len; ++i) {
EXPECT_EQ(1, counts[i]);
}
}
}
}
}
struct NoDefaultCtor {
explicit NoDefaultCtor(int) {}
};
struct NoCopy {
NoCopy() {}
NoCopy(const NoCopy&) = delete;
};
struct NoAssign {
NoAssign() {}
NoAssign& operator=(const NoAssign&) = delete;
};
struct MoveOnly {
MoveOnly() {}
MoveOnly(MoveOnly&&) = default;
MoveOnly& operator=(MoveOnly&&) = default;
};
TEST(InlinedVectorTest, NoDefaultCtor) {
absl::InlinedVector<NoDefaultCtor, 1> v(10, NoDefaultCtor(2));
(void)v;
}
TEST(InlinedVectorTest, NoCopy) {
absl::InlinedVector<NoCopy, 1> v(10);
(void)v;
}
TEST(InlinedVectorTest, NoAssign) {
absl::InlinedVector<NoAssign, 1> v(10);
(void)v;
}
TEST(InlinedVectorTest, MoveOnly) {
absl::InlinedVector<MoveOnly, 2> v;
v.push_back(MoveOnly{});
v.push_back(MoveOnly{});
v.push_back(MoveOnly{});
v.erase(v.begin());
v.push_back(MoveOnly{});
v.erase(v.begin(), v.begin() + 1);
v.insert(v.begin(), MoveOnly{});
v.emplace(v.begin());
v.emplace(v.begin(), MoveOnly{});
}
TEST(InlinedVectorTest, Noexcept) {
EXPECT_TRUE(std::is_nothrow_move_constructible<IntVec>::value);
EXPECT_TRUE((std::is_nothrow_move_constructible<
absl::InlinedVector<MoveOnly, 2>>::value));
struct MoveCanThrow {
MoveCanThrow(MoveCanThrow&&) {}
};
EXPECT_EQ(absl::default_allocator_is_nothrow::value,
(std::is_nothrow_move_constructible<
absl::InlinedVector<MoveCanThrow, 2>>::value));
}
TEST(InlinedVectorTest, EmplaceBack) {
absl::InlinedVector<std::pair<std::string, int>, 1> v;
auto& inlined_element = v.emplace_back("answer", 42);
EXPECT_EQ(&inlined_element, &v[0]);
EXPECT_EQ(inlined_element.first, "answer");
EXPECT_EQ(inlined_element.second, 42);
auto& allocated_element = v.emplace_back("taxicab", 1729);
EXPECT_EQ(&allocated_element, &v[1]);
EXPECT_EQ(allocated_element.first, "taxicab");
EXPECT_EQ(allocated_element.second, 1729);
}
7 years ago
TEST(InlinedVectorTest, ShrinkToFitGrowingVector) {
absl::InlinedVector<std::pair<std::string, int>, 1> v;
v.shrink_to_fit();
EXPECT_EQ(v.capacity(), 1);
v.emplace_back("answer", 42);
v.shrink_to_fit();
EXPECT_EQ(v.capacity(), 1);
v.emplace_back("taxicab", 1729);
EXPECT_GE(v.capacity(), 2);
v.shrink_to_fit();
EXPECT_EQ(v.capacity(), 2);
v.reserve(100);
EXPECT_GE(v.capacity(), 100);
v.shrink_to_fit();
EXPECT_EQ(v.capacity(), 2);
}
TEST(InlinedVectorTest, ShrinkToFitEdgeCases) {
{
absl::InlinedVector<std::pair<std::string, int>, 1> v;
v.emplace_back("answer", 42);
v.emplace_back("taxicab", 1729);
EXPECT_GE(v.capacity(), 2);
v.pop_back();
v.shrink_to_fit();
EXPECT_EQ(v.capacity(), 1);
EXPECT_EQ(v[0].first, "answer");
EXPECT_EQ(v[0].second, 42);
}
{
absl::InlinedVector<std::string, 2> v(100);
v.resize(0);
v.shrink_to_fit();
EXPECT_EQ(v.capacity(), 2); // inlined capacity
}
{
absl::InlinedVector<std::string, 2> v(100);
v.resize(1);
v.shrink_to_fit();
EXPECT_EQ(v.capacity(), 2); // inlined capacity
}
{
absl::InlinedVector<std::string, 2> v(100);
v.resize(2);
v.shrink_to_fit();
EXPECT_EQ(v.capacity(), 2);
}
{
absl::InlinedVector<std::string, 2> v(100);
v.resize(3);
v.shrink_to_fit();
EXPECT_EQ(v.capacity(), 3);
}
}
7 years ago
TEST(IntVec, Insert) {
for (int len = 0; len < 20; len++) {
for (int pos = 0; pos <= len; pos++) {
{
// Single element
std::vector<int> std_v;
Fill(&std_v, len);
IntVec v;
Fill(&v, len);
std_v.insert(std_v.begin() + pos, 9999);
IntVec::iterator it = v.insert(v.cbegin() + pos, 9999);
EXPECT_THAT(v, ElementsAreArray(std_v));
EXPECT_EQ(it, v.cbegin() + pos);
}
{
// n elements
std::vector<int> std_v;
Fill(&std_v, len);
IntVec v;
Fill(&v, len);
IntVec::size_type n = 5;
std_v.insert(std_v.begin() + pos, n, 9999);
IntVec::iterator it = v.insert(v.cbegin() + pos, n, 9999);
EXPECT_THAT(v, ElementsAreArray(std_v));
EXPECT_EQ(it, v.cbegin() + pos);
}
{
// Iterator range (random access iterator)
std::vector<int> std_v;
Fill(&std_v, len);
IntVec v;
Fill(&v, len);
const std::vector<int> input = {9999, 8888, 7777};
std_v.insert(std_v.begin() + pos, input.cbegin(), input.cend());
IntVec::iterator it =
v.insert(v.cbegin() + pos, input.cbegin(), input.cend());
EXPECT_THAT(v, ElementsAreArray(std_v));
EXPECT_EQ(it, v.cbegin() + pos);
}
{
// Iterator range (forward iterator)
std::vector<int> std_v;
Fill(&std_v, len);
IntVec v;
Fill(&v, len);
const std::forward_list<int> input = {9999, 8888, 7777};
std_v.insert(std_v.begin() + pos, input.cbegin(), input.cend());
IntVec::iterator it =
v.insert(v.cbegin() + pos, input.cbegin(), input.cend());
EXPECT_THAT(v, ElementsAreArray(std_v));
EXPECT_EQ(it, v.cbegin() + pos);
}
{
// Iterator range (input iterator)
std::vector<int> std_v;
Fill(&std_v, len);
IntVec v;
Fill(&v, len);
std_v.insert(std_v.begin() + pos, {9999, 8888, 7777});
std::istringstream input("9999 8888 7777");
IntVec::iterator it =
v.insert(v.cbegin() + pos, std::istream_iterator<int>(input),
std::istream_iterator<int>());
EXPECT_THAT(v, ElementsAreArray(std_v));
EXPECT_EQ(it, v.cbegin() + pos);
}
{
// Initializer list
std::vector<int> std_v;
Fill(&std_v, len);
IntVec v;
Fill(&v, len);
std_v.insert(std_v.begin() + pos, {9999, 8888});
IntVec::iterator it = v.insert(v.cbegin() + pos, {9999, 8888});
EXPECT_THAT(v, ElementsAreArray(std_v));
EXPECT_EQ(it, v.cbegin() + pos);
}
}
}
}
TEST(RefCountedVec, InsertConstructorDestructor) {
// Make sure the proper construction/destruction happen during insert
// operations.
for (int len = 0; len < 20; len++) {
SCOPED_TRACE(len);
for (int pos = 0; pos <= len; pos++) {
SCOPED_TRACE(pos);
std::vector<int> counts(len, 0);
int inserted_count = 0;
RefCountedVec v;
for (int i = 0; i < len; ++i) {
SCOPED_TRACE(i);
v.push_back(RefCounted(i, &counts[i]));
}
EXPECT_THAT(counts, Each(Eq(1)));
RefCounted insert_element(9999, &inserted_count);
EXPECT_EQ(1, inserted_count);
v.insert(v.begin() + pos, insert_element);
EXPECT_EQ(2, inserted_count);
// Check that the elements at the end are preserved.
EXPECT_THAT(counts, Each(Eq(1)));
EXPECT_EQ(2, inserted_count);
}
}
}
TEST(IntVec, Resize) {
for (int len = 0; len < 20; len++) {
IntVec v;
Fill(&v, len);
// Try resizing up and down by k elements
static const int kResizeElem = 1000000;
for (int k = 0; k < 10; k++) {
// Enlarging resize
v.resize(len+k, kResizeElem);
EXPECT_EQ(len+k, v.size());
EXPECT_LE(len+k, v.capacity());
for (int i = 0; i < len+k; i++) {
if (i < len) {
EXPECT_EQ(i, v[i]);
} else {
EXPECT_EQ(kResizeElem, v[i]);
}
}
// Shrinking resize
v.resize(len, kResizeElem);
EXPECT_EQ(len, v.size());
EXPECT_LE(len, v.capacity());
for (int i = 0; i < len; i++) {
EXPECT_EQ(i, v[i]);
}
}
}
}
TEST(IntVec, InitWithLength) {
for (int len = 0; len < 20; len++) {
IntVec v(len, 7);
EXPECT_EQ(len, v.size());
EXPECT_LE(len, v.capacity());
for (int i = 0; i < len; i++) {
EXPECT_EQ(7, v[i]);
}
}
}
TEST(IntVec, CopyConstructorAndAssignment) {
for (int len = 0; len < 20; len++) {
IntVec v;
Fill(&v, len);
EXPECT_EQ(len, v.size());
EXPECT_LE(len, v.capacity());
IntVec v2(v);
EXPECT_TRUE(v == v2) << PrintToString(v) << PrintToString(v2);
for (int start_len = 0; start_len < 20; start_len++) {
IntVec v3;
Fill(&v3, start_len, 99); // Add dummy elements that should go away
v3 = v;
EXPECT_TRUE(v == v3) << PrintToString(v) << PrintToString(v3);
}
}
}
TEST(IntVec, AliasingCopyAssignment) {
for (int len = 0; len < 20; ++len) {
IntVec original;
Fill(&original, len);
IntVec dup = original;
dup = *&dup;
EXPECT_EQ(dup, original);
}
}
7 years ago
TEST(IntVec, MoveConstructorAndAssignment) {
for (int len = 0; len < 20; len++) {
IntVec v_in;
const int inlined_capacity = v_in.capacity();
Fill(&v_in, len);
EXPECT_EQ(len, v_in.size());
EXPECT_LE(len, v_in.capacity());
{
IntVec v_temp(v_in);
auto* old_data = v_temp.data();
IntVec v_out(std::move(v_temp));
EXPECT_TRUE(v_in == v_out) << PrintToString(v_in) << PrintToString(v_out);
if (v_in.size() > inlined_capacity) {
// Allocation is moved as a whole, data stays in place.
EXPECT_TRUE(v_out.data() == old_data);
} else {
EXPECT_FALSE(v_out.data() == old_data);
}
}
for (int start_len = 0; start_len < 20; start_len++) {
IntVec v_out;
Fill(&v_out, start_len, 99); // Add dummy elements that should go away
IntVec v_temp(v_in);
auto* old_data = v_temp.data();
v_out = std::move(v_temp);
EXPECT_TRUE(v_in == v_out) << PrintToString(v_in) << PrintToString(v_out);
if (v_in.size() > inlined_capacity) {
// Allocation is moved as a whole, data stays in place.
EXPECT_TRUE(v_out.data() == old_data);
} else {
EXPECT_FALSE(v_out.data() == old_data);
}
}
}
}
class NotTriviallyDestructible {
public:
NotTriviallyDestructible() : p_(new int(1)) {}
explicit NotTriviallyDestructible(int i) : p_(new int(i)) {}
NotTriviallyDestructible(const NotTriviallyDestructible& other)
: p_(new int(*other.p_)) {}
NotTriviallyDestructible& operator=(const NotTriviallyDestructible& other) {
p_ = absl::make_unique<int>(*other.p_);
return *this;
}
bool operator==(const NotTriviallyDestructible& other) const {
return *p_ == *other.p_;
}
private:
std::unique_ptr<int> p_;
};
TEST(AliasingTest, Emplace) {
for (int i = 2; i < 20; ++i) {
absl::InlinedVector<NotTriviallyDestructible, 10> vec;
for (int j = 0; j < i; ++j) {
vec.push_back(NotTriviallyDestructible(j));
}
vec.emplace(vec.begin(), vec[0]);
EXPECT_EQ(vec[0], vec[1]);
vec.emplace(vec.begin() + i / 2, vec[i / 2]);
EXPECT_EQ(vec[i / 2], vec[i / 2 + 1]);
vec.emplace(vec.end() - 1, vec.back());
EXPECT_EQ(vec[vec.size() - 2], vec.back());
}
}
TEST(AliasingTest, InsertWithCount) {
for (int i = 1; i < 20; ++i) {
absl::InlinedVector<NotTriviallyDestructible, 10> vec;
for (int j = 0; j < i; ++j) {
vec.push_back(NotTriviallyDestructible(j));
}
for (int n = 0; n < 5; ++n) {
// We use back where we can because it's guaranteed to become invalidated
vec.insert(vec.begin(), n, vec.back());
auto b = vec.begin();
EXPECT_TRUE(
std::all_of(b, b + n, [&vec](const NotTriviallyDestructible& x) {
return x == vec.back();
}));
auto m_idx = vec.size() / 2;
vec.insert(vec.begin() + m_idx, n, vec.back());
auto m = vec.begin() + m_idx;
EXPECT_TRUE(
std::all_of(m, m + n, [&vec](const NotTriviallyDestructible& x) {
return x == vec.back();
}));
// We want distinct values so the equality test is meaningful,
// vec[vec.size() - 1] is also almost always invalidated.
auto old_e = vec.size() - 1;
auto val = vec[old_e];
vec.insert(vec.end(), n, vec[old_e]);
auto e = vec.begin() + old_e;
EXPECT_TRUE(std::all_of(
e, e + n,
[&val](const NotTriviallyDestructible& x) { return x == val; }));
}
}
}
7 years ago
TEST(OverheadTest, Storage) {
// Check for size overhead.
// In particular, ensure that std::allocator doesn't cost anything to store.
// The union should be absorbing some of the allocation bookkeeping overhead
// in the larger vectors, leaving only the size_ field as overhead.
EXPECT_EQ(2 * sizeof(int*),
sizeof(absl::InlinedVector<int*, 1>) - 1 * sizeof(int*));
EXPECT_EQ(1 * sizeof(int*),
sizeof(absl::InlinedVector<int*, 2>) - 2 * sizeof(int*));
EXPECT_EQ(1 * sizeof(int*),
sizeof(absl::InlinedVector<int*, 3>) - 3 * sizeof(int*));
EXPECT_EQ(1 * sizeof(int*),
sizeof(absl::InlinedVector<int*, 4>) - 4 * sizeof(int*));
EXPECT_EQ(1 * sizeof(int*),
sizeof(absl::InlinedVector<int*, 5>) - 5 * sizeof(int*));
EXPECT_EQ(1 * sizeof(int*),
sizeof(absl::InlinedVector<int*, 6>) - 6 * sizeof(int*));
EXPECT_EQ(1 * sizeof(int*),
sizeof(absl::InlinedVector<int*, 7>) - 7 * sizeof(int*));
EXPECT_EQ(1 * sizeof(int*),
sizeof(absl::InlinedVector<int*, 8>) - 8 * sizeof(int*));
}
TEST(IntVec, Clear) {
for (int len = 0; len < 20; len++) {
SCOPED_TRACE(len);
IntVec v;
Fill(&v, len);
v.clear();
EXPECT_EQ(0, v.size());
EXPECT_EQ(v.begin(), v.end());
}
}
TEST(IntVec, Reserve) {
for (int len = 0; len < 20; len++) {
IntVec v;
Fill(&v, len);
for (int newlen = 0; newlen < 100; newlen++) {
const int* start_rep = v.data();
v.reserve(newlen);
const int* final_rep = v.data();
if (newlen <= len) {
EXPECT_EQ(start_rep, final_rep);
}
EXPECT_LE(newlen, v.capacity());
// Filling up to newlen should not change rep
while (v.size() < newlen) {
v.push_back(0);
}
EXPECT_EQ(final_rep, v.data());
}
}
}
TEST(StringVec, SelfRefPushBack) {
std::vector<std::string> std_v;
absl::InlinedVector<std::string, 4> v;
const std::string s = "A quite long std::string to ensure heap.";
std_v.push_back(s);
v.push_back(s);
for (int i = 0; i < 20; ++i) {
EXPECT_THAT(v, ElementsAreArray(std_v));
v.push_back(v.back());
std_v.push_back(std_v.back());
}
EXPECT_THAT(v, ElementsAreArray(std_v));
}
TEST(StringVec, SelfRefPushBackWithMove) {
std::vector<std::string> std_v;
absl::InlinedVector<std::string, 4> v;
const std::string s = "A quite long std::string to ensure heap.";
std_v.push_back(s);
v.push_back(s);
for (int i = 0; i < 20; ++i) {
EXPECT_EQ(v.back(), std_v.back());
v.push_back(std::move(v.back()));
std_v.push_back(std::move(std_v.back()));
}
EXPECT_EQ(v.back(), std_v.back());
}
TEST(StringVec, SelfMove) {
const std::string s = "A quite long std::string to ensure heap.";
for (int len = 0; len < 20; len++) {
SCOPED_TRACE(len);
absl::InlinedVector<std::string, 8> v;
for (int i = 0; i < len; ++i) {
SCOPED_TRACE(i);
v.push_back(s);
}
// Indirection necessary to avoid compiler warning.
v = std::move(*(&v));
// Ensure that the inlined vector is still in a valid state by copying it.
// We don't expect specific contents since a self-move results in an
// unspecified valid state.
std::vector<std::string> copy(v.begin(), v.end());
}
}
TEST(IntVec, Swap) {
for (int l1 = 0; l1 < 20; l1++) {
SCOPED_TRACE(l1);
for (int l2 = 0; l2 < 20; l2++) {
SCOPED_TRACE(l2);
IntVec a = Fill(l1, 0);
IntVec b = Fill(l2, 100);
{
using std::swap;
swap(a, b);
}
EXPECT_EQ(l1, b.size());
EXPECT_EQ(l2, a.size());
for (int i = 0; i < l1; i++) {
SCOPED_TRACE(i);
EXPECT_EQ(i, b[i]);
}
for (int i = 0; i < l2; i++) {
SCOPED_TRACE(i);
EXPECT_EQ(100 + i, a[i]);
}
}
}
}
TYPED_TEST_P(InstanceTest, Swap) {
using Instance = TypeParam;
using InstanceVec = absl::InlinedVector<Instance, 8>;
for (int l1 = 0; l1 < 20; l1++) {
SCOPED_TRACE(l1);
for (int l2 = 0; l2 < 20; l2++) {
SCOPED_TRACE(l2);
InstanceTracker tracker;
InstanceVec a, b;
const size_t inlined_capacity = a.capacity();
Export of internal Abseil changes. -- ba4dd47492748bd630462eb68b7959037fc6a11a by Abseil Team <absl-team@google.com>: Work around nvcc 9.0 compiler bug for open-source Tensorflow build. With the current implementation, when I (unintentionally and transitively) include absl/types/optional.h in a CUDA compilation unit, I get the following nvcc error message: INFO: From Compiling tensorflow/core/kernels/crop_and_resize_op_gpu.cu.cc: external/com_google_absl/absl/types/optional.h: In member function 'void absl::optional_internal::optional_data_dtor_base<T, <anonymous> >::destruct()': external/com_google_absl/absl/types/optional.h:185:50: error: '__T0' was not declared in this scope data_.~T(); I've also seen similar compilation failures online, for flat_hash_map: https://devtalk.nvidia.com/default/topic/1042599/nvcc-preprocessor-bug-causes-compilation-failure/ The bug is always around unnamed template parameters. Therefore, the workaround is to make them named. PiperOrigin-RevId: 219208288 -- dad2f40cb2e8d5017660985ef6fb57f3c3cdcc80 by CJ Johnson <johnsoncj@google.com>: Adds internal macros for catching and throwing unknown exception types PiperOrigin-RevId: 219207362 -- 0a9840328d2d86e8420b853435fdbf1f7a19d931 by Abseil Team <absl-team@google.com>: Fix typo in mutex.h comments. PiperOrigin-RevId: 219199397 -- 0d576dc7597564210bfdf91518075064756f0bf4 by Matt Calabrese <calabrese@google.com>: Internal change. PiperOrigin-RevId: 219185475 -- 66be156095571959fb19a76da8ad0b53ec37658e by Abseil Team <absl-team@google.com>: Fix alignment conformance for VS 2017 >= 15.8 (fix #193) PiperOrigin-RevId: 219129894 -- a6e1825a12587945f8194677ccfdcaba6f7aad1d by Abseil Team <absl-team@google.com>: Reapply PR #173 PiperOrigin-RevId: 219129361 -- cf72ade4881b25acc6ccaea468f69793a0fdce32 by Abseil Team <absl-team@google.com>: Update .gitignore PiperOrigin-RevId: 219127495 -- 0537490c6348a2cb489abe15638928ac5aa6982a by Jon Cohen <cohenjon@google.com>: Small refactor and reformat of error messages from the exception safety test framework. PiperOrigin-RevId: 218927773 -- 4c556ca45fa25698ad12002a00c713aeceefab73 by CJ Johnson <johnsoncj@google.com>: Updates the inlined vector swap tests to check for number of moves that took place if available PiperOrigin-RevId: 218900777 -- dcbfda0021a1e6dfa9586986b1269c06ec394053 by Mark Barolak <mbar@google.com>: Add parens around calls to std::numeric_limits<>::min and std::numeric_limits<>::max to prevent compilation errors on Windows platforms where min and max are defined as macros. PiperOrigin-RevId: 218888700 GitOrigin-RevId: ba4dd47492748bd630462eb68b7959037fc6a11a Change-Id: I0e393958eb8cb501b85f6114979f6d4d86ed996c
6 years ago
auto min_len = std::min(l1, l2);
auto max_len = std::max(l1, l2);
7 years ago
for (int i = 0; i < l1; i++) a.push_back(Instance(i));
for (int i = 0; i < l2; i++) b.push_back(Instance(100+i));
EXPECT_EQ(tracker.instances(), l1 + l2);
tracker.ResetCopiesMovesSwaps();
{
using std::swap;
swap(a, b);
}
EXPECT_EQ(tracker.instances(), l1 + l2);
if (a.size() > inlined_capacity && b.size() > inlined_capacity) {
EXPECT_EQ(tracker.swaps(), 0); // Allocations are swapped.
EXPECT_EQ(tracker.moves(), 0);
} else if (a.size() <= inlined_capacity && b.size() <= inlined_capacity) {
Export of internal Abseil changes. -- ba4dd47492748bd630462eb68b7959037fc6a11a by Abseil Team <absl-team@google.com>: Work around nvcc 9.0 compiler bug for open-source Tensorflow build. With the current implementation, when I (unintentionally and transitively) include absl/types/optional.h in a CUDA compilation unit, I get the following nvcc error message: INFO: From Compiling tensorflow/core/kernels/crop_and_resize_op_gpu.cu.cc: external/com_google_absl/absl/types/optional.h: In member function 'void absl::optional_internal::optional_data_dtor_base<T, <anonymous> >::destruct()': external/com_google_absl/absl/types/optional.h:185:50: error: '__T0' was not declared in this scope data_.~T(); I've also seen similar compilation failures online, for flat_hash_map: https://devtalk.nvidia.com/default/topic/1042599/nvcc-preprocessor-bug-causes-compilation-failure/ The bug is always around unnamed template parameters. Therefore, the workaround is to make them named. PiperOrigin-RevId: 219208288 -- dad2f40cb2e8d5017660985ef6fb57f3c3cdcc80 by CJ Johnson <johnsoncj@google.com>: Adds internal macros for catching and throwing unknown exception types PiperOrigin-RevId: 219207362 -- 0a9840328d2d86e8420b853435fdbf1f7a19d931 by Abseil Team <absl-team@google.com>: Fix typo in mutex.h comments. PiperOrigin-RevId: 219199397 -- 0d576dc7597564210bfdf91518075064756f0bf4 by Matt Calabrese <calabrese@google.com>: Internal change. PiperOrigin-RevId: 219185475 -- 66be156095571959fb19a76da8ad0b53ec37658e by Abseil Team <absl-team@google.com>: Fix alignment conformance for VS 2017 >= 15.8 (fix #193) PiperOrigin-RevId: 219129894 -- a6e1825a12587945f8194677ccfdcaba6f7aad1d by Abseil Team <absl-team@google.com>: Reapply PR #173 PiperOrigin-RevId: 219129361 -- cf72ade4881b25acc6ccaea468f69793a0fdce32 by Abseil Team <absl-team@google.com>: Update .gitignore PiperOrigin-RevId: 219127495 -- 0537490c6348a2cb489abe15638928ac5aa6982a by Jon Cohen <cohenjon@google.com>: Small refactor and reformat of error messages from the exception safety test framework. PiperOrigin-RevId: 218927773 -- 4c556ca45fa25698ad12002a00c713aeceefab73 by CJ Johnson <johnsoncj@google.com>: Updates the inlined vector swap tests to check for number of moves that took place if available PiperOrigin-RevId: 218900777 -- dcbfda0021a1e6dfa9586986b1269c06ec394053 by Mark Barolak <mbar@google.com>: Add parens around calls to std::numeric_limits<>::min and std::numeric_limits<>::max to prevent compilation errors on Windows platforms where min and max are defined as macros. PiperOrigin-RevId: 218888700 GitOrigin-RevId: ba4dd47492748bd630462eb68b7959037fc6a11a Change-Id: I0e393958eb8cb501b85f6114979f6d4d86ed996c
6 years ago
EXPECT_EQ(tracker.swaps(), min_len);
EXPECT_EQ((tracker.moves() ? tracker.moves() : tracker.copies()),
max_len - min_len);
7 years ago
} else {
// One is allocated and the other isn't. The allocation is transferred
// without copying elements, and the inlined instances are copied/moved.
EXPECT_EQ(tracker.swaps(), 0);
Export of internal Abseil changes. -- ba4dd47492748bd630462eb68b7959037fc6a11a by Abseil Team <absl-team@google.com>: Work around nvcc 9.0 compiler bug for open-source Tensorflow build. With the current implementation, when I (unintentionally and transitively) include absl/types/optional.h in a CUDA compilation unit, I get the following nvcc error message: INFO: From Compiling tensorflow/core/kernels/crop_and_resize_op_gpu.cu.cc: external/com_google_absl/absl/types/optional.h: In member function 'void absl::optional_internal::optional_data_dtor_base<T, <anonymous> >::destruct()': external/com_google_absl/absl/types/optional.h:185:50: error: '__T0' was not declared in this scope data_.~T(); I've also seen similar compilation failures online, for flat_hash_map: https://devtalk.nvidia.com/default/topic/1042599/nvcc-preprocessor-bug-causes-compilation-failure/ The bug is always around unnamed template parameters. Therefore, the workaround is to make them named. PiperOrigin-RevId: 219208288 -- dad2f40cb2e8d5017660985ef6fb57f3c3cdcc80 by CJ Johnson <johnsoncj@google.com>: Adds internal macros for catching and throwing unknown exception types PiperOrigin-RevId: 219207362 -- 0a9840328d2d86e8420b853435fdbf1f7a19d931 by Abseil Team <absl-team@google.com>: Fix typo in mutex.h comments. PiperOrigin-RevId: 219199397 -- 0d576dc7597564210bfdf91518075064756f0bf4 by Matt Calabrese <calabrese@google.com>: Internal change. PiperOrigin-RevId: 219185475 -- 66be156095571959fb19a76da8ad0b53ec37658e by Abseil Team <absl-team@google.com>: Fix alignment conformance for VS 2017 >= 15.8 (fix #193) PiperOrigin-RevId: 219129894 -- a6e1825a12587945f8194677ccfdcaba6f7aad1d by Abseil Team <absl-team@google.com>: Reapply PR #173 PiperOrigin-RevId: 219129361 -- cf72ade4881b25acc6ccaea468f69793a0fdce32 by Abseil Team <absl-team@google.com>: Update .gitignore PiperOrigin-RevId: 219127495 -- 0537490c6348a2cb489abe15638928ac5aa6982a by Jon Cohen <cohenjon@google.com>: Small refactor and reformat of error messages from the exception safety test framework. PiperOrigin-RevId: 218927773 -- 4c556ca45fa25698ad12002a00c713aeceefab73 by CJ Johnson <johnsoncj@google.com>: Updates the inlined vector swap tests to check for number of moves that took place if available PiperOrigin-RevId: 218900777 -- dcbfda0021a1e6dfa9586986b1269c06ec394053 by Mark Barolak <mbar@google.com>: Add parens around calls to std::numeric_limits<>::min and std::numeric_limits<>::max to prevent compilation errors on Windows platforms where min and max are defined as macros. PiperOrigin-RevId: 218888700 GitOrigin-RevId: ba4dd47492748bd630462eb68b7959037fc6a11a Change-Id: I0e393958eb8cb501b85f6114979f6d4d86ed996c
6 years ago
EXPECT_EQ((tracker.moves() ? tracker.moves() : tracker.copies()),
min_len);
7 years ago
}
EXPECT_EQ(l1, b.size());
EXPECT_EQ(l2, a.size());
for (int i = 0; i < l1; i++) {
EXPECT_EQ(i, b[i].value());
}
for (int i = 0; i < l2; i++) {
EXPECT_EQ(100 + i, a[i].value());
}
}
}
}
TEST(IntVec, EqualAndNotEqual) {
IntVec a, b;
EXPECT_TRUE(a == b);
EXPECT_FALSE(a != b);
a.push_back(3);
EXPECT_FALSE(a == b);
EXPECT_TRUE(a != b);
b.push_back(3);
EXPECT_TRUE(a == b);
EXPECT_FALSE(a != b);
b.push_back(7);
EXPECT_FALSE(a == b);
EXPECT_TRUE(a != b);
a.push_back(6);
EXPECT_FALSE(a == b);
EXPECT_TRUE(a != b);
a.clear();
b.clear();
for (int i = 0; i < 100; i++) {
a.push_back(i);
b.push_back(i);
EXPECT_TRUE(a == b);
EXPECT_FALSE(a != b);
b[i] = b[i] + 1;
EXPECT_FALSE(a == b);
EXPECT_TRUE(a != b);
b[i] = b[i] - 1; // Back to before
EXPECT_TRUE(a == b);
EXPECT_FALSE(a != b);
}
}
TEST(IntVec, RelationalOps) {
IntVec a, b;
EXPECT_FALSE(a < b);
EXPECT_FALSE(b < a);
EXPECT_FALSE(a > b);
EXPECT_FALSE(b > a);
EXPECT_TRUE(a <= b);
EXPECT_TRUE(b <= a);
EXPECT_TRUE(a >= b);
EXPECT_TRUE(b >= a);
b.push_back(3);
EXPECT_TRUE(a < b);
EXPECT_FALSE(b < a);
EXPECT_FALSE(a > b);
EXPECT_TRUE(b > a);
EXPECT_TRUE(a <= b);
EXPECT_FALSE(b <= a);
EXPECT_FALSE(a >= b);
EXPECT_TRUE(b >= a);
}
TYPED_TEST_P(InstanceTest, CountConstructorsDestructors) {
using Instance = TypeParam;
using InstanceVec = absl::InlinedVector<Instance, 8>;
InstanceTracker tracker;
for (int len = 0; len < 20; len++) {
SCOPED_TRACE(len);
tracker.ResetCopiesMovesSwaps();
InstanceVec v;
const size_t inlined_capacity = v.capacity();
for (int i = 0; i < len; i++) {
v.push_back(Instance(i));
}
EXPECT_EQ(tracker.instances(), len);
EXPECT_GE(tracker.copies() + tracker.moves(),
len); // More due to reallocation.
tracker.ResetCopiesMovesSwaps();
// Enlarging resize() must construct some objects
tracker.ResetCopiesMovesSwaps();
v.resize(len + 10, Instance(100));
EXPECT_EQ(tracker.instances(), len + 10);
if (len <= inlined_capacity && len + 10 > inlined_capacity) {
EXPECT_EQ(tracker.copies() + tracker.moves(), 10 + len);
} else {
// Only specify a minimum number of copies + moves. We don't want to
// depend on the reallocation policy here.
EXPECT_GE(tracker.copies() + tracker.moves(),
10); // More due to reallocation.
}
// Shrinking resize() must destroy some objects
tracker.ResetCopiesMovesSwaps();
v.resize(len, Instance(100));
EXPECT_EQ(tracker.instances(), len);
EXPECT_EQ(tracker.copies(), 0);
EXPECT_EQ(tracker.moves(), 0);
// reserve() must not increase the number of initialized objects
SCOPED_TRACE("reserve");
v.reserve(len+1000);
EXPECT_EQ(tracker.instances(), len);
EXPECT_EQ(tracker.copies() + tracker.moves(), len);
// pop_back() and erase() must destroy one object
if (len > 0) {
tracker.ResetCopiesMovesSwaps();
v.pop_back();
EXPECT_EQ(tracker.instances(), len - 1);
EXPECT_EQ(tracker.copies(), 0);
EXPECT_EQ(tracker.moves(), 0);
if (!v.empty()) {
tracker.ResetCopiesMovesSwaps();
v.erase(v.begin());
EXPECT_EQ(tracker.instances(), len - 2);
EXPECT_EQ(tracker.copies() + tracker.moves(), len - 2);
}
}
tracker.ResetCopiesMovesSwaps();
int instances_before_empty_erase = tracker.instances();
v.erase(v.begin(), v.begin());
EXPECT_EQ(tracker.instances(), instances_before_empty_erase);
EXPECT_EQ(tracker.copies() + tracker.moves(), 0);
}
}
TYPED_TEST_P(InstanceTest, CountConstructorsDestructorsOnCopyConstruction) {
using Instance = TypeParam;
using InstanceVec = absl::InlinedVector<Instance, 8>;
InstanceTracker tracker;
for (int len = 0; len < 20; len++) {
SCOPED_TRACE(len);
tracker.ResetCopiesMovesSwaps();
InstanceVec v;
for (int i = 0; i < len; i++) {
v.push_back(Instance(i));
}
EXPECT_EQ(tracker.instances(), len);
EXPECT_GE(tracker.copies() + tracker.moves(),
len); // More due to reallocation.
tracker.ResetCopiesMovesSwaps();
{ // Copy constructor should create 'len' more instances.
InstanceVec v_copy(v);
EXPECT_EQ(tracker.instances(), len + len);
EXPECT_EQ(tracker.copies(), len);
EXPECT_EQ(tracker.moves(), 0);
}
EXPECT_EQ(tracker.instances(), len);
}
}
TYPED_TEST_P(InstanceTest, CountConstructorsDestructorsOnMoveConstruction) {
using Instance = TypeParam;
using InstanceVec = absl::InlinedVector<Instance, 8>;
InstanceTracker tracker;
for (int len = 0; len < 20; len++) {
SCOPED_TRACE(len);
tracker.ResetCopiesMovesSwaps();
InstanceVec v;
const size_t inlined_capacity = v.capacity();
for (int i = 0; i < len; i++) {
v.push_back(Instance(i));
}
EXPECT_EQ(tracker.instances(), len);
EXPECT_GE(tracker.copies() + tracker.moves(),
len); // More due to reallocation.
tracker.ResetCopiesMovesSwaps();
{
InstanceVec v_copy(std::move(v));
if (len > inlined_capacity) {
// Allocation is moved as a whole.
EXPECT_EQ(tracker.instances(), len);
EXPECT_EQ(tracker.live_instances(), len);
// Tests an implementation detail, don't rely on this in your code.
EXPECT_EQ(v.size(), 0); // NOLINT misc-use-after-move
EXPECT_EQ(tracker.copies(), 0);
EXPECT_EQ(tracker.moves(), 0);
} else {
EXPECT_EQ(tracker.instances(), len + len);
if (Instance::supports_move()) {
EXPECT_EQ(tracker.live_instances(), len);
EXPECT_EQ(tracker.copies(), 0);
EXPECT_EQ(tracker.moves(), len);
} else {
EXPECT_EQ(tracker.live_instances(), len + len);
EXPECT_EQ(tracker.copies(), len);
EXPECT_EQ(tracker.moves(), 0);
}
}
EXPECT_EQ(tracker.swaps(), 0);
}
}
}
TYPED_TEST_P(InstanceTest, CountConstructorsDestructorsOnAssignment) {
using Instance = TypeParam;
using InstanceVec = absl::InlinedVector<Instance, 8>;
InstanceTracker tracker;
for (int len = 0; len < 20; len++) {
SCOPED_TRACE(len);
for (int longorshort = 0; longorshort <= 1; ++longorshort) {
SCOPED_TRACE(longorshort);
tracker.ResetCopiesMovesSwaps();
InstanceVec longer, shorter;
for (int i = 0; i < len; i++) {
longer.push_back(Instance(i));
shorter.push_back(Instance(i));
}
longer.push_back(Instance(len));
EXPECT_EQ(tracker.instances(), len + len + 1);
EXPECT_GE(tracker.copies() + tracker.moves(),
len + len + 1); // More due to reallocation.
tracker.ResetCopiesMovesSwaps();
if (longorshort) {
shorter = longer;
EXPECT_EQ(tracker.instances(), (len + 1) + (len + 1));
EXPECT_GE(tracker.copies() + tracker.moves(),
len + 1); // More due to reallocation.
} else {
longer = shorter;
EXPECT_EQ(tracker.instances(), len + len);
EXPECT_EQ(tracker.copies() + tracker.moves(), len);
}
}
}
}
TYPED_TEST_P(InstanceTest, CountConstructorsDestructorsOnMoveAssignment) {
using Instance = TypeParam;
using InstanceVec = absl::InlinedVector<Instance, 8>;
InstanceTracker tracker;
for (int len = 0; len < 20; len++) {
SCOPED_TRACE(len);
for (int longorshort = 0; longorshort <= 1; ++longorshort) {
SCOPED_TRACE(longorshort);
tracker.ResetCopiesMovesSwaps();
InstanceVec longer, shorter;
const int inlined_capacity = longer.capacity();
for (int i = 0; i < len; i++) {
longer.push_back(Instance(i));
shorter.push_back(Instance(i));
}
longer.push_back(Instance(len));
EXPECT_EQ(tracker.instances(), len + len + 1);
EXPECT_GE(tracker.copies() + tracker.moves(),
len + len + 1); // More due to reallocation.
tracker.ResetCopiesMovesSwaps();
int src_len;
if (longorshort) {
src_len = len + 1;
shorter = std::move(longer);
} else {
src_len = len;
longer = std::move(shorter);
}
if (src_len > inlined_capacity) {
// Allocation moved as a whole.
EXPECT_EQ(tracker.instances(), src_len);
EXPECT_EQ(tracker.live_instances(), src_len);
EXPECT_EQ(tracker.copies(), 0);
EXPECT_EQ(tracker.moves(), 0);
} else {
// Elements are all copied.
EXPECT_EQ(tracker.instances(), src_len + src_len);
if (Instance::supports_move()) {
EXPECT_EQ(tracker.copies(), 0);
EXPECT_EQ(tracker.moves(), src_len);
EXPECT_EQ(tracker.live_instances(), src_len);
} else {
EXPECT_EQ(tracker.copies(), src_len);
EXPECT_EQ(tracker.moves(), 0);
EXPECT_EQ(tracker.live_instances(), src_len + src_len);
}
}
EXPECT_EQ(tracker.swaps(), 0);
}
}
}
TEST(CountElemAssign, SimpleTypeWithInlineBacking) {
for (size_t original_size = 0; original_size <= 5; ++original_size) {
SCOPED_TRACE(original_size);
// Original contents are [12345, 12345, ...]
std::vector<int> original_contents(original_size, 12345);
absl::InlinedVector<int, 2> v(original_contents.begin(),
original_contents.end());
v.assign(2, 123);
EXPECT_THAT(v, AllOf(SizeIs(2), ElementsAre(123, 123)));
if (original_size <= 2) {
// If the original had inline backing, it should stay inline.
EXPECT_EQ(2, v.capacity());
}
}
}
TEST(CountElemAssign, SimpleTypeWithAllocation) {
for (size_t original_size = 0; original_size <= 5; ++original_size) {
SCOPED_TRACE(original_size);
// Original contents are [12345, 12345, ...]
std::vector<int> original_contents(original_size, 12345);
absl::InlinedVector<int, 2> v(original_contents.begin(),
original_contents.end());
v.assign(3, 123);
EXPECT_THAT(v, AllOf(SizeIs(3), ElementsAre(123, 123, 123)));
EXPECT_LE(v.size(), v.capacity());
}
}
TYPED_TEST_P(InstanceTest, CountElemAssignInlineBacking) {
using Instance = TypeParam;
for (size_t original_size = 0; original_size <= 5; ++original_size) {
SCOPED_TRACE(original_size);
// Original contents are [12345, 12345, ...]
std::vector<Instance> original_contents(original_size, Instance(12345));
absl::InlinedVector<Instance, 2> v(original_contents.begin(),
original_contents.end());
v.assign(2, Instance(123));
EXPECT_THAT(v, AllOf(SizeIs(2), ElementsAre(ValueIs(123), ValueIs(123))));
if (original_size <= 2) {
// If the original had inline backing, it should stay inline.
EXPECT_EQ(2, v.capacity());
}
}
}
template <typename Instance>
void InstanceCountElemAssignWithAllocationTest() {
for (size_t original_size = 0; original_size <= 5; ++original_size) {
SCOPED_TRACE(original_size);
// Original contents are [12345, 12345, ...]
std::vector<Instance> original_contents(original_size, Instance(12345));
absl::InlinedVector<Instance, 2> v(original_contents.begin(),
original_contents.end());
v.assign(3, Instance(123));
EXPECT_THAT(v,
AllOf(SizeIs(3),
ElementsAre(ValueIs(123), ValueIs(123), ValueIs(123))));
EXPECT_LE(v.size(), v.capacity());
}
}
TEST(CountElemAssign, WithAllocationCopyableInstance) {
InstanceCountElemAssignWithAllocationTest<CopyableOnlyInstance>();
}
TEST(CountElemAssign, WithAllocationCopyableMovableInstance) {
InstanceCountElemAssignWithAllocationTest<CopyableMovableInstance>();
}
TEST(RangedConstructor, SimpleType) {
std::vector<int> source_v = {4, 5, 6};
// First try to fit in inline backing
absl::InlinedVector<int, 4> v(source_v.begin(), source_v.end());
EXPECT_EQ(3, v.size());
EXPECT_EQ(4, v.capacity()); // Indication that we're still on inlined storage
EXPECT_EQ(4, v[0]);
EXPECT_EQ(5, v[1]);
EXPECT_EQ(6, v[2]);
// Now, force a re-allocate
absl::InlinedVector<int, 2> realloc_v(source_v.begin(), source_v.end());
EXPECT_EQ(3, realloc_v.size());
EXPECT_LT(2, realloc_v.capacity());
EXPECT_EQ(4, realloc_v[0]);
EXPECT_EQ(5, realloc_v[1]);
EXPECT_EQ(6, realloc_v[2]);
}
// Test for ranged constructors using Instance as the element type and
// SourceContainer as the source container type.
template <typename Instance, typename SourceContainer, int inlined_capacity>
void InstanceRangedConstructorTestForContainer() {
InstanceTracker tracker;
SourceContainer source_v = {Instance(0), Instance(1)};
tracker.ResetCopiesMovesSwaps();
absl::InlinedVector<Instance, inlined_capacity> v(source_v.begin(),
source_v.end());
EXPECT_EQ(2, v.size());
EXPECT_LT(1, v.capacity());
EXPECT_EQ(0, v[0].value());
EXPECT_EQ(1, v[1].value());
EXPECT_EQ(tracker.copies(), 2);
EXPECT_EQ(tracker.moves(), 0);
}
template <typename Instance, int inlined_capacity>
void InstanceRangedConstructorTestWithCapacity() {
// Test with const and non-const, random access and non-random-access sources.
// TODO(bsamwel): Test with an input iterator source.
{
SCOPED_TRACE("std::list");
InstanceRangedConstructorTestForContainer<Instance, std::list<Instance>,
inlined_capacity>();
{
SCOPED_TRACE("const std::list");
InstanceRangedConstructorTestForContainer<
Instance, const std::list<Instance>, inlined_capacity>();
}
{
SCOPED_TRACE("std::vector");
InstanceRangedConstructorTestForContainer<Instance, std::vector<Instance>,
inlined_capacity>();
}
{
SCOPED_TRACE("const std::vector");
InstanceRangedConstructorTestForContainer<
Instance, const std::vector<Instance>, inlined_capacity>();
}
}
}
TYPED_TEST_P(InstanceTest, RangedConstructor) {
using Instance = TypeParam;
SCOPED_TRACE("capacity=1");
InstanceRangedConstructorTestWithCapacity<Instance, 1>();
SCOPED_TRACE("capacity=2");
InstanceRangedConstructorTestWithCapacity<Instance, 2>();
}
TEST(RangedConstructor, ElementsAreConstructed) {
std::vector<std::string> source_v = {"cat", "dog"};
// Force expansion and re-allocation of v. Ensures that when the vector is
// expanded that new elements are constructed.
absl::InlinedVector<std::string, 1> v(source_v.begin(), source_v.end());
EXPECT_EQ("cat", v[0]);
EXPECT_EQ("dog", v[1]);
}
TEST(RangedAssign, SimpleType) {
// Test for all combinations of original sizes (empty and non-empty inline,
// and out of line) and target sizes.
for (size_t original_size = 0; original_size <= 5; ++original_size) {
SCOPED_TRACE(original_size);
// Original contents are [12345, 12345, ...]
std::vector<int> original_contents(original_size, 12345);
for (size_t target_size = 0; target_size <= 5; ++target_size) {
SCOPED_TRACE(target_size);
// New contents are [3, 4, ...]
std::vector<int> new_contents;
for (size_t i = 0; i < target_size; ++i) {
new_contents.push_back(i + 3);
}
absl::InlinedVector<int, 3> v(original_contents.begin(),
original_contents.end());
v.assign(new_contents.begin(), new_contents.end());
EXPECT_EQ(new_contents.size(), v.size());
EXPECT_LE(new_contents.size(), v.capacity());
if (target_size <= 3 && original_size <= 3) {
// Storage should stay inline when target size is small.
EXPECT_EQ(3, v.capacity());
}
EXPECT_THAT(v, ElementsAreArray(new_contents));
}
}
}
// Returns true if lhs and rhs have the same value.
template <typename Instance>
static bool InstanceValuesEqual(const Instance& lhs, const Instance& rhs) {
return lhs.value() == rhs.value();
}
// Test for ranged assign() using Instance as the element type and
// SourceContainer as the source container type.
template <typename Instance, typename SourceContainer>
void InstanceRangedAssignTestForContainer() {
// Test for all combinations of original sizes (empty and non-empty inline,
// and out of line) and target sizes.
for (size_t original_size = 0; original_size <= 5; ++original_size) {
SCOPED_TRACE(original_size);
// Original contents are [12345, 12345, ...]
std::vector<Instance> original_contents(original_size, Instance(12345));
for (size_t target_size = 0; target_size <= 5; ++target_size) {
SCOPED_TRACE(target_size);
// New contents are [3, 4, ...]
// Generate data using a non-const container, because SourceContainer
// itself may be const.
// TODO(bsamwel): Test with an input iterator.
std::vector<Instance> new_contents_in;
for (size_t i = 0; i < target_size; ++i) {
new_contents_in.push_back(Instance(i + 3));
}
SourceContainer new_contents(new_contents_in.begin(),
new_contents_in.end());
absl::InlinedVector<Instance, 3> v(original_contents.begin(),
original_contents.end());
v.assign(new_contents.begin(), new_contents.end());
EXPECT_EQ(new_contents.size(), v.size());
EXPECT_LE(new_contents.size(), v.capacity());
if (target_size <= 3 && original_size <= 3) {
// Storage should stay inline when target size is small.
EXPECT_EQ(3, v.capacity());
}
EXPECT_TRUE(std::equal(v.begin(), v.end(), new_contents.begin(),
InstanceValuesEqual<Instance>));
}
}
}
TYPED_TEST_P(InstanceTest, RangedAssign) {
using Instance = TypeParam;
// Test with const and non-const, random access and non-random-access sources.
// TODO(bsamwel): Test with an input iterator source.
SCOPED_TRACE("std::list");
InstanceRangedAssignTestForContainer<Instance, std::list<Instance>>();
SCOPED_TRACE("const std::list");
InstanceRangedAssignTestForContainer<Instance, const std::list<Instance>>();
SCOPED_TRACE("std::vector");
InstanceRangedAssignTestForContainer<Instance, std::vector<Instance>>();
SCOPED_TRACE("const std::vector");
InstanceRangedAssignTestForContainer<Instance, const std::vector<Instance>>();
}
TEST(InitializerListConstructor, SimpleTypeWithInlineBacking) {
EXPECT_THAT((absl::InlinedVector<int, 4>{4, 5, 6}),
AllOf(SizeIs(3), CapacityIs(4), ElementsAre(4, 5, 6)));
}
TEST(InitializerListConstructor, SimpleTypeWithReallocationRequired) {
EXPECT_THAT((absl::InlinedVector<int, 2>{4, 5, 6}),
AllOf(SizeIs(3), CapacityIs(Gt(2)), ElementsAre(4, 5, 6)));
}
TEST(InitializerListConstructor, DisparateTypesInList) {
EXPECT_THAT((absl::InlinedVector<int, 2>{-7, 8ULL}), ElementsAre(-7, 8));
EXPECT_THAT((absl::InlinedVector<std::string, 2>{"foo", std::string("bar")}),
ElementsAre("foo", "bar"));
}
TEST(InitializerListConstructor, ComplexTypeWithInlineBacking) {
EXPECT_THAT((absl::InlinedVector<CopyableMovableInstance, 1>{
CopyableMovableInstance(0)}),
AllOf(SizeIs(1), CapacityIs(1), ElementsAre(ValueIs(0))));
}
TEST(InitializerListConstructor, ComplexTypeWithReallocationRequired) {
EXPECT_THAT(
(absl::InlinedVector<CopyableMovableInstance, 1>{
CopyableMovableInstance(0), CopyableMovableInstance(1)}),
AllOf(SizeIs(2), CapacityIs(Gt(1)), ElementsAre(ValueIs(0), ValueIs(1))));
}
TEST(InitializerListAssign, SimpleTypeFitsInlineBacking) {
for (size_t original_size = 0; original_size <= 4; ++original_size) {
SCOPED_TRACE(original_size);
absl::InlinedVector<int, 2> v1(original_size, 12345);
const size_t original_capacity_v1 = v1.capacity();
v1.assign({3});
EXPECT_THAT(
v1, AllOf(SizeIs(1), CapacityIs(original_capacity_v1), ElementsAre(3)));
absl::InlinedVector<int, 2> v2(original_size, 12345);
const size_t original_capacity_v2 = v2.capacity();
v2 = {3};
EXPECT_THAT(
v2, AllOf(SizeIs(1), CapacityIs(original_capacity_v2), ElementsAre(3)));
}
}
TEST(InitializerListAssign, SimpleTypeDoesNotFitInlineBacking) {
for (size_t original_size = 0; original_size <= 4; ++original_size) {
SCOPED_TRACE(original_size);
absl::InlinedVector<int, 2> v1(original_size, 12345);
v1.assign({3, 4, 5});
EXPECT_THAT(v1, AllOf(SizeIs(3), ElementsAre(3, 4, 5)));
EXPECT_LE(3, v1.capacity());
absl::InlinedVector<int, 2> v2(original_size, 12345);
v2 = {3, 4, 5};
EXPECT_THAT(v2, AllOf(SizeIs(3), ElementsAre(3, 4, 5)));
EXPECT_LE(3, v2.capacity());
}
}
TEST(InitializerListAssign, DisparateTypesInList) {
absl::InlinedVector<int, 2> v_int1;
v_int1.assign({-7, 8ULL});
EXPECT_THAT(v_int1, ElementsAre(-7, 8));
absl::InlinedVector<int, 2> v_int2;
v_int2 = {-7, 8ULL};
EXPECT_THAT(v_int2, ElementsAre(-7, 8));
absl::InlinedVector<std::string, 2> v_string1;
v_string1.assign({"foo", std::string("bar")});
EXPECT_THAT(v_string1, ElementsAre("foo", "bar"));
absl::InlinedVector<std::string, 2> v_string2;
v_string2 = {"foo", std::string("bar")};
EXPECT_THAT(v_string2, ElementsAre("foo", "bar"));
}
TYPED_TEST_P(InstanceTest, InitializerListAssign) {
using Instance = TypeParam;
for (size_t original_size = 0; original_size <= 4; ++original_size) {
SCOPED_TRACE(original_size);
absl::InlinedVector<Instance, 2> v(original_size, Instance(12345));
const size_t original_capacity = v.capacity();
v.assign({Instance(3)});
EXPECT_THAT(v, AllOf(SizeIs(1), CapacityIs(original_capacity),
ElementsAre(ValueIs(3))));
}
for (size_t original_size = 0; original_size <= 4; ++original_size) {
SCOPED_TRACE(original_size);
absl::InlinedVector<Instance, 2> v(original_size, Instance(12345));
v.assign({Instance(3), Instance(4), Instance(5)});
EXPECT_THAT(v, AllOf(SizeIs(3),
ElementsAre(ValueIs(3), ValueIs(4), ValueIs(5))));
EXPECT_LE(3, v.capacity());
}
}
REGISTER_TYPED_TEST_CASE_P(InstanceTest, Swap, CountConstructorsDestructors,
CountConstructorsDestructorsOnCopyConstruction,
CountConstructorsDestructorsOnMoveConstruction,
CountConstructorsDestructorsOnAssignment,
CountConstructorsDestructorsOnMoveAssignment,
CountElemAssignInlineBacking, RangedConstructor,
RangedAssign, InitializerListAssign);
using InstanceTypes =
::testing::Types<CopyableOnlyInstance, CopyableMovableInstance>;
INSTANTIATE_TYPED_TEST_CASE_P(InstanceTestOnTypes, InstanceTest, InstanceTypes);
TEST(DynamicVec, DynamicVecCompiles) {
DynamicVec v;
(void)v;
}
TEST(AllocatorSupportTest, Constructors) {
using MyAlloc = CountingAllocator<int>;
using AllocVec = absl::InlinedVector<int, 4, MyAlloc>;
const int ia[] = { 0, 1, 2, 3, 4, 5, 6, 7 };
int64_t allocated = 0;
MyAlloc alloc(&allocated);
{ AllocVec ABSL_ATTRIBUTE_UNUSED v; }
{ AllocVec ABSL_ATTRIBUTE_UNUSED v(alloc); }
{ AllocVec ABSL_ATTRIBUTE_UNUSED v(ia, ia + ABSL_ARRAYSIZE(ia), alloc); }
{ AllocVec ABSL_ATTRIBUTE_UNUSED v({1, 2, 3}, alloc); }
AllocVec v2;
{ AllocVec ABSL_ATTRIBUTE_UNUSED v(v2, alloc); }
{ AllocVec ABSL_ATTRIBUTE_UNUSED v(std::move(v2), alloc); }
}
TEST(AllocatorSupportTest, CountAllocations) {
using MyAlloc = CountingAllocator<int>;
using AllocVec = absl::InlinedVector<int, 4, MyAlloc>;
const int ia[] = { 0, 1, 2, 3, 4, 5, 6, 7 };
int64_t allocated = 0;
MyAlloc alloc(&allocated);
{
AllocVec ABSL_ATTRIBUTE_UNUSED v(ia, ia + 4, alloc);
EXPECT_THAT(allocated, 0);
}
EXPECT_THAT(allocated, 0);
{
AllocVec ABSL_ATTRIBUTE_UNUSED v(ia, ia + ABSL_ARRAYSIZE(ia), alloc);
EXPECT_THAT(allocated, v.size() * sizeof(int));
}
EXPECT_THAT(allocated, 0);
{
AllocVec v(4, 1, alloc);
EXPECT_THAT(allocated, 0);
int64_t allocated2 = 0;
MyAlloc alloc2(&allocated2);
AllocVec v2(v, alloc2);
EXPECT_THAT(allocated2, 0);
int64_t allocated3 = 0;
MyAlloc alloc3(&allocated3);
AllocVec v3(std::move(v), alloc3);
EXPECT_THAT(allocated3, 0);
}
EXPECT_THAT(allocated, 0);
{
AllocVec v(8, 2, alloc);
EXPECT_THAT(allocated, v.size() * sizeof(int));
int64_t allocated2 = 0;
MyAlloc alloc2(&allocated2);
AllocVec v2(v, alloc2);
EXPECT_THAT(allocated2, v2.size() * sizeof(int));
int64_t allocated3 = 0;
MyAlloc alloc3(&allocated3);
AllocVec v3(std::move(v), alloc3);
EXPECT_THAT(allocated3, v3.size() * sizeof(int));
}
EXPECT_EQ(allocated, 0);
{
// Test shrink_to_fit deallocations.
AllocVec v(8, 2, alloc);
EXPECT_EQ(allocated, 8 * sizeof(int));
v.resize(5);
EXPECT_EQ(allocated, 8 * sizeof(int));
v.shrink_to_fit();
EXPECT_EQ(allocated, 5 * sizeof(int));
v.resize(4);
EXPECT_EQ(allocated, 5 * sizeof(int));
v.shrink_to_fit();
EXPECT_EQ(allocated, 0);
}
7 years ago
}
TEST(AllocatorSupportTest, SwapBothAllocated) {
using MyAlloc = CountingAllocator<int>;
using AllocVec = absl::InlinedVector<int, 4, MyAlloc>;
int64_t allocated1 = 0;
int64_t allocated2 = 0;
{
const int ia1[] = { 0, 1, 2, 3, 4, 5, 6, 7 };
const int ia2[] = { 0, 1, 2, 3, 4, 5, 6, 7, 8 };
MyAlloc a1(&allocated1);
MyAlloc a2(&allocated2);
AllocVec v1(ia1, ia1 + ABSL_ARRAYSIZE(ia1), a1);
AllocVec v2(ia2, ia2 + ABSL_ARRAYSIZE(ia2), a2);
EXPECT_LT(v1.capacity(), v2.capacity());
EXPECT_THAT(allocated1, v1.capacity() * sizeof(int));
EXPECT_THAT(allocated2, v2.capacity() * sizeof(int));
v1.swap(v2);
EXPECT_THAT(v1, ElementsAreArray(ia2));
EXPECT_THAT(v2, ElementsAreArray(ia1));
EXPECT_THAT(allocated1, v2.capacity() * sizeof(int));
EXPECT_THAT(allocated2, v1.capacity() * sizeof(int));
}
EXPECT_THAT(allocated1, 0);
EXPECT_THAT(allocated2, 0);
}
TEST(AllocatorSupportTest, SwapOneAllocated) {
using MyAlloc = CountingAllocator<int>;
using AllocVec = absl::InlinedVector<int, 4, MyAlloc>;
int64_t allocated1 = 0;
int64_t allocated2 = 0;
{
const int ia1[] = { 0, 1, 2, 3, 4, 5, 6, 7 };
const int ia2[] = { 0, 1, 2, 3 };
MyAlloc a1(&allocated1);
MyAlloc a2(&allocated2);
AllocVec v1(ia1, ia1 + ABSL_ARRAYSIZE(ia1), a1);
AllocVec v2(ia2, ia2 + ABSL_ARRAYSIZE(ia2), a2);
EXPECT_THAT(allocated1, v1.capacity() * sizeof(int));
EXPECT_THAT(allocated2, 0);
v1.swap(v2);
EXPECT_THAT(v1, ElementsAreArray(ia2));
EXPECT_THAT(v2, ElementsAreArray(ia1));
EXPECT_THAT(allocated1, v2.capacity() * sizeof(int));
EXPECT_THAT(allocated2, 0);
EXPECT_TRUE(v2.get_allocator() == a1);
EXPECT_TRUE(v1.get_allocator() == a2);
}
EXPECT_THAT(allocated1, 0);
EXPECT_THAT(allocated2, 0);
}
TEST(AllocatorSupportTest, ScopedAllocatorWorks) {
using StdVector = std::vector<int, CountingAllocator<int>>;
using MyAlloc =
std::scoped_allocator_adaptor<CountingAllocator<StdVector>>;
using AllocVec = absl::InlinedVector<StdVector, 4, MyAlloc>;
Export of internal Abseil changes. -- 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 by Jon Cohen <cohenjon@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 220523164 -- 59ef14fe7034a3148f1e9cef1f128b8ca264b444 by Jon Cohen <cohenjon@google.com>: Don't assume how much std::vector's constructors allocate in InlinedVector's test for scoped_allocator_adaptor support. PiperOrigin-RevId: 220464683 -- 6f8351be43a44a8f10bf20612b2cc744a4a911c7 by Jon Cohen <cohenjon@google.com>: Add VS Code and some Bazel output files to absl/.gitignore PiperOrigin-RevId: 220464362 -- 43fac22f8af6b6ed55309a784a9d25d837393d0e by Abseil Team <absl-team@google.com>: absl: fix SpinLock::EncodeWaitCycles If a thread has ever observed or set kSpinLockSleeper, it must never leave 0 in kWaitTimeMask because at this point it is expected to wake subsequent threads. Current calculations in EncodeWaitCycles can result in 0 in kWaitTimeMask and lead to missed wakeups. This is mostly theoretical today, because the futex call needs to finish within 128 cycles (futex can return immediately without waiting, but 128 cycles still look too low for this). But this can well fire in future if we bump granularity and/or threshold for recording contention. Use kSpinLockSleeper instead of 0. PiperOrigin-RevId: 220463123 -- def9b7e3d45c99d68cc52a4429256116d7f421f2 by Abseil Team <absl-team@google.com>: absl: optimize SpinLock::SlowLock Currently we record contention even after the first initial spin. This leads to several performance issues: 1. If we succeed in acquiring the lock after the initial spin, overall wait time can be within tens/hundreds of nanoseconds. Recording such low wait time looks completely unnecessary and excessive. From some point of view this is not even a wait, because we did not sleep. And, for example, Mutex does not record contention in this case. In majority of cases the lock should be acquired exactly during the initial spin, yet we still go through full overhead of submitting contention. 2. Whenever a thread submits contention it also calls FUTEX_WAKE (there is no way to understand if it's necessary or not when wait value is stored in the lock). So if there are just 2 threads and a brief contention, the second thread will still call FUTEX_WAKE which is completely unnecessary overhead. Don't record contention after the initial spin wait. FWIW this also removes 2 CycleClock::Now calls and EncodeWaitCycles from the common hot path. PiperOrigin-RevId: 220379972 -- 75b0c0cb214de904ea622f81ec3f4eabdc8695b0 by Derek Mauro <dmauro@google.com>: Supress MSVC warnings in raw_hash_set's use of TrailingZeros and LeadingZeros. https://github.com/abseil/abseil-cpp/issues/208 PiperOrigin-RevId: 220372204 GitOrigin-RevId: 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 Change-Id: I3a66af4e050810a3274e45d4e055b2aa19ffba1b
6 years ago
// MSVC 2017's std::vector allocates different amounts of memory in debug
// versus opt mode.
int64_t test_allocated = 0;
StdVector v(CountingAllocator<int>{&test_allocated});
// The amount of memory allocated by a default constructed vector<int>
auto default_std_vec_allocated = test_allocated;
v.push_back(1);
// The amound of memory allocated by a copy-constructed vector<int> with one
// element.
int64_t one_element_std_vec_copy_allocated = test_allocated;
7 years ago
int64_t allocated = 0;
AllocVec vec(MyAlloc{CountingAllocator<StdVector>{&allocated}});
EXPECT_EQ(allocated, 0);
// This default constructs a vector<int>, but the allocator should pass itself
Export of internal Abseil changes. -- 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 by Jon Cohen <cohenjon@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 220523164 -- 59ef14fe7034a3148f1e9cef1f128b8ca264b444 by Jon Cohen <cohenjon@google.com>: Don't assume how much std::vector's constructors allocate in InlinedVector's test for scoped_allocator_adaptor support. PiperOrigin-RevId: 220464683 -- 6f8351be43a44a8f10bf20612b2cc744a4a911c7 by Jon Cohen <cohenjon@google.com>: Add VS Code and some Bazel output files to absl/.gitignore PiperOrigin-RevId: 220464362 -- 43fac22f8af6b6ed55309a784a9d25d837393d0e by Abseil Team <absl-team@google.com>: absl: fix SpinLock::EncodeWaitCycles If a thread has ever observed or set kSpinLockSleeper, it must never leave 0 in kWaitTimeMask because at this point it is expected to wake subsequent threads. Current calculations in EncodeWaitCycles can result in 0 in kWaitTimeMask and lead to missed wakeups. This is mostly theoretical today, because the futex call needs to finish within 128 cycles (futex can return immediately without waiting, but 128 cycles still look too low for this). But this can well fire in future if we bump granularity and/or threshold for recording contention. Use kSpinLockSleeper instead of 0. PiperOrigin-RevId: 220463123 -- def9b7e3d45c99d68cc52a4429256116d7f421f2 by Abseil Team <absl-team@google.com>: absl: optimize SpinLock::SlowLock Currently we record contention even after the first initial spin. This leads to several performance issues: 1. If we succeed in acquiring the lock after the initial spin, overall wait time can be within tens/hundreds of nanoseconds. Recording such low wait time looks completely unnecessary and excessive. From some point of view this is not even a wait, because we did not sleep. And, for example, Mutex does not record contention in this case. In majority of cases the lock should be acquired exactly during the initial spin, yet we still go through full overhead of submitting contention. 2. Whenever a thread submits contention it also calls FUTEX_WAKE (there is no way to understand if it's necessary or not when wait value is stored in the lock). So if there are just 2 threads and a brief contention, the second thread will still call FUTEX_WAKE which is completely unnecessary overhead. Don't record contention after the initial spin wait. FWIW this also removes 2 CycleClock::Now calls and EncodeWaitCycles from the common hot path. PiperOrigin-RevId: 220379972 -- 75b0c0cb214de904ea622f81ec3f4eabdc8695b0 by Derek Mauro <dmauro@google.com>: Supress MSVC warnings in raw_hash_set's use of TrailingZeros and LeadingZeros. https://github.com/abseil/abseil-cpp/issues/208 PiperOrigin-RevId: 220372204 GitOrigin-RevId: 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 Change-Id: I3a66af4e050810a3274e45d4e055b2aa19ffba1b
6 years ago
// into the vector<int>, so check allocation compared to that.
7 years ago
// The absl::InlinedVector does not allocate any memory.
Export of internal Abseil changes. -- 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 by Jon Cohen <cohenjon@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 220523164 -- 59ef14fe7034a3148f1e9cef1f128b8ca264b444 by Jon Cohen <cohenjon@google.com>: Don't assume how much std::vector's constructors allocate in InlinedVector's test for scoped_allocator_adaptor support. PiperOrigin-RevId: 220464683 -- 6f8351be43a44a8f10bf20612b2cc744a4a911c7 by Jon Cohen <cohenjon@google.com>: Add VS Code and some Bazel output files to absl/.gitignore PiperOrigin-RevId: 220464362 -- 43fac22f8af6b6ed55309a784a9d25d837393d0e by Abseil Team <absl-team@google.com>: absl: fix SpinLock::EncodeWaitCycles If a thread has ever observed or set kSpinLockSleeper, it must never leave 0 in kWaitTimeMask because at this point it is expected to wake subsequent threads. Current calculations in EncodeWaitCycles can result in 0 in kWaitTimeMask and lead to missed wakeups. This is mostly theoretical today, because the futex call needs to finish within 128 cycles (futex can return immediately without waiting, but 128 cycles still look too low for this). But this can well fire in future if we bump granularity and/or threshold for recording contention. Use kSpinLockSleeper instead of 0. PiperOrigin-RevId: 220463123 -- def9b7e3d45c99d68cc52a4429256116d7f421f2 by Abseil Team <absl-team@google.com>: absl: optimize SpinLock::SlowLock Currently we record contention even after the first initial spin. This leads to several performance issues: 1. If we succeed in acquiring the lock after the initial spin, overall wait time can be within tens/hundreds of nanoseconds. Recording such low wait time looks completely unnecessary and excessive. From some point of view this is not even a wait, because we did not sleep. And, for example, Mutex does not record contention in this case. In majority of cases the lock should be acquired exactly during the initial spin, yet we still go through full overhead of submitting contention. 2. Whenever a thread submits contention it also calls FUTEX_WAKE (there is no way to understand if it's necessary or not when wait value is stored in the lock). So if there are just 2 threads and a brief contention, the second thread will still call FUTEX_WAKE which is completely unnecessary overhead. Don't record contention after the initial spin wait. FWIW this also removes 2 CycleClock::Now calls and EncodeWaitCycles from the common hot path. PiperOrigin-RevId: 220379972 -- 75b0c0cb214de904ea622f81ec3f4eabdc8695b0 by Derek Mauro <dmauro@google.com>: Supress MSVC warnings in raw_hash_set's use of TrailingZeros and LeadingZeros. https://github.com/abseil/abseil-cpp/issues/208 PiperOrigin-RevId: 220372204 GitOrigin-RevId: 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 Change-Id: I3a66af4e050810a3274e45d4e055b2aa19ffba1b
6 years ago
// The vector<int> may allocate any memory.
auto expected = default_std_vec_allocated;
7 years ago
vec.resize(1);
Export of internal Abseil changes. -- 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 by Jon Cohen <cohenjon@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 220523164 -- 59ef14fe7034a3148f1e9cef1f128b8ca264b444 by Jon Cohen <cohenjon@google.com>: Don't assume how much std::vector's constructors allocate in InlinedVector's test for scoped_allocator_adaptor support. PiperOrigin-RevId: 220464683 -- 6f8351be43a44a8f10bf20612b2cc744a4a911c7 by Jon Cohen <cohenjon@google.com>: Add VS Code and some Bazel output files to absl/.gitignore PiperOrigin-RevId: 220464362 -- 43fac22f8af6b6ed55309a784a9d25d837393d0e by Abseil Team <absl-team@google.com>: absl: fix SpinLock::EncodeWaitCycles If a thread has ever observed or set kSpinLockSleeper, it must never leave 0 in kWaitTimeMask because at this point it is expected to wake subsequent threads. Current calculations in EncodeWaitCycles can result in 0 in kWaitTimeMask and lead to missed wakeups. This is mostly theoretical today, because the futex call needs to finish within 128 cycles (futex can return immediately without waiting, but 128 cycles still look too low for this). But this can well fire in future if we bump granularity and/or threshold for recording contention. Use kSpinLockSleeper instead of 0. PiperOrigin-RevId: 220463123 -- def9b7e3d45c99d68cc52a4429256116d7f421f2 by Abseil Team <absl-team@google.com>: absl: optimize SpinLock::SlowLock Currently we record contention even after the first initial spin. This leads to several performance issues: 1. If we succeed in acquiring the lock after the initial spin, overall wait time can be within tens/hundreds of nanoseconds. Recording such low wait time looks completely unnecessary and excessive. From some point of view this is not even a wait, because we did not sleep. And, for example, Mutex does not record contention in this case. In majority of cases the lock should be acquired exactly during the initial spin, yet we still go through full overhead of submitting contention. 2. Whenever a thread submits contention it also calls FUTEX_WAKE (there is no way to understand if it's necessary or not when wait value is stored in the lock). So if there are just 2 threads and a brief contention, the second thread will still call FUTEX_WAKE which is completely unnecessary overhead. Don't record contention after the initial spin wait. FWIW this also removes 2 CycleClock::Now calls and EncodeWaitCycles from the common hot path. PiperOrigin-RevId: 220379972 -- 75b0c0cb214de904ea622f81ec3f4eabdc8695b0 by Derek Mauro <dmauro@google.com>: Supress MSVC warnings in raw_hash_set's use of TrailingZeros and LeadingZeros. https://github.com/abseil/abseil-cpp/issues/208 PiperOrigin-RevId: 220372204 GitOrigin-RevId: 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 Change-Id: I3a66af4e050810a3274e45d4e055b2aa19ffba1b
6 years ago
EXPECT_EQ(allocated, expected);
7 years ago
// We make vector<int> allocate memory.
// It must go through the allocator even though we didn't construct the
Export of internal Abseil changes. -- 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 by Jon Cohen <cohenjon@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 220523164 -- 59ef14fe7034a3148f1e9cef1f128b8ca264b444 by Jon Cohen <cohenjon@google.com>: Don't assume how much std::vector's constructors allocate in InlinedVector's test for scoped_allocator_adaptor support. PiperOrigin-RevId: 220464683 -- 6f8351be43a44a8f10bf20612b2cc744a4a911c7 by Jon Cohen <cohenjon@google.com>: Add VS Code and some Bazel output files to absl/.gitignore PiperOrigin-RevId: 220464362 -- 43fac22f8af6b6ed55309a784a9d25d837393d0e by Abseil Team <absl-team@google.com>: absl: fix SpinLock::EncodeWaitCycles If a thread has ever observed or set kSpinLockSleeper, it must never leave 0 in kWaitTimeMask because at this point it is expected to wake subsequent threads. Current calculations in EncodeWaitCycles can result in 0 in kWaitTimeMask and lead to missed wakeups. This is mostly theoretical today, because the futex call needs to finish within 128 cycles (futex can return immediately without waiting, but 128 cycles still look too low for this). But this can well fire in future if we bump granularity and/or threshold for recording contention. Use kSpinLockSleeper instead of 0. PiperOrigin-RevId: 220463123 -- def9b7e3d45c99d68cc52a4429256116d7f421f2 by Abseil Team <absl-team@google.com>: absl: optimize SpinLock::SlowLock Currently we record contention even after the first initial spin. This leads to several performance issues: 1. If we succeed in acquiring the lock after the initial spin, overall wait time can be within tens/hundreds of nanoseconds. Recording such low wait time looks completely unnecessary and excessive. From some point of view this is not even a wait, because we did not sleep. And, for example, Mutex does not record contention in this case. In majority of cases the lock should be acquired exactly during the initial spin, yet we still go through full overhead of submitting contention. 2. Whenever a thread submits contention it also calls FUTEX_WAKE (there is no way to understand if it's necessary or not when wait value is stored in the lock). So if there are just 2 threads and a brief contention, the second thread will still call FUTEX_WAKE which is completely unnecessary overhead. Don't record contention after the initial spin wait. FWIW this also removes 2 CycleClock::Now calls and EncodeWaitCycles from the common hot path. PiperOrigin-RevId: 220379972 -- 75b0c0cb214de904ea622f81ec3f4eabdc8695b0 by Derek Mauro <dmauro@google.com>: Supress MSVC warnings in raw_hash_set's use of TrailingZeros and LeadingZeros. https://github.com/abseil/abseil-cpp/issues/208 PiperOrigin-RevId: 220372204 GitOrigin-RevId: 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 Change-Id: I3a66af4e050810a3274e45d4e055b2aa19ffba1b
6 years ago
// vector directly. This assumes that vec[0] doesn't need to grow its
// allocation.
expected += sizeof(int);
7 years ago
vec[0].push_back(1);
Export of internal Abseil changes. -- 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 by Jon Cohen <cohenjon@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 220523164 -- 59ef14fe7034a3148f1e9cef1f128b8ca264b444 by Jon Cohen <cohenjon@google.com>: Don't assume how much std::vector's constructors allocate in InlinedVector's test for scoped_allocator_adaptor support. PiperOrigin-RevId: 220464683 -- 6f8351be43a44a8f10bf20612b2cc744a4a911c7 by Jon Cohen <cohenjon@google.com>: Add VS Code and some Bazel output files to absl/.gitignore PiperOrigin-RevId: 220464362 -- 43fac22f8af6b6ed55309a784a9d25d837393d0e by Abseil Team <absl-team@google.com>: absl: fix SpinLock::EncodeWaitCycles If a thread has ever observed or set kSpinLockSleeper, it must never leave 0 in kWaitTimeMask because at this point it is expected to wake subsequent threads. Current calculations in EncodeWaitCycles can result in 0 in kWaitTimeMask and lead to missed wakeups. This is mostly theoretical today, because the futex call needs to finish within 128 cycles (futex can return immediately without waiting, but 128 cycles still look too low for this). But this can well fire in future if we bump granularity and/or threshold for recording contention. Use kSpinLockSleeper instead of 0. PiperOrigin-RevId: 220463123 -- def9b7e3d45c99d68cc52a4429256116d7f421f2 by Abseil Team <absl-team@google.com>: absl: optimize SpinLock::SlowLock Currently we record contention even after the first initial spin. This leads to several performance issues: 1. If we succeed in acquiring the lock after the initial spin, overall wait time can be within tens/hundreds of nanoseconds. Recording such low wait time looks completely unnecessary and excessive. From some point of view this is not even a wait, because we did not sleep. And, for example, Mutex does not record contention in this case. In majority of cases the lock should be acquired exactly during the initial spin, yet we still go through full overhead of submitting contention. 2. Whenever a thread submits contention it also calls FUTEX_WAKE (there is no way to understand if it's necessary or not when wait value is stored in the lock). So if there are just 2 threads and a brief contention, the second thread will still call FUTEX_WAKE which is completely unnecessary overhead. Don't record contention after the initial spin wait. FWIW this also removes 2 CycleClock::Now calls and EncodeWaitCycles from the common hot path. PiperOrigin-RevId: 220379972 -- 75b0c0cb214de904ea622f81ec3f4eabdc8695b0 by Derek Mauro <dmauro@google.com>: Supress MSVC warnings in raw_hash_set's use of TrailingZeros and LeadingZeros. https://github.com/abseil/abseil-cpp/issues/208 PiperOrigin-RevId: 220372204 GitOrigin-RevId: 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 Change-Id: I3a66af4e050810a3274e45d4e055b2aa19ffba1b
6 years ago
EXPECT_EQ(allocated, expected);
7 years ago
// Another allocating vector.
Export of internal Abseil changes. -- 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 by Jon Cohen <cohenjon@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 220523164 -- 59ef14fe7034a3148f1e9cef1f128b8ca264b444 by Jon Cohen <cohenjon@google.com>: Don't assume how much std::vector's constructors allocate in InlinedVector's test for scoped_allocator_adaptor support. PiperOrigin-RevId: 220464683 -- 6f8351be43a44a8f10bf20612b2cc744a4a911c7 by Jon Cohen <cohenjon@google.com>: Add VS Code and some Bazel output files to absl/.gitignore PiperOrigin-RevId: 220464362 -- 43fac22f8af6b6ed55309a784a9d25d837393d0e by Abseil Team <absl-team@google.com>: absl: fix SpinLock::EncodeWaitCycles If a thread has ever observed or set kSpinLockSleeper, it must never leave 0 in kWaitTimeMask because at this point it is expected to wake subsequent threads. Current calculations in EncodeWaitCycles can result in 0 in kWaitTimeMask and lead to missed wakeups. This is mostly theoretical today, because the futex call needs to finish within 128 cycles (futex can return immediately without waiting, but 128 cycles still look too low for this). But this can well fire in future if we bump granularity and/or threshold for recording contention. Use kSpinLockSleeper instead of 0. PiperOrigin-RevId: 220463123 -- def9b7e3d45c99d68cc52a4429256116d7f421f2 by Abseil Team <absl-team@google.com>: absl: optimize SpinLock::SlowLock Currently we record contention even after the first initial spin. This leads to several performance issues: 1. If we succeed in acquiring the lock after the initial spin, overall wait time can be within tens/hundreds of nanoseconds. Recording such low wait time looks completely unnecessary and excessive. From some point of view this is not even a wait, because we did not sleep. And, for example, Mutex does not record contention in this case. In majority of cases the lock should be acquired exactly during the initial spin, yet we still go through full overhead of submitting contention. 2. Whenever a thread submits contention it also calls FUTEX_WAKE (there is no way to understand if it's necessary or not when wait value is stored in the lock). So if there are just 2 threads and a brief contention, the second thread will still call FUTEX_WAKE which is completely unnecessary overhead. Don't record contention after the initial spin wait. FWIW this also removes 2 CycleClock::Now calls and EncodeWaitCycles from the common hot path. PiperOrigin-RevId: 220379972 -- 75b0c0cb214de904ea622f81ec3f4eabdc8695b0 by Derek Mauro <dmauro@google.com>: Supress MSVC warnings in raw_hash_set's use of TrailingZeros and LeadingZeros. https://github.com/abseil/abseil-cpp/issues/208 PiperOrigin-RevId: 220372204 GitOrigin-RevId: 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 Change-Id: I3a66af4e050810a3274e45d4e055b2aa19ffba1b
6 years ago
expected += one_element_std_vec_copy_allocated;
7 years ago
vec.push_back(vec[0]);
Export of internal Abseil changes. -- 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 by Jon Cohen <cohenjon@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 220523164 -- 59ef14fe7034a3148f1e9cef1f128b8ca264b444 by Jon Cohen <cohenjon@google.com>: Don't assume how much std::vector's constructors allocate in InlinedVector's test for scoped_allocator_adaptor support. PiperOrigin-RevId: 220464683 -- 6f8351be43a44a8f10bf20612b2cc744a4a911c7 by Jon Cohen <cohenjon@google.com>: Add VS Code and some Bazel output files to absl/.gitignore PiperOrigin-RevId: 220464362 -- 43fac22f8af6b6ed55309a784a9d25d837393d0e by Abseil Team <absl-team@google.com>: absl: fix SpinLock::EncodeWaitCycles If a thread has ever observed or set kSpinLockSleeper, it must never leave 0 in kWaitTimeMask because at this point it is expected to wake subsequent threads. Current calculations in EncodeWaitCycles can result in 0 in kWaitTimeMask and lead to missed wakeups. This is mostly theoretical today, because the futex call needs to finish within 128 cycles (futex can return immediately without waiting, but 128 cycles still look too low for this). But this can well fire in future if we bump granularity and/or threshold for recording contention. Use kSpinLockSleeper instead of 0. PiperOrigin-RevId: 220463123 -- def9b7e3d45c99d68cc52a4429256116d7f421f2 by Abseil Team <absl-team@google.com>: absl: optimize SpinLock::SlowLock Currently we record contention even after the first initial spin. This leads to several performance issues: 1. If we succeed in acquiring the lock after the initial spin, overall wait time can be within tens/hundreds of nanoseconds. Recording such low wait time looks completely unnecessary and excessive. From some point of view this is not even a wait, because we did not sleep. And, for example, Mutex does not record contention in this case. In majority of cases the lock should be acquired exactly during the initial spin, yet we still go through full overhead of submitting contention. 2. Whenever a thread submits contention it also calls FUTEX_WAKE (there is no way to understand if it's necessary or not when wait value is stored in the lock). So if there are just 2 threads and a brief contention, the second thread will still call FUTEX_WAKE which is completely unnecessary overhead. Don't record contention after the initial spin wait. FWIW this also removes 2 CycleClock::Now calls and EncodeWaitCycles from the common hot path. PiperOrigin-RevId: 220379972 -- 75b0c0cb214de904ea622f81ec3f4eabdc8695b0 by Derek Mauro <dmauro@google.com>: Supress MSVC warnings in raw_hash_set's use of TrailingZeros and LeadingZeros. https://github.com/abseil/abseil-cpp/issues/208 PiperOrigin-RevId: 220372204 GitOrigin-RevId: 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 Change-Id: I3a66af4e050810a3274e45d4e055b2aa19ffba1b
6 years ago
EXPECT_EQ(allocated, expected);
7 years ago
// Overflow the inlined memory.
// The absl::InlinedVector will now allocate.
Export of internal Abseil changes. -- 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 by Jon Cohen <cohenjon@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 220523164 -- 59ef14fe7034a3148f1e9cef1f128b8ca264b444 by Jon Cohen <cohenjon@google.com>: Don't assume how much std::vector's constructors allocate in InlinedVector's test for scoped_allocator_adaptor support. PiperOrigin-RevId: 220464683 -- 6f8351be43a44a8f10bf20612b2cc744a4a911c7 by Jon Cohen <cohenjon@google.com>: Add VS Code and some Bazel output files to absl/.gitignore PiperOrigin-RevId: 220464362 -- 43fac22f8af6b6ed55309a784a9d25d837393d0e by Abseil Team <absl-team@google.com>: absl: fix SpinLock::EncodeWaitCycles If a thread has ever observed or set kSpinLockSleeper, it must never leave 0 in kWaitTimeMask because at this point it is expected to wake subsequent threads. Current calculations in EncodeWaitCycles can result in 0 in kWaitTimeMask and lead to missed wakeups. This is mostly theoretical today, because the futex call needs to finish within 128 cycles (futex can return immediately without waiting, but 128 cycles still look too low for this). But this can well fire in future if we bump granularity and/or threshold for recording contention. Use kSpinLockSleeper instead of 0. PiperOrigin-RevId: 220463123 -- def9b7e3d45c99d68cc52a4429256116d7f421f2 by Abseil Team <absl-team@google.com>: absl: optimize SpinLock::SlowLock Currently we record contention even after the first initial spin. This leads to several performance issues: 1. If we succeed in acquiring the lock after the initial spin, overall wait time can be within tens/hundreds of nanoseconds. Recording such low wait time looks completely unnecessary and excessive. From some point of view this is not even a wait, because we did not sleep. And, for example, Mutex does not record contention in this case. In majority of cases the lock should be acquired exactly during the initial spin, yet we still go through full overhead of submitting contention. 2. Whenever a thread submits contention it also calls FUTEX_WAKE (there is no way to understand if it's necessary or not when wait value is stored in the lock). So if there are just 2 threads and a brief contention, the second thread will still call FUTEX_WAKE which is completely unnecessary overhead. Don't record contention after the initial spin wait. FWIW this also removes 2 CycleClock::Now calls and EncodeWaitCycles from the common hot path. PiperOrigin-RevId: 220379972 -- 75b0c0cb214de904ea622f81ec3f4eabdc8695b0 by Derek Mauro <dmauro@google.com>: Supress MSVC warnings in raw_hash_set's use of TrailingZeros and LeadingZeros. https://github.com/abseil/abseil-cpp/issues/208 PiperOrigin-RevId: 220372204 GitOrigin-RevId: 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 Change-Id: I3a66af4e050810a3274e45d4e055b2aa19ffba1b
6 years ago
expected += sizeof(StdVector) * 8 + default_std_vec_allocated * 3;
7 years ago
vec.resize(5);
Export of internal Abseil changes. -- 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 by Jon Cohen <cohenjon@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 220523164 -- 59ef14fe7034a3148f1e9cef1f128b8ca264b444 by Jon Cohen <cohenjon@google.com>: Don't assume how much std::vector's constructors allocate in InlinedVector's test for scoped_allocator_adaptor support. PiperOrigin-RevId: 220464683 -- 6f8351be43a44a8f10bf20612b2cc744a4a911c7 by Jon Cohen <cohenjon@google.com>: Add VS Code and some Bazel output files to absl/.gitignore PiperOrigin-RevId: 220464362 -- 43fac22f8af6b6ed55309a784a9d25d837393d0e by Abseil Team <absl-team@google.com>: absl: fix SpinLock::EncodeWaitCycles If a thread has ever observed or set kSpinLockSleeper, it must never leave 0 in kWaitTimeMask because at this point it is expected to wake subsequent threads. Current calculations in EncodeWaitCycles can result in 0 in kWaitTimeMask and lead to missed wakeups. This is mostly theoretical today, because the futex call needs to finish within 128 cycles (futex can return immediately without waiting, but 128 cycles still look too low for this). But this can well fire in future if we bump granularity and/or threshold for recording contention. Use kSpinLockSleeper instead of 0. PiperOrigin-RevId: 220463123 -- def9b7e3d45c99d68cc52a4429256116d7f421f2 by Abseil Team <absl-team@google.com>: absl: optimize SpinLock::SlowLock Currently we record contention even after the first initial spin. This leads to several performance issues: 1. If we succeed in acquiring the lock after the initial spin, overall wait time can be within tens/hundreds of nanoseconds. Recording such low wait time looks completely unnecessary and excessive. From some point of view this is not even a wait, because we did not sleep. And, for example, Mutex does not record contention in this case. In majority of cases the lock should be acquired exactly during the initial spin, yet we still go through full overhead of submitting contention. 2. Whenever a thread submits contention it also calls FUTEX_WAKE (there is no way to understand if it's necessary or not when wait value is stored in the lock). So if there are just 2 threads and a brief contention, the second thread will still call FUTEX_WAKE which is completely unnecessary overhead. Don't record contention after the initial spin wait. FWIW this also removes 2 CycleClock::Now calls and EncodeWaitCycles from the common hot path. PiperOrigin-RevId: 220379972 -- 75b0c0cb214de904ea622f81ec3f4eabdc8695b0 by Derek Mauro <dmauro@google.com>: Supress MSVC warnings in raw_hash_set's use of TrailingZeros and LeadingZeros. https://github.com/abseil/abseil-cpp/issues/208 PiperOrigin-RevId: 220372204 GitOrigin-RevId: 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 Change-Id: I3a66af4e050810a3274e45d4e055b2aa19ffba1b
6 years ago
EXPECT_EQ(allocated, expected);
7 years ago
// Adding one more in external mode should also work.
Export of internal Abseil changes. -- 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 by Jon Cohen <cohenjon@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 220523164 -- 59ef14fe7034a3148f1e9cef1f128b8ca264b444 by Jon Cohen <cohenjon@google.com>: Don't assume how much std::vector's constructors allocate in InlinedVector's test for scoped_allocator_adaptor support. PiperOrigin-RevId: 220464683 -- 6f8351be43a44a8f10bf20612b2cc744a4a911c7 by Jon Cohen <cohenjon@google.com>: Add VS Code and some Bazel output files to absl/.gitignore PiperOrigin-RevId: 220464362 -- 43fac22f8af6b6ed55309a784a9d25d837393d0e by Abseil Team <absl-team@google.com>: absl: fix SpinLock::EncodeWaitCycles If a thread has ever observed or set kSpinLockSleeper, it must never leave 0 in kWaitTimeMask because at this point it is expected to wake subsequent threads. Current calculations in EncodeWaitCycles can result in 0 in kWaitTimeMask and lead to missed wakeups. This is mostly theoretical today, because the futex call needs to finish within 128 cycles (futex can return immediately without waiting, but 128 cycles still look too low for this). But this can well fire in future if we bump granularity and/or threshold for recording contention. Use kSpinLockSleeper instead of 0. PiperOrigin-RevId: 220463123 -- def9b7e3d45c99d68cc52a4429256116d7f421f2 by Abseil Team <absl-team@google.com>: absl: optimize SpinLock::SlowLock Currently we record contention even after the first initial spin. This leads to several performance issues: 1. If we succeed in acquiring the lock after the initial spin, overall wait time can be within tens/hundreds of nanoseconds. Recording such low wait time looks completely unnecessary and excessive. From some point of view this is not even a wait, because we did not sleep. And, for example, Mutex does not record contention in this case. In majority of cases the lock should be acquired exactly during the initial spin, yet we still go through full overhead of submitting contention. 2. Whenever a thread submits contention it also calls FUTEX_WAKE (there is no way to understand if it's necessary or not when wait value is stored in the lock). So if there are just 2 threads and a brief contention, the second thread will still call FUTEX_WAKE which is completely unnecessary overhead. Don't record contention after the initial spin wait. FWIW this also removes 2 CycleClock::Now calls and EncodeWaitCycles from the common hot path. PiperOrigin-RevId: 220379972 -- 75b0c0cb214de904ea622f81ec3f4eabdc8695b0 by Derek Mauro <dmauro@google.com>: Supress MSVC warnings in raw_hash_set's use of TrailingZeros and LeadingZeros. https://github.com/abseil/abseil-cpp/issues/208 PiperOrigin-RevId: 220372204 GitOrigin-RevId: 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 Change-Id: I3a66af4e050810a3274e45d4e055b2aa19ffba1b
6 years ago
expected += one_element_std_vec_copy_allocated;
7 years ago
vec.push_back(vec[0]);
Export of internal Abseil changes. -- 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 by Jon Cohen <cohenjon@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 220523164 -- 59ef14fe7034a3148f1e9cef1f128b8ca264b444 by Jon Cohen <cohenjon@google.com>: Don't assume how much std::vector's constructors allocate in InlinedVector's test for scoped_allocator_adaptor support. PiperOrigin-RevId: 220464683 -- 6f8351be43a44a8f10bf20612b2cc744a4a911c7 by Jon Cohen <cohenjon@google.com>: Add VS Code and some Bazel output files to absl/.gitignore PiperOrigin-RevId: 220464362 -- 43fac22f8af6b6ed55309a784a9d25d837393d0e by Abseil Team <absl-team@google.com>: absl: fix SpinLock::EncodeWaitCycles If a thread has ever observed or set kSpinLockSleeper, it must never leave 0 in kWaitTimeMask because at this point it is expected to wake subsequent threads. Current calculations in EncodeWaitCycles can result in 0 in kWaitTimeMask and lead to missed wakeups. This is mostly theoretical today, because the futex call needs to finish within 128 cycles (futex can return immediately without waiting, but 128 cycles still look too low for this). But this can well fire in future if we bump granularity and/or threshold for recording contention. Use kSpinLockSleeper instead of 0. PiperOrigin-RevId: 220463123 -- def9b7e3d45c99d68cc52a4429256116d7f421f2 by Abseil Team <absl-team@google.com>: absl: optimize SpinLock::SlowLock Currently we record contention even after the first initial spin. This leads to several performance issues: 1. If we succeed in acquiring the lock after the initial spin, overall wait time can be within tens/hundreds of nanoseconds. Recording such low wait time looks completely unnecessary and excessive. From some point of view this is not even a wait, because we did not sleep. And, for example, Mutex does not record contention in this case. In majority of cases the lock should be acquired exactly during the initial spin, yet we still go through full overhead of submitting contention. 2. Whenever a thread submits contention it also calls FUTEX_WAKE (there is no way to understand if it's necessary or not when wait value is stored in the lock). So if there are just 2 threads and a brief contention, the second thread will still call FUTEX_WAKE which is completely unnecessary overhead. Don't record contention after the initial spin wait. FWIW this also removes 2 CycleClock::Now calls and EncodeWaitCycles from the common hot path. PiperOrigin-RevId: 220379972 -- 75b0c0cb214de904ea622f81ec3f4eabdc8695b0 by Derek Mauro <dmauro@google.com>: Supress MSVC warnings in raw_hash_set's use of TrailingZeros and LeadingZeros. https://github.com/abseil/abseil-cpp/issues/208 PiperOrigin-RevId: 220372204 GitOrigin-RevId: 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 Change-Id: I3a66af4e050810a3274e45d4e055b2aa19ffba1b
6 years ago
EXPECT_EQ(allocated, expected);
7 years ago
Export of internal Abseil changes. -- 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 by Jon Cohen <cohenjon@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 220523164 -- 59ef14fe7034a3148f1e9cef1f128b8ca264b444 by Jon Cohen <cohenjon@google.com>: Don't assume how much std::vector's constructors allocate in InlinedVector's test for scoped_allocator_adaptor support. PiperOrigin-RevId: 220464683 -- 6f8351be43a44a8f10bf20612b2cc744a4a911c7 by Jon Cohen <cohenjon@google.com>: Add VS Code and some Bazel output files to absl/.gitignore PiperOrigin-RevId: 220464362 -- 43fac22f8af6b6ed55309a784a9d25d837393d0e by Abseil Team <absl-team@google.com>: absl: fix SpinLock::EncodeWaitCycles If a thread has ever observed or set kSpinLockSleeper, it must never leave 0 in kWaitTimeMask because at this point it is expected to wake subsequent threads. Current calculations in EncodeWaitCycles can result in 0 in kWaitTimeMask and lead to missed wakeups. This is mostly theoretical today, because the futex call needs to finish within 128 cycles (futex can return immediately without waiting, but 128 cycles still look too low for this). But this can well fire in future if we bump granularity and/or threshold for recording contention. Use kSpinLockSleeper instead of 0. PiperOrigin-RevId: 220463123 -- def9b7e3d45c99d68cc52a4429256116d7f421f2 by Abseil Team <absl-team@google.com>: absl: optimize SpinLock::SlowLock Currently we record contention even after the first initial spin. This leads to several performance issues: 1. If we succeed in acquiring the lock after the initial spin, overall wait time can be within tens/hundreds of nanoseconds. Recording such low wait time looks completely unnecessary and excessive. From some point of view this is not even a wait, because we did not sleep. And, for example, Mutex does not record contention in this case. In majority of cases the lock should be acquired exactly during the initial spin, yet we still go through full overhead of submitting contention. 2. Whenever a thread submits contention it also calls FUTEX_WAKE (there is no way to understand if it's necessary or not when wait value is stored in the lock). So if there are just 2 threads and a brief contention, the second thread will still call FUTEX_WAKE which is completely unnecessary overhead. Don't record contention after the initial spin wait. FWIW this also removes 2 CycleClock::Now calls and EncodeWaitCycles from the common hot path. PiperOrigin-RevId: 220379972 -- 75b0c0cb214de904ea622f81ec3f4eabdc8695b0 by Derek Mauro <dmauro@google.com>: Supress MSVC warnings in raw_hash_set's use of TrailingZeros and LeadingZeros. https://github.com/abseil/abseil-cpp/issues/208 PiperOrigin-RevId: 220372204 GitOrigin-RevId: 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 Change-Id: I3a66af4e050810a3274e45d4e055b2aa19ffba1b
6 years ago
// And extending these should still work. This assumes that vec[0] does not
// need to grow its allocation.
expected += sizeof(int);
7 years ago
vec[0].push_back(1);
Export of internal Abseil changes. -- 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 by Jon Cohen <cohenjon@google.com>: Import of CCTZ from GitHub. PiperOrigin-RevId: 220523164 -- 59ef14fe7034a3148f1e9cef1f128b8ca264b444 by Jon Cohen <cohenjon@google.com>: Don't assume how much std::vector's constructors allocate in InlinedVector's test for scoped_allocator_adaptor support. PiperOrigin-RevId: 220464683 -- 6f8351be43a44a8f10bf20612b2cc744a4a911c7 by Jon Cohen <cohenjon@google.com>: Add VS Code and some Bazel output files to absl/.gitignore PiperOrigin-RevId: 220464362 -- 43fac22f8af6b6ed55309a784a9d25d837393d0e by Abseil Team <absl-team@google.com>: absl: fix SpinLock::EncodeWaitCycles If a thread has ever observed or set kSpinLockSleeper, it must never leave 0 in kWaitTimeMask because at this point it is expected to wake subsequent threads. Current calculations in EncodeWaitCycles can result in 0 in kWaitTimeMask and lead to missed wakeups. This is mostly theoretical today, because the futex call needs to finish within 128 cycles (futex can return immediately without waiting, but 128 cycles still look too low for this). But this can well fire in future if we bump granularity and/or threshold for recording contention. Use kSpinLockSleeper instead of 0. PiperOrigin-RevId: 220463123 -- def9b7e3d45c99d68cc52a4429256116d7f421f2 by Abseil Team <absl-team@google.com>: absl: optimize SpinLock::SlowLock Currently we record contention even after the first initial spin. This leads to several performance issues: 1. If we succeed in acquiring the lock after the initial spin, overall wait time can be within tens/hundreds of nanoseconds. Recording such low wait time looks completely unnecessary and excessive. From some point of view this is not even a wait, because we did not sleep. And, for example, Mutex does not record contention in this case. In majority of cases the lock should be acquired exactly during the initial spin, yet we still go through full overhead of submitting contention. 2. Whenever a thread submits contention it also calls FUTEX_WAKE (there is no way to understand if it's necessary or not when wait value is stored in the lock). So if there are just 2 threads and a brief contention, the second thread will still call FUTEX_WAKE which is completely unnecessary overhead. Don't record contention after the initial spin wait. FWIW this also removes 2 CycleClock::Now calls and EncodeWaitCycles from the common hot path. PiperOrigin-RevId: 220379972 -- 75b0c0cb214de904ea622f81ec3f4eabdc8695b0 by Derek Mauro <dmauro@google.com>: Supress MSVC warnings in raw_hash_set's use of TrailingZeros and LeadingZeros. https://github.com/abseil/abseil-cpp/issues/208 PiperOrigin-RevId: 220372204 GitOrigin-RevId: 178e7a9a76fc8fcd6df6335b59139cbe644a16b9 Change-Id: I3a66af4e050810a3274e45d4e055b2aa19ffba1b
6 years ago
EXPECT_EQ(allocated, expected);
7 years ago
vec.clear();
EXPECT_EQ(allocated, 0);
}
TEST(AllocatorSupportTest, SizeAllocConstructor) {
constexpr int inlined_size = 4;
using Alloc = CountingAllocator<int>;
using AllocVec = absl::InlinedVector<int, inlined_size, Alloc>;
{
auto len = inlined_size / 2;
int64_t allocated = 0;
auto v = AllocVec(len, Alloc(&allocated));
// Inline storage used; allocator should not be invoked
EXPECT_THAT(allocated, 0);
EXPECT_THAT(v, AllOf(SizeIs(len), Each(0)));
}
{
auto len = inlined_size * 2;
int64_t allocated = 0;
auto v = AllocVec(len, Alloc(&allocated));
// Out of line storage used; allocation of 8 elements expected
EXPECT_THAT(allocated, len * sizeof(int));
EXPECT_THAT(v, AllOf(SizeIs(len), Each(0)));
}
}
Export of internal Abseil changes. -- 1c1d6e2404dfc6caa022b335df5acdac6da50fe1 by Derek Mauro <dmauro@google.com>: Fix the internal namespacing in unaligned_access.h PiperOrigin-RevId: 215434506 -- 17d4400aebf025a230690fc1c7a968ef8d85bbba by Eric Fiselier <ericwf@google.com>: gtest depends on the GCC extension allowing variadic macros to be passed a empty parameter pack for ..., but LLVM diagnoses this as a GNU extension. This patch suppresses the warning when building the absl tests. PiperOrigin-RevId: 215426161 -- f2c49dde23a9f445b9de963f1bbe840ebb568b30 by Eric Fiselier <ericwf@google.com>: Use EXPECT_DEATH_IF_SUPPORTED instead of EXPECT_DEATH. This avoids breaking the test when gtest doesn't support death tests. PiperOrigin-RevId: 215423849 -- cd687c1e121709603f4fc3726b534f6a9c52cc89 by Eric Fiselier <ericwf@google.com>: Disable LLVM's -Wmissing-variable-declarations in tests. GCC's configuration already disables this via -Wno-missing-declarations, this change makes LLVM do the same. The warning would otherwise flag most tests which use ABSL_FLAG. PiperOrigin-RevId: 215407429 -- d14098824c84e3a8c8f6fb920e0335fb48fe2010 by Eric Fiselier <ericwf@google.com>: Fix local variable shadowing in city hash implementation. PiperOrigin-RevId: 215407249 -- 4b5e140ba743f0d231790a26c49083abb4329e2c by Abseil Team <absl-team@google.com>: Make raw_hash_set::reserve 2X fast when reserve doesn't do any allocation. Make raw_hash_set::reserve ~1% faster when reserve does some (128~4k) allocation. PiperOrigin-RevId: 215348727 -- 461161e65e04b801480aa117af2534c594654ccf by Eric Fiselier <ericwf@google.com>: Internal change PiperOrigin-RevId: 215272283 -- 50413ae31ad3d3a177257416acd8ede47a17bff2 by Eric Fiselier <ericwf@google.com>: Internal Change PiperOrigin-RevId: 215233183 -- 477be54c43d61019a8fe4e190e340eb52737d383 by Abseil Team <absl-team@google.com>: Clarify misleading comment on ABSL_ATTRIBUTE_UNUSED PiperOrigin-RevId: 215185496 -- 2cafa2b5287507d3a946682aee9ab13af6d471c9 by Matt Kulukundis <kfm@google.com>: Add support for absl::Hash to various absl in types. PiperOrigin-RevId: 215039569 -- 082248901991aa3d29be0ea3689c7f213cf0fd83 by Derek Mauro <dmauro@google.com>: Remove an instance of HAS_GLOBAL_STRING from hash_function_defaults.h PiperOrigin-RevId: 214989094 -- b929f61907f0786a6133e3a9d7287e339c0a0acb by Derek Mauro <dmauro@google.com>: Internal import of Github #174 Fix code snippet in comment https://github.com/abseil/abseil-cpp/pull/174 PiperOrigin-RevId: 214958849 -- f2c5e829eca11c352e121f56eefbf87083305023 by Derek Mauro <dmauro@google.com>: Internal import of GitHub #173 Fix CMake build for absl::container. https://github.com/abseil/abseil-cpp/pull/173 PiperOrigin-RevId: 214957796 -- d704f860f9fddafb99e34e6c5032e49f73874e10 by Abseil Team <absl-team@google.com>: Internal change PiperOrigin-RevId: 214828181 GitOrigin-RevId: 1c1d6e2404dfc6caa022b335df5acdac6da50fe1 Change-Id: I551de2b1ba0cbaf6856cd5959358cf6651179dea
6 years ago
7 years ago
} // anonymous namespace