Skip to content

[GPU] fix dGPU func testcases in smoke_MatMulCompressedWeights_extra_multiply #35442

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

Open
yuanxion wants to merge 2 commits into
base: master
Choose a base branch
from
+130 −1
Open

[GPU] fix dGPU func testcases in smoke_MatMulCompressedWeights_extra_multiply #35442

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
4814d4e
fix smoke_MatMulCompressedWeights_extra_multiply/MatmulWeightsDecompr…
yuanxion Apr 21, 2026
a0cef9b
refine testcases
yuanxion Apr 22, 2026
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
39 changes: 39 additions & 0 deletions src/plugins/intel_gpu/src/plugin/transformations/convert_matmul_to_fc.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -54,6 +54,45 @@ ConvertMatMulToFullyConnected::ConvertMatMulToFullyConnected(bool supports_immad
auto fc_input_a = pattern_map.at(activations_m);
auto fc_input_b = pattern_map.at(weights_m);

auto introduces_non_trivial_batch_broadcast = [](const ov::PartialShape& original_shape,
const ov::PartialShape& broadcasted_shape) {
if (!original_shape.rank().is_static() || !broadcasted_shape.rank().is_static()) {
return false;
}

const auto original_rank = static_cast<size_t>(original_shape.rank().get_length());
const auto broadcasted_rank = static_cast<size_t>(broadcasted_shape.rank().get_length());
if (broadcasted_rank < 2 || original_rank > broadcasted_rank) {
return false;
}

ov::PartialShape aligned_original_shape = original_shape;
for (size_t i = 0, cnt = broadcasted_rank - original_rank; i < cnt; ++i) {
aligned_original_shape.insert(aligned_original_shape.begin(), 1);
}

for (size_t i = 0; i < broadcasted_rank - 2; ++i) {
const auto& original_dim = aligned_original_shape[i];
const auto& broadcasted_dim = broadcasted_shape[i];
if (original_dim == 1 && broadcasted_dim.is_static() && broadcasted_dim.get_length() != 1) {
return true;
}
}

return false;
};

auto mul2_it = pattern_map.find(mul2_m);
if (mul2_it != pattern_map.end() && mul2_it->second.get_node_shared_ptr() == fc_input_b.get_node_shared_ptr()) {
const auto reshape_output = pattern_map.at(reshape_m);
// Keep valid 3D compressed FC cases enabled. Only reject the extra post-reshape multiply when broadcasting changes the weights
// from a shared matrix into data with real batch dimensions. For example, reshape may first squeeze the weights to [16, 32],
// then an extra multiply with scale [8, 1, 32] broadcasts them to [8, 16, 32], which makes the weights effectively batched again.
if (introduces_non_trivial_batch_broadcast(reshape_output.get_partial_shape(), fc_input_b.get_partial_shape())) {
return false;
}
}

// If 'fc_input_b' is shared with another matmul, transposing 'fc_input_b' is restricted.
// If it is connected to the 'input_a' of another matmul, do not transpose
// If it is connected to the 'input_b' of another matmul and the transpose option differs between the two matmuls, do not transpose.
Expand Down
26 changes: 25 additions & 1 deletion ...lugins/intel_gpu/tests/functional/subgraph_tests/dynamic/matmul_weights_decompression.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -411,6 +411,15 @@ const std::vector<ShapeParams> input_shapes_basic = {
{{{}, {{11, 339, 377}}}, {377, 335}}
};

const std::vector<ShapeParams> input_shapes_extra_multiply = {
{{{}, {{1, 4, 2}}}, {2, 32}, 2ul},
{{{}, {{1, 4, 16}}}, {1, 16, 32}},
};

const std::vector<ShapeParams> input_shapes_extra_multiply_non_trivial_batch_broadcast = {
{{{}, {{1, 4, 16}}}, {16, 32}, 2ul},
};

INSTANTIATE_TEST_SUITE_P(smoke_MatMulCompressedWeights_basic,
MatmulWeightsDecompression,
::testing::Combine(::testing::ValuesIn(input_shapes_basic),
Expand All @@ -428,7 +437,22 @@ INSTANTIATE_TEST_SUITE_P(smoke_MatMulCompressedWeights_basic,

INSTANTIATE_TEST_SUITE_P(smoke_MatMulCompressedWeights_extra_multiply,
MatmulWeightsDecompression,
::testing::Combine(::testing::ValuesIn(input_shapes_basic),
::testing::Combine(::testing::ValuesIn(input_shapes_extra_multiply),
::testing::ValuesIn(weights_precisions),
::testing::ValuesIn(activations_precisions),
::testing::Values(false),
::testing::Values(false),
::testing::Values(false),
::testing::Values(true),
::testing::Values(false),
::testing::ValuesIn(param_weights),
::testing::Values(0),
::testing::Values(1.0f)),
MatmulWeightsDecompression::get_test_case_name);

INSTANTIATE_TEST_SUITE_P(smoke_MatMulCompressedWeights_extra_multiply_non_trivial_batch_broadcast_no_convert,
MatmulWeightsDecompression,
Comment thread
yuanxion marked this conversation as resolved.
::testing::Combine(::testing::ValuesIn(input_shapes_extra_multiply_non_trivial_batch_broadcast),
::testing::ValuesIn(weights_precisions),
::testing::ValuesIn(activations_precisions),
::testing::Values(false),
Expand Down
66 changes: 66 additions & 0 deletions src/plugins/intel_gpu/tests/unit/transformations/convert_matmul_to_fc_test.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -474,6 +474,72 @@ TEST_F(TransformationTestsF, ConvertMatMulToFullyConnectedTest_compressed_u8_par
}
}

TEST_F(TransformationTestsF, ConvertMatMulToFullyConnectedTest_compressed_u8_weights_extra_multiply) {
{
auto data = std::make_shared<ov::opset1::Parameter>(ov::element::f16, ov::Shape{1, 4, 2});
auto weights = ov::opset1::Constant::create(ov::element::u8, ov::Shape{1, 2, 32}, {1});
auto convert = std::make_shared<ov::opset1::Convert>(weights, ov::element::f16);
auto mul_const = ov::opset1::Constant::create(ov::element::f16, ov::Shape{1, 1, 32}, {1});
auto mul = std::make_shared<ov::opset1::Multiply>(convert, mul_const);
auto reshape_const = ov::opset1::Constant::create(ov::element::i32, ov::Shape{2}, {2, 32});
auto reshape = std::make_shared<ov::opset1::Reshape>(mul, reshape_const, false);
auto extra_mul = std::make_shared<ov::opset1::Multiply>(reshape, mul_const);
auto matmul = std::make_shared<ov::opset1::MatMul>(data, extra_mul);

model = std::make_shared<ov::Model>(ov::OutputVector{matmul}, ov::ParameterVector{data});
bool support_immad = true;
manager.register_pass<ConvertMatMulToFullyConnected>(support_immad);
}
{
auto data = std::make_shared<ov::opset1::Parameter>(ov::element::f16, ov::Shape{1, 4, 2});
auto weights = ov::opset1::Constant::create(ov::element::u8, ov::Shape{1, 2, 32}, {1});
auto convert = std::make_shared<ov::opset1::Convert>(weights, ov::element::f16);
auto mul_const = ov::opset1::Constant::create(ov::element::f16, ov::Shape{1, 1, 32}, {1});
auto mul = std::make_shared<ov::opset1::Multiply>(convert, mul_const);
auto reshape_const = ov::opset1::Constant::create(ov::element::i32, ov::Shape{2}, {2, 32});
auto reshape = std::make_shared<ov::opset1::Reshape>(mul, reshape_const, false);
auto extra_mul = std::make_shared<ov::opset1::Multiply>(reshape, mul_const);

auto transpose_const = ov::opset1::Constant::create(ov::element::i32, {3}, {0, 2, 1});
auto transpose = std::make_shared<ov::opset1::Transpose>(extra_mul, transpose_const);
auto no_bias = std::make_shared<ov::intel_gpu::op::Placeholder>();
auto matmul = std::make_shared<op::FullyConnected>(data, transpose, no_bias);

model_ref = std::make_shared<ov::Model>(ov::OutputVector{matmul}, ov::ParameterVector{data});
}
}

TEST_F(TransformationTestsF, ConvertMatMulToFullyConnectedTest_compressed_u8_weights_extra_multiply_non_trivial_batch_broadcast) {
{
auto data = std::make_shared<ov::opset1::Parameter>(ov::element::f16, ov::Shape{1, 4, 16});
auto weights = ov::opset1::Constant::create(ov::element::u8, ov::Shape{8, 2, 32}, {1});
auto convert = std::make_shared<ov::opset1::Convert>(weights, ov::element::f16);
auto mul_const = ov::opset1::Constant::create(ov::element::f16, ov::Shape{8, 1, 32}, {1});
auto mul = std::make_shared<ov::opset1::Multiply>(convert, mul_const);
auto reshape_const = ov::opset1::Constant::create(ov::element::i32, ov::Shape{2}, {16, 32});
auto reshape = std::make_shared<ov::opset1::Reshape>(mul, reshape_const, false);
auto extra_mul = std::make_shared<ov::opset1::Multiply>(reshape, mul_const);
auto matmul = std::make_shared<ov::opset1::MatMul>(data, extra_mul);

model = std::make_shared<ov::Model>(ov::OutputVector{matmul}, ov::ParameterVector{data});
bool support_immad = true;
manager.register_pass<ConvertMatMulToFullyConnected>(support_immad);
}
{
auto data = std::make_shared<ov::opset1::Parameter>(ov::element::f16, ov::Shape{1, 4, 16});
auto weights = ov::opset1::Constant::create(ov::element::u8, ov::Shape{8, 2, 32}, {1});
auto convert = std::make_shared<ov::opset1::Convert>(weights, ov::element::f16);
auto mul_const = ov::opset1::Constant::create(ov::element::f16, ov::Shape{8, 1, 32}, {1});
auto mul = std::make_shared<ov::opset1::Multiply>(convert, mul_const);
auto reshape_const = ov::opset1::Constant::create(ov::element::i32, ov::Shape{2}, {16, 32});
auto reshape = std::make_shared<ov::opset1::Reshape>(mul, reshape_const, false);
auto extra_mul = std::make_shared<ov::opset1::Multiply>(reshape, mul_const);
auto matmul = std::make_shared<ov::opset1::MatMul>(data, extra_mul);

model_ref = std::make_shared<ov::Model>(ov::OutputVector{matmul}, ov::ParameterVector{data});
}
}

TEST_F(TransformationTestsF, ConvertMatMulToFullyConnectedTest_compressed_u4_weights_3D) {
{
auto data = std::make_shared<ov::opset1::Parameter>(ov::element::f16, ov::Shape{3, 2, 2});
Expand Down
Loading