diff options
author | ramelg01 <ramy.elgammal@arm.com> | 2022-04-08 03:52:28 +0100 |
---|---|---|
committer | Ramy Elgammal <ramy.elgammal@arm.com> | 2022-04-25 15:35:59 +0000 |
commit | c827e99fc46521f43719b0c2d1b6f05d66abf68c (patch) | |
tree | 31df1002673b2a4c4aae66608ad85b1ad6517050 /src/core/NEON/kernels/assembly/pool_common.hpp | |
parent | 0a3948394e7e77344201b8732e9c20fcb5fa9a38 (diff) | |
download | ComputeLibrary-c827e99fc46521f43719b0c2d1b6f05d66abf68c.tar.gz |
Update Neon™ pooling kernel
- Reduce duplication and simplify overall structure.
- Improve multi-threaded performance by sharing more data
in lower-level caches.
Partially Resolves: COMPMID-5054
Signed-off-by: Ramy Elgammal<ramy.elgammal@arm.com>
Change-Id: I5f4dc50913401d5c1cbfc10b866fae9490cbc4d7
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/7404
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Andrew Mundy
Reviewed-by: Sheri Zhang <sheri.zhang@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/core/NEON/kernels/assembly/pool_common.hpp')
-rw-r--r-- | src/core/NEON/kernels/assembly/pool_common.hpp | 22 |
1 files changed, 10 insertions, 12 deletions
diff --git a/src/core/NEON/kernels/assembly/pool_common.hpp b/src/core/NEON/kernels/assembly/pool_common.hpp index b6a0a0abed..599e18ac59 100644 --- a/src/core/NEON/kernels/assembly/pool_common.hpp +++ b/src/core/NEON/kernels/assembly/pool_common.hpp @@ -1,5 +1,5 @@ /* - * Copyright (c) 2021 Arm Limited. + * Copyright (c) 2021-2022 Arm Limited. * * SPDX-License-Identifier: MIT * @@ -23,8 +23,9 @@ */ #pragma once - -#include "common.hpp" +#ifdef CYCLE_PROFILING +#include "profiler.hpp" +#endif namespace arm_conv { @@ -53,6 +54,11 @@ struct PoolingStride unsigned int rows, cols; }; +struct PaddingValues +{ + unsigned int left, top, right, bottom; +}; + class IPoolingCommon { public: @@ -60,6 +66,7 @@ public: // Determine the amount of working space required. virtual size_t get_working_size(unsigned int num_threads) const = 0; + virtual size_t get_working_size(unsigned int num_threads, unsigned int n_channels) const = 0; // Execute pooling over the specified area of memory. virtual void execute( @@ -103,14 +110,5 @@ public: unsigned int num_threads) const = 0; }; -struct Nothing -{ -}; - -template <typename TInput, typename TOutput, class OutputStage = Nothing> -class PoolingCommon : public IPoolingCommon -{ -}; - } // namespace pooling } // namespace arm_conv |