aboutsummaryrefslogtreecommitdiff
path: root/src/core/NEON/kernels/arm_gemm/indirect-interleaves/a64_interleave4_block16_s8_s8_summing.hpp
diff options
context:
space:
mode:
Diffstat (limited to 'src/core/NEON/kernels/arm_gemm/indirect-interleaves/a64_interleave4_block16_s8_s8_summing.hpp')
-rw-r--r--src/core/NEON/kernels/arm_gemm/indirect-interleaves/a64_interleave4_block16_s8_s8_summing.hpp4
1 files changed, 2 insertions, 2 deletions
diff --git a/src/core/NEON/kernels/arm_gemm/indirect-interleaves/a64_interleave4_block16_s8_s8_summing.hpp b/src/core/NEON/kernels/arm_gemm/indirect-interleaves/a64_interleave4_block16_s8_s8_summing.hpp
index 56ca49a36e..23800edf20 100644
--- a/src/core/NEON/kernels/arm_gemm/indirect-interleaves/a64_interleave4_block16_s8_s8_summing.hpp
+++ b/src/core/NEON/kernels/arm_gemm/indirect-interleaves/a64_interleave4_block16_s8_s8_summing.hpp
@@ -210,8 +210,8 @@ void interleave_block<4, 16, VLType::None, true>(
"sadalp v22.4s, v26.8h\n"
"sadalp 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"