commit 2217f1e60964fe586cae7ef996af7ef1c0bef2ab
author Pablo Marquez Tello <pablo.tello@arm.com> Tue May 14 07:54:19 2024 +0100
committer Pablo Marquez Tello <pablo.tello@arm.com> Tue May 14 11:58:27 2024 +0000
tree 593dad484fe922efeb655335d37b0066d2af4bcd
parent 21fb2ad16a30a5ff29929515abe28c14b2c6b5a1

Refactor arm_gemm to enable FP16 in all multi_isa builds

* Resolves MLCE-1285

Change-Id: I22a37972aefe1c0f04accbc798baa18358ed8959
Signed-off-by: Pablo Marquez Tello <pablo.tello@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/11552
Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>

src/core/NEON/kernels/arm_gemm/interleave_indirect.cpp

diff --git a/src/core/NEON/kernels/arm_gemm/interleave_indirect.cpp b/src/core/NEON/kernels/arm_gemm/interleave_indirect.cpp
index 5959193..7c09608 100644
--- a/src/core/NEON/kernels/arm_gemm/interleave_indirect.cpp
+++ b/src/core/NEON/kernels/arm_gemm/interleave_indirect.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2020-2022 Arm Limited.
+ * Copyright (c) 2020-2022, 2024 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -330,11 +330,11 @@
 #endif // ARM_COMPUTE_ENABLE_SVE && ARM_COMPUTE_ENABLE_SVEF32MM
 
 /* FP16 */
-#if defined(FP16_KERNELS) || defined(__ARM_FEATURE_FP16_VECTOR_ARITHMETIC)
+#if defined(FP16_KERNELS) || defined(ARM_COMPUTE_ENABLE_FP16)
 template void IndirectInterleave<8, 1, VLType::None>(__fp16 *, const __fp16 * const * const *, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 template void ConvolutionInterleave<8, 1, VLType::None>(__fp16 *, const __fp16 *, size_t, const convolver<__fp16> &, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 template void Interleave<8, 1, VLType::None>(__fp16 *, const __fp16 *, size_t, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
-#endif // FP16_KERNELS ar __ARM_FEATURE_FP16_VECTOR_ARITHMETIC
+#endif // FP16_KERNELS ar ARM_COMPUTE_ENABLE_FP16
 
 template void IndirectInterleave<8, 1, VLType::None>(float *, const __fp16 * const * const *, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);
 template void ConvolutionInterleave<8, 1, VLType::None>(float *, const __fp16 *, size_t, const convolver<__fp16> &, unsigned int, unsigned int, unsigned int, unsigned int, unsigned int, bool, int32_t);