tvl-depot/absl/base/spinlock_test_common.cc
Abseil Team 070f6e47b3 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
2018-11-07 16:54:23 -05:00

267 lines
9.4 KiB
C++

// 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.
// A bunch of threads repeatedly hash an array of ints protected by a
// spinlock. If the spinlock is working properly, all elements of the
// array should be equal at the end of the test.
#include <cstdint>
#include <limits>
#include <random>
#include <thread> // NOLINT(build/c++11)
#include <vector>
#include "gtest/gtest.h"
#include "absl/base/attributes.h"
#include "absl/base/internal/low_level_scheduling.h"
#include "absl/base/internal/scheduling_mode.h"
#include "absl/base/internal/spinlock.h"
#include "absl/base/internal/sysinfo.h"
#include "absl/base/macros.h"
#include "absl/synchronization/blocking_counter.h"
#include "absl/synchronization/notification.h"
constexpr int32_t kNumThreads = 10;
constexpr int32_t kIters = 1000;
namespace absl {
namespace base_internal {
// This is defined outside of anonymous namespace so that it can be
// a friend of SpinLock to access protected methods for testing.
struct SpinLockTest {
static uint32_t EncodeWaitCycles(int64_t wait_start_time,
int64_t wait_end_time) {
return SpinLock::EncodeWaitCycles(wait_start_time, wait_end_time);
}
static uint64_t DecodeWaitCycles(uint32_t lock_value) {
return SpinLock::DecodeWaitCycles(lock_value);
}
};
namespace {
static constexpr int kArrayLength = 10;
static uint32_t values[kArrayLength];
static SpinLock static_spinlock(base_internal::kLinkerInitialized);
static SpinLock static_cooperative_spinlock(
base_internal::kLinkerInitialized,
base_internal::SCHEDULE_COOPERATIVE_AND_KERNEL);
static SpinLock static_noncooperative_spinlock(
base_internal::kLinkerInitialized, base_internal::SCHEDULE_KERNEL_ONLY);
// Simple integer hash function based on the public domain lookup2 hash.
// http://burtleburtle.net/bob/c/lookup2.c
static uint32_t Hash32(uint32_t a, uint32_t c) {
uint32_t b = 0x9e3779b9UL; // The golden ratio; an arbitrary value.
a -= b; a -= c; a ^= (c >> 13);
b -= c; b -= a; b ^= (a << 8);
c -= a; c -= b; c ^= (b >> 13);
a -= b; a -= c; a ^= (c >> 12);
b -= c; b -= a; b ^= (a << 16);
c -= a; c -= b; c ^= (b >> 5);
a -= b; a -= c; a ^= (c >> 3);
b -= c; b -= a; b ^= (a << 10);
c -= a; c -= b; c ^= (b >> 15);
return c;
}
static void TestFunction(int thread_salt, SpinLock* spinlock) {
for (int i = 0; i < kIters; i++) {
SpinLockHolder h(spinlock);
for (int j = 0; j < kArrayLength; j++) {
const int index = (j + thread_salt) % kArrayLength;
values[index] = Hash32(values[index], thread_salt);
std::this_thread::yield();
}
}
}
static void ThreadedTest(SpinLock* spinlock) {
std::vector<std::thread> threads;
for (int i = 0; i < kNumThreads; ++i) {
threads.push_back(std::thread(TestFunction, i, spinlock));
}
for (auto& thread : threads) {
thread.join();
}
SpinLockHolder h(spinlock);
for (int i = 1; i < kArrayLength; i++) {
EXPECT_EQ(values[0], values[i]);
}
}
TEST(SpinLock, StackNonCooperativeDisablesScheduling) {
SpinLock spinlock(base_internal::SCHEDULE_KERNEL_ONLY);
spinlock.Lock();
EXPECT_FALSE(base_internal::SchedulingGuard::ReschedulingIsAllowed());
spinlock.Unlock();
}
TEST(SpinLock, StaticNonCooperativeDisablesScheduling) {
static_noncooperative_spinlock.Lock();
EXPECT_FALSE(base_internal::SchedulingGuard::ReschedulingIsAllowed());
static_noncooperative_spinlock.Unlock();
}
TEST(SpinLock, WaitCyclesEncoding) {
// These are implementation details not exported by SpinLock.
const int kProfileTimestampShift = 7;
const int kLockwordReservedShift = 3;
const uint32_t kSpinLockSleeper = 8;
// We should be able to encode up to (1^kMaxCycleBits - 1) without clamping
// but the lower kProfileTimestampShift will be dropped.
const int kMaxCyclesShift =
32 - kLockwordReservedShift + kProfileTimestampShift;
const uint64_t kMaxCycles = (int64_t{1} << kMaxCyclesShift) - 1;
// These bits should be zero after encoding.
const uint32_t kLockwordReservedMask = (1 << kLockwordReservedShift) - 1;
// These bits are dropped when wait cycles are encoded.
const uint64_t kProfileTimestampMask = (1 << kProfileTimestampShift) - 1;
// Test a bunch of random values
std::default_random_engine generator;
// Shift to avoid overflow below.
std::uniform_int_distribution<uint64_t> time_distribution(
0, std::numeric_limits<uint64_t>::max() >> 4);
std::uniform_int_distribution<uint64_t> cycle_distribution(0, kMaxCycles);
for (int i = 0; i < 100; i++) {
int64_t start_time = time_distribution(generator);
int64_t cycles = cycle_distribution(generator);
int64_t end_time = start_time + cycles;
uint32_t lock_value = SpinLockTest::EncodeWaitCycles(start_time, end_time);
EXPECT_EQ(0, lock_value & kLockwordReservedMask);
uint64_t decoded = SpinLockTest::DecodeWaitCycles(lock_value);
EXPECT_EQ(0, decoded & kProfileTimestampMask);
EXPECT_EQ(cycles & ~kProfileTimestampMask, decoded);
}
// Test corner cases
int64_t start_time = time_distribution(generator);
EXPECT_EQ(kSpinLockSleeper,
SpinLockTest::EncodeWaitCycles(start_time, start_time));
EXPECT_EQ(0, SpinLockTest::DecodeWaitCycles(0));
EXPECT_EQ(0, SpinLockTest::DecodeWaitCycles(kLockwordReservedMask));
EXPECT_EQ(kMaxCycles & ~kProfileTimestampMask,
SpinLockTest::DecodeWaitCycles(~kLockwordReservedMask));
// Check that we cannot produce kSpinLockSleeper during encoding.
int64_t sleeper_cycles =
kSpinLockSleeper << (kProfileTimestampShift - kLockwordReservedShift);
uint32_t sleeper_value =
SpinLockTest::EncodeWaitCycles(start_time, start_time + sleeper_cycles);
EXPECT_NE(sleeper_value, kSpinLockSleeper);
// Test clamping
uint32_t max_value =
SpinLockTest::EncodeWaitCycles(start_time, start_time + kMaxCycles);
uint64_t max_value_decoded = SpinLockTest::DecodeWaitCycles(max_value);
uint64_t expected_max_value_decoded = kMaxCycles & ~kProfileTimestampMask;
EXPECT_EQ(expected_max_value_decoded, max_value_decoded);
const int64_t step = (1 << kProfileTimestampShift);
uint32_t after_max_value =
SpinLockTest::EncodeWaitCycles(start_time, start_time + kMaxCycles + step);
uint64_t after_max_value_decoded =
SpinLockTest::DecodeWaitCycles(after_max_value);
EXPECT_EQ(expected_max_value_decoded, after_max_value_decoded);
uint32_t before_max_value = SpinLockTest::EncodeWaitCycles(
start_time, start_time + kMaxCycles - step);
uint64_t before_max_value_decoded =
SpinLockTest::DecodeWaitCycles(before_max_value);
EXPECT_GT(expected_max_value_decoded, before_max_value_decoded);
}
TEST(SpinLockWithThreads, StaticSpinLock) {
ThreadedTest(&static_spinlock);
}
TEST(SpinLockWithThreads, StackSpinLock) {
SpinLock spinlock;
ThreadedTest(&spinlock);
}
TEST(SpinLockWithThreads, StackCooperativeSpinLock) {
SpinLock spinlock(base_internal::SCHEDULE_COOPERATIVE_AND_KERNEL);
ThreadedTest(&spinlock);
}
TEST(SpinLockWithThreads, StackNonCooperativeSpinLock) {
SpinLock spinlock(base_internal::SCHEDULE_KERNEL_ONLY);
ThreadedTest(&spinlock);
}
TEST(SpinLockWithThreads, StaticCooperativeSpinLock) {
ThreadedTest(&static_cooperative_spinlock);
}
TEST(SpinLockWithThreads, StaticNonCooperativeSpinLock) {
ThreadedTest(&static_noncooperative_spinlock);
}
TEST(SpinLockWithThreads, DoesNotDeadlock) {
struct Helper {
static void NotifyThenLock(Notification* locked, SpinLock* spinlock,
BlockingCounter* b) {
locked->WaitForNotification(); // Wait for LockThenWait() to hold "s".
b->DecrementCount();
SpinLockHolder l(spinlock);
}
static void LockThenWait(Notification* locked, SpinLock* spinlock,
BlockingCounter* b) {
SpinLockHolder l(spinlock);
locked->Notify();
b->Wait();
}
static void DeadlockTest(SpinLock* spinlock, int num_spinners) {
Notification locked;
BlockingCounter counter(num_spinners);
std::vector<std::thread> threads;
threads.push_back(
std::thread(Helper::LockThenWait, &locked, spinlock, &counter));
for (int i = 0; i < num_spinners; ++i) {
threads.push_back(
std::thread(Helper::NotifyThenLock, &locked, spinlock, &counter));
}
for (auto& thread : threads) {
thread.join();
}
}
};
SpinLock stack_cooperative_spinlock(
base_internal::SCHEDULE_COOPERATIVE_AND_KERNEL);
SpinLock stack_noncooperative_spinlock(base_internal::SCHEDULE_KERNEL_ONLY);
Helper::DeadlockTest(&stack_cooperative_spinlock,
base_internal::NumCPUs() * 2);
Helper::DeadlockTest(&stack_noncooperative_spinlock,
base_internal::NumCPUs() * 2);
Helper::DeadlockTest(&static_cooperative_spinlock,
base_internal::NumCPUs() * 2);
Helper::DeadlockTest(&static_noncooperative_spinlock,
base_internal::NumCPUs() * 2);
}
} // namespace
} // namespace base_internal
} // namespace absl