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

[GPU] Fix strided slice kernel with begin/end/stride as inputs #16302

Merged
merged 1 commit into from
Mar 15, 2023
Merged

[GPU] Fix strided slice kernel with begin/end/stride as inputs #16302

Show file tree
Hide file tree
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
48 changes: 21 additions & 27 deletions src/plugins/intel_gpu/src/kernel_selector/cl_kernels/strided_slice_ref.cl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,17 +10,15 @@ inline void FUNC(get_slice_step)(OPTIONAL_SHAPE_INFO_ARG
int* step_batch, int* step_feature,
int* step_z, int* step_y, int* step_x)
{
const uint batch_index = 0;
const uint feature_index = 1;
#ifdef OUTPUT_LAYOUT_BFYX
const uint batch_index = STRIDE_GET_INDEX(0, 0, 0, 0);
const uint feature_index = STRIDE_GET_INDEX(1, 0, 0, 0);
const uint y_index = STRIDE_GET_INDEX(2, 0, 0, 0);
const uint x_index = STRIDE_GET_INDEX(3, 0, 0, 0);
const uint y_index = 2;
const uint x_index = 3;
#elif OUTPUT_LAYOUT_BFZYX
const uint batch_index = STRIDE_GET_INDEX(0, 0, 0, 0, 0);
const uint feature_index = STRIDE_GET_INDEX(1, 0, 0, 0, 0);
const uint z_index = STRIDE_GET_INDEX(2, 0, 0, 0, 0);
const uint y_index = STRIDE_GET_INDEX(3, 0, 0, 0, 0);
const uint x_index = STRIDE_GET_INDEX(4, 0, 0, 0, 0);
const uint z_index = 2;
const uint y_index = 3;
const uint x_index = 4;
#endif

*step_batch = batch_index < STRIDE_DIMS ? stride[batch_index] : 1;
Expand Down Expand Up @@ -59,17 +57,15 @@ inline void FUNC(get_slice_end)(OPTIONAL_SHAPE_INFO_ARG
const uint out_z_num = INPUT0_SIZE_Z;
const uint out_y_num = INPUT0_SIZE_Y;
const uint out_x_num = INPUT0_SIZE_X;
const uint batch_index = 0;
const uint feature_index = 1;
#ifdef OUTPUT_LAYOUT_BFYX
const uint batch_index = END_GET_INDEX(0, 0, 0, 0);
const uint feature_index = END_GET_INDEX(1, 0, 0, 0);
const uint y_index = END_GET_INDEX(2, 0, 0, 0);
const uint x_index = END_GET_INDEX(3, 0, 0, 0);
const uint y_index = 2;
const uint x_index = 3;
#elif OUTPUT_LAYOUT_BFZYX
const uint batch_index = END_GET_INDEX(0, 0, 0, 0, 0);
const uint feature_index = END_GET_INDEX(1, 0, 0, 0, 0);
const uint z_index = END_GET_INDEX(2, 0, 0, 0, 0);
const uint y_index = END_GET_INDEX(3, 0, 0, 0, 0);
const uint x_index = END_GET_INDEX(4, 0, 0, 0, 0);
const uint z_index = 2;
const uint y_index = 3;
const uint x_index = 4;
#endif
END_TYPE batch = batch_index < END_DIMS ? end[batch_index] : 0;
END_TYPE feature = feature_index < END_DIMS ? end[feature_index] : 0;
Expand Down Expand Up @@ -176,17 +172,15 @@ inline void FUNC(get_slice_begin)(OPTIONAL_SHAPE_INFO_ARG
const uint out_z_num = INPUT0_SIZE_Z;
const uint out_y_num = INPUT0_SIZE_Y;
const uint out_x_num = INPUT0_SIZE_X;
const uint batch_index = 0;
const uint feature_index = 1;
#ifdef OUTPUT_LAYOUT_BFYX
const uint batch_index = STRIDE_GET_INDEX(0, 0, 0, 0);
const uint feature_index = STRIDE_GET_INDEX(1, 0, 0, 0);
const uint y_index = STRIDE_GET_INDEX(2, 0, 0, 0);
const uint x_index = STRIDE_GET_INDEX(3, 0, 0, 0);
const uint y_index = 2;
const uint x_index = 3;
#elif OUTPUT_LAYOUT_BFZYX
const uint batch_index = STRIDE_GET_INDEX(0, 0, 0, 0, 0);
const uint feature_index = STRIDE_GET_INDEX(1, 0, 0, 0, 0);
const uint z_index = STRIDE_GET_INDEX(2, 0, 0, 0, 0);
const uint y_index = STRIDE_GET_INDEX(3, 0, 0, 0, 0);
const uint x_index = STRIDE_GET_INDEX(4, 0, 0, 0, 0);
const uint z_index = 2;
const uint y_index = 3;
const uint x_index = 4;
#endif
BEGIN_TYPE batch = batch_index < BEGIN_DIMS ? begin[batch_index] : 0;
BEGIN_TYPE feature = feature_index < BEGIN_DIMS ? begin[feature_index] : 0;
Expand Down
59 changes: 58 additions & 1 deletion src/plugins/intel_gpu/tests/test_cases/strided_slice_gpu_test.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -620,6 +620,59 @@ class strided_slice_gpu: public ::testing::Test {
}
}

void test_2x2x2_all_dynamic_bcast() {
auto& engine = get_test_engine();
auto input_lay = layout{ ov::PartialShape::dynamic(3), data_types::f32, format::bfyx };
auto input = engine.allocate_memory({ ov::PartialShape{ 2, 2, 2 }, data_types::f32, format::bfyx });
auto begin = engine.allocate_memory({ ov::PartialShape{ 1 }, data_types::i64, format::bfyx });
auto end = engine.allocate_memory({ ov::PartialShape{ 1 }, data_types::i64, format::bfyx });
auto strides = engine.allocate_memory({ ov::PartialShape{ 1 }, data_types::i64, format::bfyx });

set_values(input, {0.0f, 1.0f, 2.0f, 3.0f, 4.0f, 5.0f, 6.0f, 7.0f});
set_values<int64_t>(begin, {1});
set_values<int64_t>(end, {2});
set_values<int64_t>(strides, {1});

topology topology;
topology.add(input_layout("input", input_lay));
topology.add(data("input2", begin));
topology.add(data("input3", end));
topology.add(data("input4", strides));
topology.add(strided_slice("strided_slice", input_info("input"), input_info("input2"), input_info("input3"), input_info("input4"), {}, {}, {}, {}, {}, {}));

ExecutionConfig config;
config.set_property(ov::intel_gpu::allow_new_shape_infer(true));
network network(engine, topology, config);

network.set_input_data("input", input);

auto inst = network.get_primitive("strided_slice");
auto impl = inst->get_impl();
ASSERT_TRUE(impl != nullptr);
ASSERT_TRUE(impl->is_dynamic());

auto outputs = network.execute();

ASSERT_EQ(outputs.size(), size_t(1));
ASSERT_EQ(outputs.begin()->first, "strided_slice");

auto output = outputs.at("strided_slice").get_memory();

ov::PartialShape expected_shape{1, 2, 2};

ASSERT_EQ(output->get_layout().get_partial_shape(), expected_shape);

std::vector<float> answers = {
4.0f, 5.0f, 6.0f, 7.0f
};

cldnn::mem_lock<float> output_ptr(output, get_test_stream());

for (size_t i = 0; i < answers.size(); ++i) {
ASSERT_EQ(answers[i], output_ptr[i]) << " i = " << i;
}
}

void test_2x2x2x1x1_2_negative_all_dynamic_begin() {
auto& engine = get_test_engine();
auto input = engine.allocate_memory({ ov::PartialShape{ 2, 2, 2 }, data_types::f32, format::bfyx });
Expand Down Expand Up @@ -1593,7 +1646,7 @@ class strided_slice_gpu_four_inputs: public ::testing::Test {
ASSERT_TRUE(are_equal(answers[i], output_ptr[i]));
}
}
};
};

class strided_slice_gpu_i8: public ::testing::Test {
public:
Expand Down Expand Up @@ -1822,6 +1875,10 @@ TEST_F(strided_slice_gpu, test_2x2x2x1x1_2_negative_all_dynamic_begin) {
this->test_2x2x2x1x1_2_negative_all_dynamic_begin();
}

TEST_F(strided_slice_gpu, test_2x2x2_all_dynamic_bcast) {
this->test_2x2x2_all_dynamic_bcast();
}

#ifdef RUN_ALL_MODEL_CACHING_TESTS
TEST_F(strided_slice_gpu, test_2x2x2x2_full_cached) {
this->test_2x2x2x2_full(true);
Expand Down