diff options
author | Georgios Pinitas <georgios.pinitas@arm.com> | 2020-06-02 23:00:41 +0100 |
---|---|---|
committer | Georgios Pinitas <georgios.pinitas@arm.com> | 2020-06-08 16:42:03 +0000 |
commit | 25ef7217ec4e13682bf37c87c0c6075a799ba1c0 (patch) | |
tree | 8ed3e4d7b811a36322560298ab6c8a0484d10ab8 /arm_compute/core/NEON/wrapper/intrinsics | |
parent | ebfb2f8701c131294b3c5c1e36547fa3658d09dd (diff) | |
download | ComputeLibrary-25ef7217ec4e13682bf37c87c0c6075a799ba1c0.tar.gz |
COMPMID-3180: Remove padding from NEThreshold
- Removes padding from NEThresholdKernel
- Alters configuration interface to use a descriptor
Change-Id: I394d5e1375454813856d9d206e61dc9a87c2cadc
Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3300
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'arm_compute/core/NEON/wrapper/intrinsics')
-rw-r--r-- | arm_compute/core/NEON/wrapper/intrinsics/cle.h | 64 | ||||
-rw-r--r-- | arm_compute/core/NEON/wrapper/intrinsics/intrinsics.h | 1 |
2 files changed, 65 insertions, 0 deletions
diff --git a/arm_compute/core/NEON/wrapper/intrinsics/cle.h b/arm_compute/core/NEON/wrapper/intrinsics/cle.h new file mode 100644 index 0000000000..83c92d6891 --- /dev/null +++ b/arm_compute/core/NEON/wrapper/intrinsics/cle.h @@ -0,0 +1,64 @@ +/* + * Copyright (c) 2020 ARM Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef ARM_COMPUTE_WRAPPER_CLE_H +#define ARM_COMPUTE_WRAPPER_CLE_H + +#include <arm_neon.h> + +namespace arm_compute +{ +namespace wrapper +{ +#define VCLE_IMPL(stype, vtype, rtype, prefix, postfix) \ + inline rtype vcle(const vtype &a, const vtype &b) \ + { \ + return prefix##_##postfix(a, b); \ + } + +VCLE_IMPL(uint8_t, uint8x8_t, uint8x8_t, vcle, u8) +VCLE_IMPL(int8_t, int8x8_t, uint8x8_t, vcle, s8) +VCLE_IMPL(uint16_t, uint16x4_t, uint16x4_t, vcle, u16) +VCLE_IMPL(int16_t, int16x4_t, uint16x4_t, vcle, s16) +VCLE_IMPL(uint32_t, uint32x2_t, uint32x2_t, vcle, u32) +VCLE_IMPL(int32_t, int32x2_t, uint32x2_t, vcle, s32) +VCLE_IMPL(float32x2_t, float32x2_t, uint32x2_t, vcle, f32) +#ifdef __ARM_FEATURE_FP16_VECTOR_ARITHMETIC +VCLE_IMPL(float16x4_t, float16x4_t, uint16x4_t, vcle, f16) +#endif // __ARM_FEATURE_FP16_VECTOR_ARITHMETIC + +VCLE_IMPL(uint8_t, uint8x16_t, uint8x16_t, vcleq, u8) +VCLE_IMPL(int8_t, int8x16_t, uint8x16_t, vcleq, s8) +VCLE_IMPL(uint16_t, uint16x8_t, uint16x8_t, vcleq, u16) +VCLE_IMPL(int16_t, int16x8_t, uint16x8_t, vcleq, s16) +VCLE_IMPL(uint32_t, uint32x4_t, uint32x4_t, vcleq, u32) +VCLE_IMPL(int32_t, int32x4_t, uint32x4_t, vcleq, s32) +VCLE_IMPL(float32x4_t, float32x4_t, uint32x4_t, vcleq, f32) +#ifdef __ARM_FEATURE_FP16_VECTOR_ARITHMETIC +VCLE_IMPL(float16x8_t, float16x8_t, uint16x8_t, vcleq, f16) +#endif // __ARM_FEATURE_FP16_VECTOR_ARITHMETIC + +#undef VCLE_IMPL +} // namespace wrapper +} // namespace arm_compute +#endif /* ARM_COMPUTE_WRAPPER_CLE_H */ diff --git a/arm_compute/core/NEON/wrapper/intrinsics/intrinsics.h b/arm_compute/core/NEON/wrapper/intrinsics/intrinsics.h index 1150daa073..14c5d615be 100644 --- a/arm_compute/core/NEON/wrapper/intrinsics/intrinsics.h +++ b/arm_compute/core/NEON/wrapper/intrinsics/intrinsics.h @@ -31,6 +31,7 @@ #include "arm_compute/core/NEON/wrapper/intrinsics/ceq.h" #include "arm_compute/core/NEON/wrapper/intrinsics/cge.h" #include "arm_compute/core/NEON/wrapper/intrinsics/cgt.h" +#include "arm_compute/core/NEON/wrapper/intrinsics/cle.h" #include "arm_compute/core/NEON/wrapper/intrinsics/clt.h" #include "arm_compute/core/NEON/wrapper/intrinsics/combine.h" #include "arm_compute/core/NEON/wrapper/intrinsics/cvt.h" |