Browse Source

Add CheckQueue Tests

0.15
Jeremy Rubin 8 years ago
parent
commit
96c7f2c345
  1. 1
      src/Makefile.test.include
  2. 442
      src/test/checkqueue_tests.cpp

1
src/Makefile.test.include

@ -67,6 +67,7 @@ BITCOIN_TESTS =\ @@ -67,6 +67,7 @@ BITCOIN_TESTS =\
test/blockencodings_tests.cpp \
test/bloom_tests.cpp \
test/bswap_tests.cpp \
test/checkqueue_tests.cpp \
test/coins_tests.cpp \
test/compress_tests.cpp \
test/crypto_tests.cpp \

442
src/test/checkqueue_tests.cpp

@ -0,0 +1,442 @@ @@ -0,0 +1,442 @@
// Copyright (c) 2012-2017 The Bitcoin Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#include "util.h"
#include "utiltime.h"
#include "validation.h"
#include "test/test_bitcoin.h"
#include "checkqueue.h"
#include <boost/test/unit_test.hpp>
#include <boost/thread.hpp>
#include <atomic>
#include <thread>
#include <vector>
#include <mutex>
#include <condition_variable>
#include <unordered_set>
#include <memory>
#include "random.h"
// BasicTestingSetup not sufficient because nScriptCheckThreads is not set
// otherwise.
BOOST_FIXTURE_TEST_SUITE(checkqueue_tests, TestingSetup)
static const int QUEUE_BATCH_SIZE = 128;
struct FakeCheck {
bool operator()()
{
return true;
}
void swap(FakeCheck& x){};
};
struct FakeCheckCheckCompletion {
static std::atomic<size_t> n_calls;
bool operator()()
{
++n_calls;
return true;
}
void swap(FakeCheckCheckCompletion& x){};
};
struct FailingCheck {
bool fails;
FailingCheck(bool fails) : fails(fails){};
FailingCheck() : fails(true){};
bool operator()()
{
return !fails;
}
void swap(FailingCheck& x)
{
std::swap(fails, x.fails);
};
};
struct UniqueCheck {
static std::mutex m;
static std::unordered_multiset<size_t> results;
size_t check_id;
UniqueCheck(size_t check_id_in) : check_id(check_id_in){};
UniqueCheck() : check_id(0){};
bool operator()()
{
std::lock_guard<std::mutex> l(m);
results.insert(check_id);
return true;
}
void swap(UniqueCheck& x) { std::swap(x.check_id, check_id); };
};
struct MemoryCheck {
static std::atomic<size_t> fake_allocated_memory;
bool b {false};
bool operator()()
{
return true;
}
MemoryCheck(){};
MemoryCheck(const MemoryCheck& x)
{
// We have to do this to make sure that destructor calls are paired
//
// Really, copy constructor should be deletable, but CCheckQueue breaks
// if it is deleted because of internal push_back.
fake_allocated_memory += b;
};
MemoryCheck(bool b_) : b(b_)
{
fake_allocated_memory += b;
};
~MemoryCheck(){
fake_allocated_memory -= b;
};
void swap(MemoryCheck& x) { std::swap(b, x.b); };
};
struct FrozenCleanupCheck {
static std::atomic<uint64_t> nFrozen;
static std::condition_variable cv;
static std::mutex m;
// Freezing can't be the default initialized behavior given how the queue
// swaps in default initialized Checks.
bool should_freeze {false};
bool operator()()
{
return true;
}
FrozenCleanupCheck() {}
~FrozenCleanupCheck()
{
if (should_freeze) {
std::unique_lock<std::mutex> l(m);
nFrozen = 1;
cv.notify_one();
cv.wait(l, []{ return nFrozen == 0;});
}
}
void swap(FrozenCleanupCheck& x){std::swap(should_freeze, x.should_freeze);};
};
// Static Allocations
std::mutex FrozenCleanupCheck::m{};
std::atomic<uint64_t> FrozenCleanupCheck::nFrozen{0};
std::condition_variable FrozenCleanupCheck::cv{};
std::mutex UniqueCheck::m;
std::unordered_multiset<size_t> UniqueCheck::results;
std::atomic<size_t> FakeCheckCheckCompletion::n_calls{0};
std::atomic<size_t> MemoryCheck::fake_allocated_memory{0};
// Queue Typedefs
typedef CCheckQueue<FakeCheckCheckCompletion> Correct_Queue;
typedef CCheckQueue<FakeCheck> Standard_Queue;
typedef CCheckQueue<FailingCheck> Failing_Queue;
typedef CCheckQueue<UniqueCheck> Unique_Queue;
typedef CCheckQueue<MemoryCheck> Memory_Queue;
typedef CCheckQueue<FrozenCleanupCheck> FrozenCleanup_Queue;
/** This test case checks that the CCheckQueue works properly
* with each specified size_t Checks pushed.
*/
void Correct_Queue_range(std::vector<size_t> range)
{
auto small_queue = std::unique_ptr<Correct_Queue>(new Correct_Queue {QUEUE_BATCH_SIZE});
boost::thread_group tg;
for (auto x = 0; x < nScriptCheckThreads; ++x) {
tg.create_thread([&]{small_queue->Thread();});
}
// Make vChecks here to save on malloc (this test can be slow...)
std::vector<FakeCheckCheckCompletion> vChecks;
for (auto i : range) {
size_t total = i;
FakeCheckCheckCompletion::n_calls = 0;
CCheckQueueControl<FakeCheckCheckCompletion> control(small_queue.get());
while (total) {
vChecks.resize(std::min(total, (size_t) GetRand(10)));
total -= vChecks.size();
control.Add(vChecks);
}
BOOST_REQUIRE(control.Wait());
if (FakeCheckCheckCompletion::n_calls != i) {
BOOST_REQUIRE_EQUAL(FakeCheckCheckCompletion::n_calls, i);
BOOST_TEST_MESSAGE("Failure on trial " << i << " expected, got " << FakeCheckCheckCompletion::n_calls);
}
}
tg.interrupt_all();
tg.join_all();
}
/** Test that 0 checks is correct
*/
BOOST_AUTO_TEST_CASE(test_CheckQueue_Correct_Zero)
{
std::vector<size_t> range;
range.push_back((size_t)0);
Correct_Queue_range(range);
}
/** Test that 1 check is correct
*/
BOOST_AUTO_TEST_CASE(test_CheckQueue_Correct_One)
{
std::vector<size_t> range;
range.push_back((size_t)1);
Correct_Queue_range(range);
}
/** Test that MAX check is correct
*/
BOOST_AUTO_TEST_CASE(test_CheckQueue_Correct_Max)
{
std::vector<size_t> range;
range.push_back(100000);
Correct_Queue_range(range);
}
/** Test that random numbers of checks are correct
*/
BOOST_AUTO_TEST_CASE(test_CheckQueue_Correct_Random)
{
std::vector<size_t> range;
range.reserve(100000/1000);
for (size_t i = 2; i < 100000; i += std::max((size_t)1, (size_t)GetRand(std::min((size_t)1000, ((size_t)100000) - i))))
range.push_back(i);
Correct_Queue_range(range);
}
/** Test that failing checks are caught */
BOOST_AUTO_TEST_CASE(test_CheckQueue_Catches_Failure)
{
auto fail_queue = std::unique_ptr<Failing_Queue>(new Failing_Queue {QUEUE_BATCH_SIZE});
boost::thread_group tg;
for (auto x = 0; x < nScriptCheckThreads; ++x) {
tg.create_thread([&]{fail_queue->Thread();});
}
for (size_t i = 0; i < 1001; ++i) {
CCheckQueueControl<FailingCheck> control(fail_queue.get());
size_t remaining = i;
while (remaining) {
size_t r = GetRand(10);
std::vector<FailingCheck> vChecks;
vChecks.reserve(r);
for (size_t k = 0; k < r && remaining; k++, remaining--)
vChecks.emplace_back(remaining == 1);
control.Add(vChecks);
}
bool success = control.Wait();
if (i > 0) {
BOOST_REQUIRE(!success);
} else if (i == 0) {
BOOST_REQUIRE(success);
}
}
tg.interrupt_all();
tg.join_all();
}
// Test that a block validation which fails does not interfere with
// future blocks, ie, the bad state is cleared.
BOOST_AUTO_TEST_CASE(test_CheckQueue_Recovers_From_Failure)
{
auto fail_queue = std::unique_ptr<Failing_Queue>(new Failing_Queue {QUEUE_BATCH_SIZE});
boost::thread_group tg;
for (auto x = 0; x < nScriptCheckThreads; ++x) {
tg.create_thread([&]{fail_queue->Thread();});
}
for (auto times = 0; times < 10; ++times) {
for (bool end_fails : {true, false}) {
CCheckQueueControl<FailingCheck> control(fail_queue.get());
{
std::vector<FailingCheck> vChecks;
vChecks.resize(100, false);
vChecks[99] = end_fails;
control.Add(vChecks);
}
bool r =control.Wait();
BOOST_REQUIRE(r || end_fails);
}
}
tg.interrupt_all();
tg.join_all();
}
// Test that unique checks are actually all called individually, rather than
// just one check being called repeatedly. Test that checks are not called
// more than once as well
BOOST_AUTO_TEST_CASE(test_CheckQueue_UniqueCheck)
{
auto queue = std::unique_ptr<Unique_Queue>(new Unique_Queue {QUEUE_BATCH_SIZE});
boost::thread_group tg;
for (auto x = 0; x < nScriptCheckThreads; ++x) {
tg.create_thread([&]{queue->Thread();});
}
size_t COUNT = 100000;
size_t total = COUNT;
{
CCheckQueueControl<UniqueCheck> control(queue.get());
while (total) {
size_t r = GetRand(10);
std::vector<UniqueCheck> vChecks;
for (size_t k = 0; k < r && total; k++)
vChecks.emplace_back(--total);
control.Add(vChecks);
}
}
bool r = true;
BOOST_REQUIRE_EQUAL(UniqueCheck::results.size(), COUNT);
for (size_t i = 0; i < COUNT; ++i)
r = r && UniqueCheck::results.count(i) == 1;
BOOST_REQUIRE(r);
tg.interrupt_all();
tg.join_all();
}
// Test that blocks which might allocate lots of memory free their memory agressively.
//
// This test attempts to catch a pathological case where by lazily freeing
// checks might mean leaving a check un-swapped out, and decreasing by 1 each
// time could leave the data hanging across a sequence of blocks.
BOOST_AUTO_TEST_CASE(test_CheckQueue_Memory)
{
auto queue = std::unique_ptr<Memory_Queue>(new Memory_Queue {QUEUE_BATCH_SIZE});
boost::thread_group tg;
for (auto x = 0; x < nScriptCheckThreads; ++x) {
tg.create_thread([&]{queue->Thread();});
}
for (size_t i = 0; i < 1000; ++i) {
size_t total = i;
{
CCheckQueueControl<MemoryCheck> control(queue.get());
while (total) {
size_t r = GetRand(10);
std::vector<MemoryCheck> vChecks;
for (size_t k = 0; k < r && total; k++) {
total--;
// Each iteration leaves data at the front, back, and middle
// to catch any sort of deallocation failure
vChecks.emplace_back(total == 0 || total == i || total == i/2);
}
control.Add(vChecks);
}
}
BOOST_REQUIRE_EQUAL(MemoryCheck::fake_allocated_memory, 0);
}
tg.interrupt_all();
tg.join_all();
}
// Test that a new verification cannot occur until all checks
// have been destructed
BOOST_AUTO_TEST_CASE(test_CheckQueue_FrozenCleanup)
{
auto queue = std::unique_ptr<FrozenCleanup_Queue>(new FrozenCleanup_Queue {QUEUE_BATCH_SIZE});
boost::thread_group tg;
bool fails = false;
for (auto x = 0; x < nScriptCheckThreads; ++x) {
tg.create_thread([&]{queue->Thread();});
}
std::thread t0([&]() {
CCheckQueueControl<FrozenCleanupCheck> control(queue.get());
std::vector<FrozenCleanupCheck> vChecks(1);
// Freezing can't be the default initialized behavior given how the queue
// swaps in default initialized Checks (otherwise freezing destructor
// would get called twice).
vChecks[0].should_freeze = true;
control.Add(vChecks);
control.Wait(); // Hangs here
});
{
std::unique_lock<std::mutex> l(FrozenCleanupCheck::m);
// Wait until the queue has finished all jobs and frozen
FrozenCleanupCheck::cv.wait(l, [](){return FrozenCleanupCheck::nFrozen == 1;});
// Try to get control of the queue a bunch of times
for (auto x = 0; x < 100 && !fails; ++x) {
fails = queue->ControlMutex.try_lock();
}
// Unfreeze
FrozenCleanupCheck::nFrozen = 0;
}
// Awaken frozen destructor
FrozenCleanupCheck::cv.notify_one();
// Wait for control to finish
t0.join();
tg.interrupt_all();
tg.join_all();
BOOST_REQUIRE(!fails);
}
/** Test that CCheckQueueControl is threadsafe */
BOOST_AUTO_TEST_CASE(test_CheckQueueControl_Locks)
{
auto queue = std::unique_ptr<Standard_Queue>(new Standard_Queue{QUEUE_BATCH_SIZE});
{
boost::thread_group tg;
std::atomic<int> nThreads {0};
std::atomic<int> fails {0};
for (size_t i = 0; i < 3; ++i) {
tg.create_thread(
[&]{
CCheckQueueControl<FakeCheck> control(queue.get());
// While sleeping, no other thread should execute to this point
auto observed = ++nThreads;
MilliSleep(10);
fails += observed != nThreads;
});
}
tg.join_all();
BOOST_REQUIRE_EQUAL(fails, 0);
}
{
boost::thread_group tg;
std::mutex m;
bool has_lock {false};
bool has_tried {false};
bool done {false};
bool done_ack {false};
std::condition_variable cv;
{
std::unique_lock<std::mutex> l(m);
tg.create_thread([&]{
CCheckQueueControl<FakeCheck> control(queue.get());
std::unique_lock<std::mutex> l(m);
has_lock = true;
cv.notify_one();
cv.wait(l, [&]{return has_tried;});
done = true;
cv.notify_one();
// Wait until the done is acknowledged
//
cv.wait(l, [&]{return done_ack;});
});
// Wait for thread to get the lock
cv.wait(l, [&](){return has_lock;});
bool fails = false;
for (auto x = 0; x < 100 && !fails; ++x) {
fails = queue->ControlMutex.try_lock();
}
has_tried = true;
cv.notify_one();
cv.wait(l, [&](){return done;});
// Acknowledge the done
done_ack = true;
cv.notify_one();
BOOST_REQUIRE(!fails);
}
tg.join_all();
}
}
BOOST_AUTO_TEST_SUITE_END()
Loading…
Cancel
Save