Add s8f32 kernels and dynamic QuantizationInfo

- Add support for QASYMM_SIGNED*QASYMM8_SIGNED->F32 in
  CpuGemmLowpMatrixMultiplyCore
- Add s8f32 kernel using existing s8->s32 kernels with a new
  DequantizeFloat OutputStage, the structure is similar to Requantize32
  but the opposite way around.
- Add SME s8f32 kernels with integrated support for DequantizeFloat.
- Add scale to CpuGemmLowpOffsetContributionKernel.
- Add virtual dequantize scale to gemm_common, only implemented for
  gemm_interleaved.
- Update year to 2024 in generate_build_files.
- Add dynamic flag to QuantizationInfo which signals to operators that
  it can change after configuration
- Add support for dynamic quantization in NEGEMMLowpMatrixMultiplyCore
- Add dynamic quantization fixture by extending
  GEMMLowpGenericMatrixMultiplyCoreValidationFixture
- Add GEMMLowpDequantizedMatrixMultiplyValidationFixture
- Store k (number of cols of A) rather than k_offset in the offset
  contribution kernels so that we can recompute it when the other
  offsets change

relates to: ONCPUML-1444 MLINFSW-439

Co-authored-by: Milos Puzovic <Milos.Puzovic@arm.com>
Co-authored-by: David Mansell <David.Mansell@arm.com>
Change-Id: I58a3acf2c09289a303e52eea6b336a696a5bc8da
Signed-off-by: Jonathan Deakin <jonathan.deakin@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/11022
Reviewed-by: Gunes Bayir <gunes.bayir@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>

1 files changed, 34 insertions, 0 deletions

diff --git a/tests/validation/NEON/GEMMLowp.cpp b/tests/validation/NEON/GEMMLowp.cpp
index 514d156983..1231c21ddb 100644
--- a/tests/validation/NEON/GEMMLowp.cpp
+++ b/tests/validation/NEON/GEMMLowp.cpp
@@ -53,6 +53,7 @@ namespace
 {
     constexpr AbsoluteTolerance<float> tolerance_batched(1);
     constexpr AbsoluteTolerance<float> tolerance_quant(1);
+    constexpr AbsoluteTolerance<float> tolerance_dequantized(0.01f);
 } // namespace
 
 
@@ -62,6 +63,11 @@ TEST_SUITE(MatrixMultiplyCore)
 
 using NEGEMMLowpMatrixMultiplyCoreFixture = GEMMLowpMatrixMultiplyCoreValidationFixture<Tensor, Accessor, NEGEMMLowpMatrixMultiplyCore>;
 using NEGEMMLowpMatrixMultiplyCoreAccumulateFixture = GEMMLowpMatrixMultiplyAccumulateValidationFixture<Tensor, Accessor, NEGEMMLowpMatrixMultiplyCore>;
+using NEGEMMLowpBatchedMatMulFixture      = GEMMLowpMatrixMultiplyCoreValidationFixture<Tensor, Accessor, NEGEMMLowpMatrixMultiplyCore, false, false, true>;
+using NEGEMMLowpMatrixMultiplyCoreDynamicQuantizationFixture = GEMMLowpMatrixMultiplyCoreDynamicQuantizationFixture<Tensor, Accessor, NEGEMMLowpMatrixMultiplyCore>;
+using NEGEMMLowpDequantizedMatrixMultiplyValidationFixture = GEMMLowpDequantizedMatrixMultiplyValidationFixture<Tensor, Accessor, NEGEMMLowpMatrixMultiplyCore>;
+
+using framework::dataset::make;
 
 DATA_TEST_CASE(Configuration, framework::DatasetMode::ALL, framework::dataset::concat(datasets::SmallGEMMLowpDataset(), datasets::LargeGEMMLowpDataset()),
                shape_a, shape_b, shape_c, a_offset, b_offset)
@@ -337,6 +343,34 @@ TEST_SUITE_END() // S32
 TEST_SUITE_END() // ACCUMULATION
 #endif // __arch64__
 
+TEST_SUITE(DynamicQuantization)
+FIXTURE_DATA_TEST_CASE(RunSmall, NEGEMMLowpMatrixMultiplyCoreDynamicQuantizationFixture, framework::DatasetMode::ALL, datasets::SmallGEMMLowpDataset())
+{
+    // Validate output
+    validate(Accessor(_target), _reference);
+}
+
+FIXTURE_DATA_TEST_CASE(RunLarge, NEGEMMLowpMatrixMultiplyCoreDynamicQuantizationFixture, framework::DatasetMode::NIGHTLY, datasets::LargeGEMMLowpDataset())
+{
+    // Validate output
+    validate(Accessor(_target), _reference);
+}
+TEST_SUITE_END() // DynamicQuantization
+
+TEST_SUITE(Dequant)
+FIXTURE_DATA_TEST_CASE(RunSmall, NEGEMMLowpDequantizedMatrixMultiplyValidationFixture, framework::DatasetMode::ALL, datasets::SmallGEMMLowpDataset())
+{
+    // Validate output
+    validate(Accessor(_target), _reference, tolerance_dequantized);
+}
+
+FIXTURE_DATA_TEST_CASE(RunLarge, NEGEMMLowpDequantizedMatrixMultiplyValidationFixture, framework::DatasetMode::NIGHTLY, datasets::LargeGEMMLowpDataset())
+{
+    // Validate output
+    validate(Accessor(_target), _reference, tolerance_dequantized);
+}
+TEST_SUITE_END() // Dequant
+
 TEST_SUITE_END() // MatrixMultiplyCore
 TEST_SUITE_END() // GEMMLowp
 TEST_SUITE_END() // NEON