Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Modify atomic tests to use atomic load #84

Merged
merged 1 commit into from
Apr 13, 2022
Merged

Modify atomic tests to use atomic load #84

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
52 changes: 27 additions & 25 deletions tests/backend.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -68,18 +68,19 @@ class backendTest : public testing::Test, public ::testing::WithParamInterface<i
TEST_F(backendTest, atomicXchgAll) {
global_char _c(argo::conew_<char>(0));
argo::backend::atomic::exchange(_c, c_const);
ASSERT_EQ(c_const, *_c);
ASSERT_EQ(c_const, argo::backend::atomic::load(_c));

global_int _i(argo::conew_<int>(0));
argo::backend::atomic::exchange(_i, i_const);
ASSERT_EQ(i_const, *_i);
ASSERT_EQ(i_const, argo::backend::atomic::load(_i));

global_uint _j(argo::conew_<unsigned>(0));
argo::backend::atomic::exchange(_j, j_const);
ASSERT_EQ(j_const, *_j);
ASSERT_EQ(j_const, argo::backend::atomic::load(_j));

global_double _d(argo::conew_<double>(0));
argo::backend::atomic::exchange(_d, d_const);
ASSERT_EQ(d_const, *_d);
ASSERT_EQ(d_const, argo::backend::atomic::load(_d));
}

/**
Expand All @@ -92,30 +93,31 @@ TEST_F(backendTest, atomicXchgOne) {
if (argo::backend::node_id() == 0)
argo::backend::atomic::exchange(_c, c_const);
argo::backend::barrier();
ASSERT_EQ(c_const, *_c);
ASSERT_EQ(c_const, argo::backend::atomic::load(_c));

global_int _i(argo::conew_<int>());
if (argo::backend::node_id() == 0)
argo::backend::atomic::exchange(_i, i_const);
argo::backend::barrier();
ASSERT_EQ(i_const, *_i);
ASSERT_EQ(i_const, argo::backend::atomic::load(_i));

global_uint _j(argo::conew_<unsigned>());
if (argo::backend::node_id() == 0)
argo::backend::atomic::exchange(_j, j_const);
argo::backend::barrier();
ASSERT_EQ(j_const, *_j);
ASSERT_EQ(j_const, argo::backend::atomic::load(_j));
argo::backend::barrier();
// The following test is here to test implicit type conversions
if (argo::backend::node_id() == 0)
argo::backend::atomic::exchange(_j, i_const);
argo::backend::barrier();
ASSERT_EQ(static_cast<unsigned>(i_const), *_j);
ASSERT_EQ(static_cast<unsigned>(i_const), argo::backend::atomic::load(_j));

global_double _d(argo::conew_<double>());
if (argo::backend::node_id() == 0)
argo::backend::atomic::exchange(_d, d_const);
argo::backend::barrier();
ASSERT_EQ(d_const, *_d);
ASSERT_EQ(d_const, argo::backend::atomic::load(_d));
}

/**
Expand All @@ -126,30 +128,31 @@ TEST_F(backendTest, storeOne) {
if (argo::backend::node_id() == 0)
argo::backend::atomic::store(_c, c_const);
argo::backend::barrier();
ASSERT_EQ(c_const, *_c);
ASSERT_EQ(c_const, argo::backend::atomic::load(_c));

global_int _i(argo::conew_<int>());
if (argo::backend::node_id() == 0)
argo::backend::atomic::store(_i, i_const);
argo::backend::barrier();
ASSERT_EQ(i_const, *_i);
ASSERT_EQ(i_const, argo::backend::atomic::load(_i));

global_uint _j(argo::conew_<unsigned>());
if (argo::backend::node_id() == 0)
argo::backend::atomic::store(_j, j_const);
argo::backend::barrier();
ASSERT_EQ(j_const, *_j);
ASSERT_EQ(j_const, argo::backend::atomic::load(_j));
argo::backend::barrier();
// The following test is here to test implicit type conversions
if (argo::backend::node_id() == 0)
argo::backend::atomic::store(_j, i_const);
argo::backend::barrier();
ASSERT_EQ(static_cast<unsigned>(i_const), *_j);
ASSERT_EQ(static_cast<unsigned>(i_const), argo::backend::atomic::load(_j));

global_double _d(argo::conew_<double>());
if (argo::backend::node_id() == 0)
argo::backend::atomic::store(_d, d_const);
argo::backend::barrier();
ASSERT_EQ(d_const, *_d);
ASSERT_EQ(d_const, argo::backend::atomic::load(_d));
}

/**
Expand All @@ -167,30 +170,27 @@ TEST_F(backendTest, loadOne) {
std::chrono::system_clock::now() + deadlock_threshold;
while(argo::backend::atomic::load(_c) != c_const)
ASSERT_LT(std::chrono::system_clock::now(), max_time);
ASSERT_EQ(c_const, *_c);
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is a slight change of behaviour, but ultimately I think it is sufficient to check that the while loop has finished in order to verify the removed ASSERT_EQ.


global_int _i(argo::conew_<int>());
if (argo::backend::node_id() == 0)
argo::backend::atomic::store(_i, i_const);
max_time = std::chrono::system_clock::now() + deadlock_threshold;
while(argo::backend::atomic::load(_i) != i_const)
ASSERT_LT(std::chrono::system_clock::now(), max_time);
ASSERT_EQ(i_const, *_i);

global_uint _j(argo::conew_<unsigned>());
if (argo::backend::node_id() == 0)
argo::backend::atomic::store(_j, j_const);
max_time = std::chrono::system_clock::now() + deadlock_threshold;
while(argo::backend::atomic::load(_j) != j_const)
ASSERT_LT(std::chrono::system_clock::now(), max_time);
ASSERT_EQ(j_const, *_j);

global_double _d(argo::conew_<double>());
if (argo::backend::node_id() == 0)
argo::backend::atomic::store(_d, d_const);
max_time = std::chrono::system_clock::now() + deadlock_threshold;
while(argo::backend::atomic::load(_d) != d_const)
ASSERT_LT(std::chrono::system_clock::now(), max_time);
ASSERT_EQ(d_const, *_d);
}

/**
Expand All @@ -215,7 +215,8 @@ TEST_F(backendTest, atomicXchgAtomicity) {
argo::barrier();
bool found = false;
for (int i = 0; i < argo::number_of_nodes(); ++i) {
if (rcs[i]) {
global_int _rcs(&rcs[i]);
if (argo::backend::atomic::load(_rcs)) {
ASSERT_FALSE(found);
found = true;
}
Expand Down Expand Up @@ -294,7 +295,8 @@ TEST_F(backendTest, atomicCASAtomicity) {
argo::barrier();
int count = 0;
for (int i = 0; i < argo::number_of_nodes(); ++i) {
if (rcs[i]) {
global_uint _rcs(&rcs[i]);
if (argo::backend::atomic::load(_rcs)) {
++count;
}
}
Expand All @@ -319,7 +321,7 @@ TEST_F(backendTest, atomicCASAtomicityStress) {
}

argo::barrier();
ASSERT_EQ(10000*argo::number_of_nodes(), *counter);
ASSERT_EQ(10000*argo::number_of_nodes(), argo::backend::atomic::load(counter));
}

/**
Expand All @@ -334,7 +336,7 @@ TEST_F(backendTest, atomicFetchAddInt) {
}

argo::barrier();
ASSERT_EQ(10000*argo::number_of_nodes(), *counter);
ASSERT_EQ(10000*argo::number_of_nodes(), argo::backend::atomic::load(counter));
}

/**
Expand All @@ -354,7 +356,7 @@ TEST_F(backendTest, atomicFetchAddUInt) {
}

argo::barrier();
ASSERT_EQ(std::numeric_limits<unsigned>::max(), *counter);
ASSERT_EQ(std::numeric_limits<unsigned>::max(), argo::backend::atomic::load(counter));
}

/**
Expand All @@ -371,7 +373,7 @@ TEST_F(backendTest, atomicFetchAddFloat) {
argo::barrier();
// If this fails, make sure it's not because of floating point precision
// issues
ASSERT_EQ(10000*argo::number_of_nodes(), *counter);
ASSERT_EQ(10000*argo::number_of_nodes(), argo::backend::atomic::load(counter));
}

/**
Expand All @@ -384,7 +386,7 @@ TEST_F(backendTest, atomicFetchAddPointer) {
// about the atomicity etc
if (argo::node_id() == 0) {
int * old_ptr = argo::backend::atomic::fetch_add(ptr, 2);
ASSERT_EQ(2, *ptr.get() - old_ptr);
ASSERT_EQ(2, argo::backend::atomic::load(ptr) - old_ptr);
}
}

Expand Down