commit 101271e1bbbf24c1a2e392426fade873f228999b
author Frank Barchard <fbarchard@google.com> Wed Feb 02 01:49:54 2022 -0800
committer XNNPACK Team <xnnpack-github-robot@google.com> Wed Feb 02 01:51:05 2022 -0800
tree c76accc771f3b71da6f49d663a93acfbcb06f777
parent f82410d541654ae496ddb2de04c3e7ab16cfe0fb

QC8 4x8 lane GEMM AArch32 microkernel for Cortex A7

- Template for Neon (armv7) FP32 quantization for Cortex A7
- Enable Neon (armv7) microkernel for Cortex A7
- Enable Non-Prefetch microkernel for Cortex A55r0

PiperOrigin-RevId: 425824335

CMakeLists.txt

diff --git a/CMakeLists.txt b/CMakeLists.txt
index f73b573..c69b907 100755
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -5803,7 +5803,9 @@
   src/f32-igemm/gen/4x8-minmax-aarch32-neon-cortex-a75.S
   src/f32-igemm/gen/4x8-minmax-aarch32-neon-ld64.S
   src/f32-igemm/gen/4x8-minmax-aarch32-neon-prfm-cortex-a75.S
+  src/qc8-gemm/gen/4x8-minmax-fp32-aarch32-neon-mlal-lane-cortex-a53.S
   src/qc8-gemm/gen/4x8-minmax-fp32-aarch32-neon-mlal-lane-ld64.S
+  src/qc8-gemm/gen/4x8-minmax-fp32-aarch32-neon-mlal-lane-prfm-cortex-a53.S
   src/qc8-gemm/gen/4x8-minmax-fp32-aarch32-neon-mlal-lane-prfm-ld64.S
   src/qc8-gemm/gen/4x8-minmax-fp32-aarch32-neonv8-mlal-lane-cortex-a53.S
   src/qc8-gemm/gen/4x8-minmax-fp32-aarch32-neonv8-mlal-lane-ld64.S