From 146138378c1587b7297d245b7177641315f6180b Mon Sep 17 00:00:00 2001 From: Georgios Pinitas Date: Fri, 1 Mar 2019 19:07:11 +0000 Subject: COMPMID-1995: Update RSH GEMM assembly kernels. -Updates u8/s8 hybrid dot product kernels to work for any N and any K >=16. -Adds hybrid FP32 kernels with generic and A55 variants. -Adds SVE native kernels for fp16/u8/s8. Change-Id: Ifc0eaba9e3c8ea5bb19d334e870e1b39e4e7e728 Signed-off-by: Georgios Pinitas Reviewed-on: https://review.mlplatform.org/c/863 Tested-by: Arm Jenkins Reviewed-by: Michele Di Giorgio Comments-Addressed: Arm Jenkins Reviewed-by: Michalis Spyrou --- src/core/NEON/kernels/arm_gemm/gemv_batched.hpp | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'src/core/NEON/kernels/arm_gemm/gemv_batched.hpp') diff --git a/src/core/NEON/kernels/arm_gemm/gemv_batched.hpp b/src/core/NEON/kernels/arm_gemm/gemv_batched.hpp index 32d668f66d..b7f9de85c4 100644 --- a/src/core/NEON/kernels/arm_gemm/gemv_batched.hpp +++ b/src/core/NEON/kernels/arm_gemm/gemv_batched.hpp @@ -44,10 +44,9 @@ public: _subgemm = gemm(newargs); } - using GemmCommon::set_arrays; void set_arrays(const To *A, const int lda, const int A_batch_stride, const int A_multi_stride, const To *B, const int ldb, const int B_multi_stride, - Tr *C, const int ldc, const int C_batch_stride, const int C_multi_stride) override { + Tr *C, const int ldc, const int C_batch_stride, const int C_multi_stride) override { /* A and C's batch stride becomes their new row stride. New batch stride is 0 as nbatches for subgemm is always 1. */ _subgemm->set_arrays(A, A_batch_stride, 0, A_multi_stride, B, ldb, B_multi_stride, @@ -86,7 +85,6 @@ public: return _subgemm->get_B_pretransposed_array_size(); } - using GemmCommon::pretranspose_B_array; void pretranspose_B_array(void *buffer, const To *B, const int ldb, const int B_multi_stride) override { _subgemm->pretranspose_B_array(buffer, B, ldb, B_multi_stride); } -- cgit v1.2.1