QU8 4x8 NEON MLA Lane microkernel AArch32 assembly language
- Template generated from QS8 source.
- Prefetch and non-prefetch version.
- RNDNU quantization.
PiperOrigin-RevId: 422755432
diff --git a/CMakeLists.txt b/CMakeLists.txt
index 12adf00..1ed6091 100755
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -5785,7 +5785,11 @@
src/qs8-gemm/gen/4x8c4-minmax-rndnu-aarch32-neondot-ld64.S
src/qs8-igemm/gen/4x8-minmax-rndnu-aarch32-neon-mlal-lane-ld64.S
src/qs8-igemm/gen/4x8-minmax-rndnu-aarch32-neon-mlal-lane-prfm-ld64.S
- src/qs8-igemm/gen/4x8c4-minmax-rndnu-aarch32-neondot-ld64.S)
+ src/qs8-igemm/gen/4x8c4-minmax-rndnu-aarch32-neondot-ld64.S
+ src/qu8-gemm/gen/4x8-minmax-rndnu-aarch32-neon-mlal-lane-ld64.S
+ src/qu8-gemm/gen/4x8-minmax-rndnu-aarch32-neon-mlal-lane-prfm-ld64.S
+ src/qu8-igemm/gen/4x8-minmax-rndnu-aarch32-neon-mlal-lane-ld64.S
+ src/qu8-igemm/gen/4x8-minmax-rndnu-aarch32-neon-mlal-lane-prfm-ld64.S)
SET(AARCH64_ASM_MICROKERNEL_SRCS
src/f16-gemm/gen-inc/1x8inc-minmax-aarch64-neonfp16arith-ld64.S