diff options
Diffstat (limited to 'src/core/NEON/kernels/convolution/winograd/output_transforms')
8 files changed, 46 insertions, 46 deletions
diff --git a/src/core/NEON/kernels/convolution/winograd/output_transforms/a64_fp16_4x4_3x3.cpp b/src/core/NEON/kernels/convolution/winograd/output_transforms/a64_fp16_4x4_3x3.cpp index 8a2837a125..295005a2ee 100644 --- a/src/core/NEON/kernels/convolution/winograd/output_transforms/a64_fp16_4x4_3x3.cpp +++ b/src/core/NEON/kernels/convolution/winograd/output_transforms/a64_fp16_4x4_3x3.cpp @@ -1,5 +1,5 @@ /* - * Copyright (c) 2022 Arm Limited. + * Copyright (c) 2022, 2024 Arm Limited. * * SPDX-License-Identifier: MIT * @@ -34,13 +34,13 @@ namespace output_transform { void a64_fp16_4x4_3x3( unsigned int n_channels, const __fp16* inptr, - const size_t matrix_stride, + size_t matrix_stride, const __fp16* bptr, __fp16* const output, - const size_t output_row_stride, - const size_t output_col_stride, - const __fp16 output_min, - const __fp16 output_max + size_t output_row_stride, + size_t output_col_stride, + __fp16 output_min, + __fp16 output_max ) { constexpr int output_tile_rows = 4, output_tile_cols = 4; diff --git a/src/core/NEON/kernels/convolution/winograd/output_transforms/arm_fp32_1x2_1x7.cpp b/src/core/NEON/kernels/convolution/winograd/output_transforms/arm_fp32_1x2_1x7.cpp index cce3745c77..8c6cf9725e 100644 --- a/src/core/NEON/kernels/convolution/winograd/output_transforms/arm_fp32_1x2_1x7.cpp +++ b/src/core/NEON/kernels/convolution/winograd/output_transforms/arm_fp32_1x2_1x7.cpp @@ -1,5 +1,5 @@ /* - * Copyright (c) 2022-2023 Arm Limited. + * Copyright (c) 2022-2024 Arm Limited. * * SPDX-License-Identifier: MIT * @@ -33,13 +33,13 @@ namespace output_transform { void arm_fp32_1x2_1x7( unsigned int n_channels, const float* inptr, - const size_t matrix_stride, + size_t matrix_stride, const float* bptr, float *outptr, size_t, // No need to stride across rows - const size_t output_col_stride, - const float output_min, - const float output_max + size_t output_col_stride, + float output_min, + float output_max ) { constexpr auto inner_tile_cols = 8u, output_tile_cols = 2u; diff --git a/src/core/NEON/kernels/convolution/winograd/output_transforms/arm_fp32_1x4_1x5.cpp b/src/core/NEON/kernels/convolution/winograd/output_transforms/arm_fp32_1x4_1x5.cpp index 10f25a4aab..ac05f23221 100644 --- a/src/core/NEON/kernels/convolution/winograd/output_transforms/arm_fp32_1x4_1x5.cpp +++ b/src/core/NEON/kernels/convolution/winograd/output_transforms/arm_fp32_1x4_1x5.cpp @@ -1,5 +1,5 @@ /* - * Copyright (c) 2022-2023 Arm Limited. + * Copyright (c) 2022-2024 Arm Limited. * * SPDX-License-Identifier: MIT * @@ -33,13 +33,13 @@ namespace output_transform { void arm_fp32_1x4_1x5( unsigned int n_channels, const float* inptr, - const size_t matrix_stride, + size_t matrix_stride, const float* bptr, float *outptr, size_t, // No need to stride across rows - const size_t output_col_stride, - const float output_min, - const float output_max + size_t output_col_stride, + float output_min, + float output_max ) { constexpr auto inner_tile_cols = 8u, output_tile_cols = 4u; diff --git a/src/core/NEON/kernels/convolution/winograd/output_transforms/arm_fp32_1x6_1x3.cpp b/src/core/NEON/kernels/convolution/winograd/output_transforms/arm_fp32_1x6_1x3.cpp index c45ff8cf2c..154dc6fe1a 100644 --- a/src/core/NEON/kernels/convolution/winograd/output_transforms/arm_fp32_1x6_1x3.cpp +++ b/src/core/NEON/kernels/convolution/winograd/output_transforms/arm_fp32_1x6_1x3.cpp @@ -1,5 +1,5 @@ /* - * Copyright (c) 2022-2023 Arm Limited. + * Copyright (c) 2022-2024 Arm Limited. * * SPDX-License-Identifier: MIT * @@ -34,13 +34,13 @@ namespace output_transform { void arm_fp32_1x6_1x3( unsigned int n_channels, const float* inptr, - const size_t matrix_stride, + size_t matrix_stride, const float* bptr, float *outptr, size_t, // No need to stride across rows - const size_t output_col_stride, - const float output_min, - const float output_max + size_t output_col_stride, + float output_min, + float output_max ) { constexpr unsigned int inner_tile_cols = 8, output_tile_cols = 6; diff --git a/src/core/NEON/kernels/convolution/winograd/output_transforms/arm_fp32_2x2_3x3.cpp b/src/core/NEON/kernels/convolution/winograd/output_transforms/arm_fp32_2x2_3x3.cpp index 7f6b838ed8..28f042bcbf 100644 --- a/src/core/NEON/kernels/convolution/winograd/output_transforms/arm_fp32_2x2_3x3.cpp +++ b/src/core/NEON/kernels/convolution/winograd/output_transforms/arm_fp32_2x2_3x3.cpp @@ -1,5 +1,5 @@ /* - * Copyright (c) 2022 Arm Limited. + * Copyright (c) 2022, 2024 Arm Limited. * * SPDX-License-Identifier: MIT * @@ -33,13 +33,13 @@ namespace output_transform { void arm_fp32_2x2_3x3( unsigned int n_channels, const float* inptr, - const size_t matrix_stride, + size_t matrix_stride, const float* bptr, float *outptr, - const size_t output_row_stride, - const size_t output_col_stride, - const float output_min, - const float output_max + size_t output_row_stride, + size_t output_col_stride, + float output_min, + float output_max ) { constexpr auto output_tile_rows = 2u, output_tile_cols = 2u; diff --git a/src/core/NEON/kernels/convolution/winograd/output_transforms/arm_fp32_2x2_5x5.cpp b/src/core/NEON/kernels/convolution/winograd/output_transforms/arm_fp32_2x2_5x5.cpp index 0dc4851084..8e5ba74ac3 100644 --- a/src/core/NEON/kernels/convolution/winograd/output_transforms/arm_fp32_2x2_5x5.cpp +++ b/src/core/NEON/kernels/convolution/winograd/output_transforms/arm_fp32_2x2_5x5.cpp @@ -1,5 +1,5 @@ /* - * Copyright (c) 2022 Arm Limited. + * Copyright (c) 2022, 2024 Arm Limited. * * SPDX-License-Identifier: MIT * @@ -33,13 +33,13 @@ namespace output_transform { void arm_fp32_2x2_5x5( unsigned int n_channels, const float* inptr, - const size_t matrix_stride, + size_t matrix_stride, const float* bptr, float *outptr, - const size_t output_row_stride, - const size_t output_col_stride, - const float output_min, - const float output_max + size_t output_row_stride, + size_t output_col_stride, + float output_min, + float output_max ) { constexpr auto output_tile_rows = 2u, output_tile_cols = 2u; diff --git a/src/core/NEON/kernels/convolution/winograd/output_transforms/arm_fp32_4x4_3x3.cpp b/src/core/NEON/kernels/convolution/winograd/output_transforms/arm_fp32_4x4_3x3.cpp index a26fbfbf5a..72c43019fa 100644 --- a/src/core/NEON/kernels/convolution/winograd/output_transforms/arm_fp32_4x4_3x3.cpp +++ b/src/core/NEON/kernels/convolution/winograd/output_transforms/arm_fp32_4x4_3x3.cpp @@ -1,5 +1,5 @@ /* - * Copyright (c) 2022 Arm Limited. + * Copyright (c) 2022, 2024 Arm Limited. * * SPDX-License-Identifier: MIT * @@ -33,13 +33,13 @@ namespace output_transform { void arm_fp32_4x4_3x3( unsigned int n_channels, const float* inptr, - const size_t matrix_stride, + size_t matrix_stride, const float* bptr, float *outptr, - const size_t output_row_stride, - const size_t output_col_stride, - const float output_min, - const float output_max + size_t output_row_stride, + size_t output_col_stride, + float output_min, + float output_max ) { constexpr auto output_tile_rows = 4u, output_tile_cols = 4u; diff --git a/src/core/NEON/kernels/convolution/winograd/output_transforms/sme_fp32_mopa_4x4_3x3.cpp b/src/core/NEON/kernels/convolution/winograd/output_transforms/sme_fp32_mopa_4x4_3x3.cpp index 55ed24cd74..043914d590 100644 --- a/src/core/NEON/kernels/convolution/winograd/output_transforms/sme_fp32_mopa_4x4_3x3.cpp +++ b/src/core/NEON/kernels/convolution/winograd/output_transforms/sme_fp32_mopa_4x4_3x3.cpp @@ -1,5 +1,5 @@ /* - * Copyright (c) 2022-2023 Arm Limited. + * Copyright (c) 2022-2024 Arm Limited. * * SPDX-License-Identifier: MIT * @@ -31,15 +31,15 @@ namespace winograd { namespace output_transform { void sme_fp32_mopa_4x4_3x3( - const unsigned int n_channels, + unsigned int n_channels, const float* inptr, - const size_t matrix_stride, + size_t matrix_stride, const float* bptr, float* const output, - const size_t output_row_stride, - const size_t output_col_stride, - const float output_min, - const float output_max + size_t output_row_stride, + size_t output_col_stride, + float output_min, + float output_max ) { // The below assembler uses the Kronecker product and the "vec trick" to |