From f1f1f87132690a8061801ef1a4638d637c780df7 Mon Sep 17 00:00:00 2001
From: Radu Salavat <radu.salavat@arm.com>
Date: Tue, 27 Feb 2024 18:32:26 +0000
Subject: Add in place summation to CPU GEMM kernels

Instead of dispatching the sum postop for GEMM kernels to a
separate kernel + add, that requires an extra destination sized
allocation, plus 3 extra load/stores per element,
just do it in the GEMM kernel.

Resolves: ONCPUML-1442

Signed-off-by: Radu Salavat <radu.salavat@arm.com>
Co-authored-by: Milos Puzovic <milos.puzovic@arm.com>
Change-Id: I7a1f2da3300875fa1ac88b705a34390969518077
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/11298
Reviewed-by: Gunes Bayir <gunes.bayir@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>
---
 tests/validation/reference/GEMMLowp.h | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

(limited to 'tests/validation/reference/GEMMLowp.h')

diff --git a/tests/validation/reference/GEMMLowp.h b/tests/validation/reference/GEMMLowp.h
index 99015d71fb..6e471fdad1 100644
--- a/tests/validation/reference/GEMMLowp.h
+++ b/tests/validation/reference/GEMMLowp.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2020 Arm Limited.
+ * Copyright (c) 2017-2020, 2024 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -21,8 +21,8 @@
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  * SOFTWARE.
  */
-#ifndef ARM_COMPUTE_TEST_GEMMLOWP_H
-#define ARM_COMPUTE_TEST_GEMMLOWP_H
+#ifndef ACL_TESTS_VALIDATION_REFERENCE_GEMMLOWP_H
+#define ACL_TESTS_VALIDATION_REFERENCE_GEMMLOWP_H
 
 #include "tests/SimpleTensor.h"
 #include "tests/validation/Helpers.h"
@@ -38,6 +38,9 @@ namespace reference
 template <typename T1, typename T2, typename T3>
 SimpleTensor<T1> gemmlowp_matrix_multiply_core(const SimpleTensor<T2> &a, const SimpleTensor<T3> &b, TensorShape shape_c, int32_t a_offset, int32_t b_offset);
 
+template <typename T1, typename T2, typename T3>
+void gemmlowp_matrix_multiply_core_accumulate(const SimpleTensor<T2> &a, const SimpleTensor<T3> &b, TensorShape shape_c, int32_t a_offset, int32_t b_offset, SimpleTensor<T1> &dst_);
+
 template <typename T1, typename T2, typename T3 = T2>
 SimpleTensor<T1> gemmlowp(const SimpleTensor<T2> &a, const SimpleTensor<T3> &b, TensorShape shape_c);
 
@@ -71,4 +74,4 @@ SimpleTensor<TOut> gemmlowp_quantize_down_scale_by_float(const SimpleTensor<TIn>
 } // namespace validation
 } // namespace test
 } // namespace arm_compute
-#endif /* ARM_COMPUTE_TEST_GEMMLOWP_H */
+#endif // ACL_TESTS_VALIDATION_REFERENCE_GEMMLOWP_H
-- 
cgit v1.2.1