diff options
| author | Omar Al Khatib <omar.alkhatib@arm.com> | 2022-11-01 17:01:24 +0000 |
|---|---|---|
| committer | Omar Al Khatib <omar.alkhatib@arm.com> | 2022-11-07 15:45:21 +0000 |
| commit | 605a928960c0c36384925a9064d12addc8f43a41 (patch) | |
| tree | 71896f7d4683751c7c87c5606b95e0c9c421384a /src/core/NEON/wrapper/intrinsics/store.h | |
| parent | a2b131bf4680f83b10a7e3544b6183279d8c2691 (diff) | |
| download | ComputeLibrary-605a928960c0c36384925a9064d12addc8f43a41.tar.gz | |
Optimize CPU mul layer on quantized data
Resolves : [COMPMID-5461]
Signed-off-by: Omar Al Khatib <omar.alkhatib@arm.com>
Change-Id: I89b99d267c32b00ef44f9bb6e7c714dfe4a0d29d
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8420
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Gunes Bayir <gunes.bayir@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/core/NEON/wrapper/intrinsics/store.h')
| -rw-r--r-- | src/core/NEON/wrapper/intrinsics/store.h | 6 |
1 files changed, 1 insertions, 5 deletions
diff --git a/src/core/NEON/wrapper/intrinsics/store.h b/src/core/NEON/wrapper/intrinsics/store.h index 6dda432ea9..ce1b9a554e 100644 --- a/src/core/NEON/wrapper/intrinsics/store.h +++ b/src/core/NEON/wrapper/intrinsics/store.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2018-2020 Arm Limited. + * Copyright (c) 2018-2020, 2022 Arm Limited. * * SPDX-License-Identifier: MIT * @@ -44,8 +44,6 @@ VSTORE_IMPL(uint16_t, uint16x4_t, vst1, u16) VSTORE_IMPL(int16_t, int16x4_t, vst1, s16) VSTORE_IMPL(uint32_t, uint32x2_t, vst1, u32) VSTORE_IMPL(int32_t, int32x2_t, vst1, s32) -//VSTORE_IMPL(uint64_t, 1, vst1, u64) -//VSTORE_IMPL(int64_t, 1, vst1, s64) VSTORE_IMPL(float, float32x2_t, vst1, f32) #ifdef __ARM_FEATURE_FP16_VECTOR_ARITHMETIC VSTORE_IMPL(float16_t, float16x4_t, vst1, f16) @@ -57,8 +55,6 @@ VSTORE_IMPL(uint16_t, uint16x8_t, vst1q, u16) VSTORE_IMPL(int16_t, int16x8_t, vst1q, s16) VSTORE_IMPL(uint32_t, uint32x4_t, vst1q, u32) VSTORE_IMPL(int32_t, int32x4_t, vst1q, s32) -//VSTORE_IMPL(uint64_t, 2, vst1q, u64) -//VSTORE_IMPL(int64_t, 2, vst1q, s64) VSTORE_IMPL(float, float32x4_t, vst1q, f32) #ifdef __ARM_FEATURE_FP16_VECTOR_ARITHMETIC VSTORE_IMPL(float16_t, float16x8_t, vst1q, f16) |