From 74921eee924625426429044decefe3673561b174 Mon Sep 17 00:00:00 2001 From: Michael Tyler Date: Wed, 12 Apr 2023 17:43:17 +0100 Subject: Update CPU kernel implementations and guard directives Resolves COMPMID-6023 Change-Id: I868975d14c4f98af6716726feda22405a6a4c891 Signed-off-by: Michael Tyler Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9686 Tested-by: Arm Jenkins Reviewed-by: Viet-Hoa Do Comments-Addressed: Arm Jenkins Benchmark: Arm Jenkins --- .../indirect-interleaves/a64_interleave4_block16_u8_u8_summing.hpp | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'src/core/NEON/kernels/arm_gemm/indirect-interleaves/a64_interleave4_block16_u8_u8_summing.hpp') diff --git a/src/core/NEON/kernels/arm_gemm/indirect-interleaves/a64_interleave4_block16_u8_u8_summing.hpp b/src/core/NEON/kernels/arm_gemm/indirect-interleaves/a64_interleave4_block16_u8_u8_summing.hpp index 4c7bb71fb2..15545c24db 100644 --- a/src/core/NEON/kernels/arm_gemm/indirect-interleaves/a64_interleave4_block16_u8_u8_summing.hpp +++ b/src/core/NEON/kernels/arm_gemm/indirect-interleaves/a64_interleave4_block16_u8_u8_summing.hpp @@ -210,8 +210,8 @@ void interleave_block<4, 16, VLType::None, true>( "uadalp v22.4s, v26.8h\n" "uadalp v21.4s, v25.8h\n" "addp v24.4s, v24.4s, v23.4s\n" - "addp v23.4s, v22.4s, v21.4s\n" - "addp v24.4s, v24.4s, v23.4s\n" + "addp v16.4s, v22.4s, v21.4s\n" + "addp v24.4s, v24.4s, v16.4s\n" "add v24.4s, v24.4s, v20.4s\n" "str q24, [%x[out_ptr], #0x0]\n" "add %x[out_ptr], %x[out_ptr], #0x10\n" -- cgit v1.2.1