diff options
author | Georgios Pinitas <georgios.pinitas@arm.com> | 2020-11-02 01:37:17 +0000 |
---|---|---|
committer | Georgios Pinitas <georgios.pinitas@arm.com> | 2020-11-12 15:59:25 +0000 |
commit | c0b6f76561580414f08633a804fc548ccad65659 (patch) | |
tree | 4d46b7f479de04f799e29095392948aeb370c029 /src/core/NEON/kernels/arm_gemm/transforms/a64_transpose_interleave_24way_16bit.hpp | |
parent | 824061d9910ebb42cbe46b677c0b843db212c9a2 (diff) | |
download | ComputeLibrary-c0b6f76561580414f08633a804fc548ccad65659.tar.gz |
COMPMID-3776: Indirect GEMM
Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com>
Change-Id: I51a1b0f098bc3a8c408c50c92221e4df3061e12c
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4343
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Sang-Hoon Park <sang-hoon.park@arm.com>
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/core/NEON/kernels/arm_gemm/transforms/a64_transpose_interleave_24way_16bit.hpp')
-rw-r--r-- | src/core/NEON/kernels/arm_gemm/transforms/a64_transpose_interleave_24way_16bit.hpp | 8 |
1 files changed, 4 insertions, 4 deletions
diff --git a/src/core/NEON/kernels/arm_gemm/transforms/a64_transpose_interleave_24way_16bit.hpp b/src/core/NEON/kernels/arm_gemm/transforms/a64_transpose_interleave_24way_16bit.hpp index a137f9360a..bcbe2b84d8 100644 --- a/src/core/NEON/kernels/arm_gemm/transforms/a64_transpose_interleave_24way_16bit.hpp +++ b/src/core/NEON/kernels/arm_gemm/transforms/a64_transpose_interleave_24way_16bit.hpp @@ -30,12 +30,12 @@ // Generic unblocked transposed 12x32-bit sized specialisation template <> template <typename T> -inline void TransformImpl<12, 1, true, 4, 4, false>::Transform( +inline void TransformImpl<12, 1, true, 4, 4, VLType::None>::Transform( T* out, const T* const in, const int stride, const int x0, const int xmax, const int k0, const int kmax ) { // Redirect to a 24 x uint16_t specialisation - TransformImpl<24, 1, true, 2, 2, false>::Transform( + TransformImpl<24, 1, true, 2, 2, VLType::None>::Transform( reinterpret_cast<uint16_t *>(out), reinterpret_cast<const uint16_t * const>(in), stride*2, x0*2, xmax*2, k0, kmax @@ -45,7 +45,7 @@ inline void TransformImpl<12, 1, true, 4, 4, false>::Transform( // Generic 24x16-bit sized specialisation template <> template <typename T> -inline void TransformImpl<24, 1, true, 2, 2, false>::Transform( +inline void TransformImpl<24, 1, true, 2, 2, VLType::None>::Transform( T* out, const T* const in, const int stride, const int x0, const int xmax, const int k0, const int kmax ) { @@ -120,7 +120,7 @@ inline void TransposeInterleaveCommon<24, uint16_t, uint16_t>::moveblock_1x4(con template <> template <> -inline void TransformImpl<24, 1, true, 2, 2, false>::Transform( +inline void TransformImpl<24, 1, true, 2, 2, VLType::None>::Transform( uint16_t* out, const uint16_t* const in, const int stride, const int x0, const int xmax, const int k0, const int kmax ) { |