aboutsummaryrefslogtreecommitdiff
path: root/src/core/NEON/kernels/arm_gemm/kernels
ModeNameSize
-rw-r--r--a32_sgemm_8x6.hpp2855logplain
d---------a32_sgemm_8x6111logplain
-rw-r--r--a64_ffhybrid_bf16fp32_mmla_6x16.hpp3020logplain
d---------a64_ffhybrid_bf16fp32_mmla_6x1639logplain
-rw-r--r--a64_ffhybrid_fp16_mla_6x32.hpp2968logplain
d---------a64_ffhybrid_fp16_mla_6x3239logplain
-rw-r--r--a64_ffhybrid_fp32_mla_6x16.hpp2960logplain
d---------a64_ffhybrid_fp32_mla_6x1639logplain
-rw-r--r--a64_ffhybrid_fp32bf16fp32_mmla_4x24.hpp3032logplain
d---------a64_ffhybrid_fp32bf16fp32_mmla_4x2439logplain
-rw-r--r--a64_ffinterleaved_bf16fp32_dot_8x12.hpp2890logplain
d---------a64_ffinterleaved_bf16fp32_dot_8x1239logplain
-rw-r--r--a64_ffinterleaved_bf16fp32_mmla_8x12.hpp3083logplain
d---------a64_ffinterleaved_bf16fp32_mmla_8x1239logplain
-rw-r--r--a64_ffinterleaved_fp16_mla_8x24.hpp2843logplain
d---------a64_ffinterleaved_fp16_mla_8x2439logplain
-rw-r--r--a64_ffinterleaved_fp32_mla_8x12.hpp2837logplain
d---------a64_ffinterleaved_fp32_mla_8x1239logplain
-rw-r--r--a64_gemm_s16_8x12.hpp2499logplain
d---------a64_gemm_s16_8x1239logplain
-rw-r--r--a64_gemm_s8_4x4.hpp3110logplain
d---------a64_gemm_s8_4x439logplain
-rw-r--r--a64_gemm_s8_8x12.hpp3632logplain
d---------a64_gemm_s8_8x12110logplain
-rw-r--r--a64_gemm_u16_8x12.hpp2172logplain
d---------a64_gemm_u16_8x1239logplain
-rw-r--r--a64_gemm_u8_4x4.hpp3462logplain
d---------a64_gemm_u8_4x439logplain
-rw-r--r--a64_gemm_u8_8x12.hpp3963logplain
d---------a64_gemm_u8_8x12110logplain
-rw-r--r--a64_hgemm_8x24.hpp3193logplain
d---------a64_hgemm_8x24110logplain
-rw-r--r--a64_hybrid_bf16fp32_dot_6x16.hpp2926logplain
d---------a64_hybrid_bf16fp32_dot_6x1639logplain
-rw-r--r--a64_hybrid_bf16fp32_mmla_6x16.hpp2925logplain
d---------a64_hybrid_bf16fp32_mmla_6x1639logplain
-rw-r--r--a64_hybrid_fp16_mla_6x32.hpp3203logplain
d---------a64_hybrid_fp16_mla_6x3274logplain
-rw-r--r--a64_hybrid_fp32_mla_4x24.hpp3369logplain
d---------a64_hybrid_fp32_mla_4x2474logplain
-rw-r--r--a64_hybrid_fp32_mla_6x16.hpp3374logplain
d---------a64_hybrid_fp32_mla_6x1674logplain
-rw-r--r--a64_hybrid_fp32_mla_8x4.hpp2633logplain
d---------a64_hybrid_fp32_mla_8x474logplain
-rw-r--r--a64_hybrid_fp32bf16fp32_mmla_4x24.hpp2931logplain
d---------a64_hybrid_fp32bf16fp32_mmla_4x2439logplain
-rw-r--r--a64_hybrid_fp32bf16fp32_mmla_6x16.hpp2932logplain
d---------a64_hybrid_fp32bf16fp32_mmla_6x1639logplain
-rw-r--r--a64_hybrid_s8qa_dot_4x16.hpp3228logplain
d---------a64_hybrid_s8qa_dot_4x1674logplain
-rw-r--r--a64_hybrid_s8qa_mmla_4x16.hpp2898logplain
d---------a64_hybrid_s8qa_mmla_4x1639logplain
-rw-r--r--a64_hybrid_s8qs_dot_6x16.hpp3228logplain
d---------a64_hybrid_s8qs_dot_6x1674logplain
-rw-r--r--a64_hybrid_s8qs_mmla_6x16.hpp2898logplain
d---------a64_hybrid_s8qs_mmla_6x1639logplain
-rw-r--r--a64_hybrid_s8s32_dot_6x16.hpp3677logplain
d---------a64_hybrid_s8s32_dot_6x1674logplain
-rw-r--r--a64_hybrid_s8s32_mmla_6x16.hpp3249logplain
d---------a64_hybrid_s8s32_mmla_6x1639logplain
-rw-r--r--a64_hybrid_u8qa_dot_4x16.hpp3235logplain
d---------a64_hybrid_u8qa_dot_4x1674logplain
-rw-r--r--a64_hybrid_u8qa_mmla_4x16.hpp2905logplain
d---------a64_hybrid_u8qa_mmla_4x1639logplain
-rw-r--r--a64_hybrid_u8u32_dot_6x16.hpp3686logplain
d---------a64_hybrid_u8u32_dot_6x1674logplain
-rw-r--r--a64_hybrid_u8u32_mmla_6x16.hpp3258logplain
d---------a64_hybrid_u8u32_mmla_6x1639logplain
-rw-r--r--a64_interleaved_bf16fp32_dot_8x12.hpp2897logplain
d---------a64_interleaved_bf16fp32_dot_8x1273logplain
-rw-r--r--a64_interleaved_bf16fp32_mmla_8x12.hpp3532logplain
d---------a64_interleaved_bf16fp32_mmla_8x1275logplain
-rw-r--r--a64_interleaved_s8s32_mmla_8x12.hpp3492logplain
d---------a64_interleaved_s8s32_mmla_8x1275logplain
-rw-r--r--a64_interleaved_u8u32_mmla_8x12.hpp3498logplain
d---------a64_interleaved_u8u32_mmla_8x1275logplain
-rw-r--r--a64_sgemm_8x12.hpp4359logplain
d---------a64_sgemm_8x12180logplain
-rw-r--r--a64_sgemm_8x6.hpp2377logplain
d---------a64_sgemm_8x639logplain
-rw-r--r--a64_sgemv_pretransposed.hpp2572logplain
d---------a64_sgemv_pretransposed39logplain
-rw-r--r--a64_smallK_hybrid_fp32_mla_6x4.hpp2414logplain
d---------a64_smallK_hybrid_fp32_mla_6x439logplain
-rw-r--r--a64_smallK_hybrid_fp32_mla_8x4.hpp2414logplain
d---------a64_smallK_hybrid_fp32_mla_8x439logplain
-rw-r--r--a64_smallK_hybrid_s8s32_dot_6x4.hpp2721logplain
d---------a64_smallK_hybrid_s8s32_dot_6x474logplain
-rw-r--r--a64_smallK_hybrid_s8s32_dot_8x4.hpp2721logplain
d---------a64_smallK_hybrid_s8s32_dot_8x474logplain
-rw-r--r--a64_smallK_hybrid_u8u32_dot_6x4.hpp2735logplain
d---------a64_smallK_hybrid_u8u32_dot_6x474logplain
-rw-r--r--a64_smallK_hybrid_u8u32_dot_8x4.hpp2735logplain
d---------a64_smallK_hybrid_u8u32_dot_8x474logplain
-rw-r--r--sme2_gemv_bf16fp32_dot_16VL.hpp2318logplain
d---------sme2_gemv_bf16fp32_dot_16VL39logplain
-rw-r--r--sme2_gemv_fp32_mla_16VL.hpp2268logplain
d---------sme2_gemv_fp32_mla_16VL39logplain
-rw-r--r--sme2_gemv_fp32bf16fp32_dot_16VL.hpp2331logplain
d---------sme2_gemv_fp32bf16fp32_dot_16VL39logplain
-rw-r--r--sme2_gemv_s8qa_dot_16VL.hpp2297logplain
d---------sme2_gemv_s8qa_dot_16VL39logplain
-rw-r--r--sme2_gemv_u8qa_dot_16VL.hpp2303logplain
d---------sme2_gemv_u8qa_dot_16VL39logplain
-rw-r--r--sme2_interleaved_nomerge_bf16fp32_mopa_1VLx4VL.hpp2816logplain
d---------sme2_interleaved_nomerge_bf16fp32_mopa_1VLx4VL39logplain
-rw-r--r--sme2_interleaved_nomerge_bf16fp32_mopa_2VLx2VL.hpp2816logplain
d---------sme2_interleaved_nomerge_bf16fp32_mopa_2VLx2VL39logplain
-rw-r--r--sme2_interleaved_nomerge_bf16fp32_mopa_4VLx1VL.hpp2816logplain
d---------sme2_interleaved_nomerge_bf16fp32_mopa_4VLx1VL39logplain
-rw-r--r--sme2_interleaved_nomerge_fp32_mopa_1VLx4VL.hpp2761logplain
d---------sme2_interleaved_nomerge_fp32_mopa_1VLx4VL39logplain
-rw-r--r--sme2_interleaved_nomerge_fp32_mopa_2VLx2VL.hpp2761logplain
d---------sme2_interleaved_nomerge_fp32_mopa_2VLx2VL39logplain
-rw-r--r--sme2_interleaved_nomerge_fp32_mopa_4VLx1VL.hpp2761logplain
d---------sme2_interleaved_nomerge_fp32_mopa_4VLx1VL39logplain
-rw-r--r--sme2_interleaved_nomerge_s8q_mopa_1VLx4VL.hpp2837logplain
d---------sme2_interleaved_nomerge_s8q_mopa_1VLx4VL39logplain
-rw-r--r--sme2_interleaved_nomerge_s8q_mopa_2VLx2VL.hpp2837logplain
d---------sme2_interleaved_nomerge_s8q_mopa_2VLx2VL39logplain
-rw-r--r--sme2_interleaved_nomerge_s8q_mopa_4VLx1VL.hpp2837logplain
d---------sme2_interleaved_nomerge_s8q_mopa_4VLx1VL39logplain
-rw-r--r--sme2_interleaved_nomerge_s8s32_mopa_1VLx4VL.hpp2807logplain
d---------sme2_interleaved_nomerge_s8s32_mopa_1VLx4VL39logplain
-rw-r--r--sme2_interleaved_nomerge_s8s32_mopa_2VLx2VL.hpp2807logplain
d---------sme2_interleaved_nomerge_s8s32_mopa_2VLx2VL39logplain
-rw-r--r--sme2_interleaved_nomerge_s8s32_mopa_4VLx1VL.hpp2807logplain
d---------sme2_interleaved_nomerge_s8s32_mopa_4VLx1VL39logplain
-rw-r--r--sme2_interleaved_nomerge_u8q_mopa_1VLx4VL.hpp2847logplain
d---------sme2_interleaved_nomerge_u8q_mopa_1VLx4VL39logplain
-rw-r--r--sme2_interleaved_nomerge_u8q_mopa_2VLx2VL.hpp2847logplain
d---------sme2_interleaved_nomerge_u8q_mopa_2VLx2VL39logplain
-rw-r--r--sme2_interleaved_nomerge_u8q_mopa_4VLx1VL.hpp2847logplain
d---------sme2_interleaved_nomerge_u8q_mopa_4VLx1VL39logplain
-rw-r--r--sve_ffhybrid_bf16fp32_mmla_6x4VL.hpp3098logplain
d---------sve_ffhybrid_bf16fp32_mmla_6x4VL39logplain
-rw-r--r--sve_ffhybrid_fp16_mla_6x4VL.hpp3308logplain
d---------sve_ffhybrid_fp16_mla_6x4VL76logplain
-rw-r--r--sve_ffhybrid_fp32_mla_6x4VL.hpp3298logplain
d---------sve_ffhybrid_fp32_mla_6x4VL76logplain
-rw-r--r--sve_ffhybrid_fp32bf16fp32_mmla_4x6VL.hpp3111logplain
d---------sve_ffhybrid_fp32bf16fp32_mmla_4x6VL39logplain
-rw-r--r--sve_ffinterleaved_bf16fp32_mmla_8x3VL.hpp3160logplain
d---------sve_ffinterleaved_bf16fp32_mmla_8x3VL39logplain
-rw-r--r--sve_ffinterleaved_fp16_mla_8x3VL.hpp3192logplain
d---------sve_ffinterleaved_fp16_mla_8x3VL76logplain
-rw-r--r--sve_ffinterleaved_fp32_mla_8x3VL.hpp3184logplain
d---------sve_ffinterleaved_fp32_mla_8x3VL76logplain
-rw-r--r--sve_hybrid_bf16fp32_dot_6x4VL.hpp2976logplain
d---------sve_hybrid_bf16fp32_dot_6x4VL39logplain
-rw-r--r--sve_hybrid_bf16fp32_mmla_6x4VL.hpp2975logplain
d---------sve_hybrid_bf16fp32_mmla_6x4VL39logplain
-rw-r--r--sve_hybrid_fp16_mla_6x4VL.hpp3175logplain
d---------sve_hybrid_fp16_mla_6x4VL76logplain
-rw-r--r--sve_hybrid_fp32_mla_6x4VL.hpp3102logplain
d---------sve_hybrid_fp32_mla_6x4VL76logplain
-rw-r--r--sve_hybrid_fp32_mla_8x1VL.hpp2664logplain
d---------sve_hybrid_fp32_mla_8x1VL76logplain
-rw-r--r--sve_hybrid_fp32bf16fp32_mmla_4x6VL.hpp2983logplain
d---------sve_hybrid_fp32bf16fp32_mmla_4x6VL39logplain
-rw-r--r--sve_hybrid_fp32bf16fp32_mmla_6x4VL.hpp2982logplain
d---------sve_hybrid_fp32bf16fp32_mmla_6x4VL39logplain
-rw-r--r--sve_hybrid_s8qa_dot_4x4VL.hpp2878logplain
d---------sve_hybrid_s8qa_dot_4x4VL39logplain
-rw-r--r--sve_hybrid_s8qa_mmla_4x4VL.hpp2877logplain
d---------sve_hybrid_s8qa_mmla_4x4VL39logplain
-rw-r--r--sve_hybrid_s8qs_dot_6x4VL.hpp2878logplain
d---------sve_hybrid_s8qs_dot_6x4VL39logplain
-rw-r--r--sve_hybrid_s8qs_mmla_6x4VL.hpp2877logplain
d---------sve_hybrid_s8qs_mmla_6x4VL39logplain
-rw-r--r--sve_hybrid_s8s32_dot_6x4VL.hpp3303logplain
d---------sve_hybrid_s8s32_dot_6x4VL76logplain
-rw-r--r--sve_hybrid_s8s32_mmla_6x4VL.hpp3303logplain
d---------sve_hybrid_s8s32_mmla_6x4VL39logplain
-rw-r--r--sve_hybrid_u8qa_dot_4x4VL.hpp2886logplain
d---------sve_hybrid_u8qa_dot_4x4VL39logplain
-rw-r--r--sve_hybrid_u8qa_mmla_4x4VL.hpp2885logplain
d---------sve_hybrid_u8qa_mmla_4x4VL39logplain
-rw-r--r--sve_hybrid_u8u32_dot_6x4VL.hpp3308logplain
d---------sve_hybrid_u8u32_dot_6x4VL76logplain
-rw-r--r--sve_hybrid_u8u32_mmla_6x4VL.hpp3313logplain
d---------sve_hybrid_u8u32_mmla_6x4VL39logplain
-rw-r--r--sve_interleaved_bf16fp32_dot_8x3VL.hpp2975logplain
d---------sve_interleaved_bf16fp32_dot_8x3VL39logplain
-rw-r--r--sve_interleaved_bf16fp32_mmla_8x3VL.hpp3339logplain
d---------sve_interleaved_bf16fp32_mmla_8x3VL39logplain
-rw-r--r--sve_interleaved_fp16_mla_8x3VL.hpp3197logplain
d---------sve_interleaved_fp16_mla_8x3VL76logplain
-rw-r--r--sve_interleaved_fp32_mla_8x3VL.hpp3104logplain
d---------sve_interleaved_fp32_mla_8x3VL76logplain
-rw-r--r--sve_interleaved_fp32_mmla_8x3VL.hpp2170logplain
d---------sve_interleaved_fp32_mmla_8x3VL39logplain
-rw-r--r--sve_interleaved_s8s32_dot_8x3VL.hpp3302logplain
d---------sve_interleaved_s8s32_dot_8x3VL76logplain
-rw-r--r--sve_interleaved_s8s32_mmla_8x3VL.hpp3308logplain
d---------sve_interleaved_s8s32_mmla_8x3VL39logplain
-rw-r--r--sve_interleaved_u8u32_dot_8x3VL.hpp3311logplain
d---------sve_interleaved_u8u32_dot_8x3VL76logplain
-rw-r--r--sve_interleaved_u8u32_mmla_8x3VL.hpp3316logplain
d---------sve_interleaved_u8u32_mmla_8x3VL39logplain
-rw-r--r--sve_smallK_hybrid_fp32_mla_8x1VL.hpp2471logplain
d---------sve_smallK_hybrid_fp32_mla_8x1VL39logplain
-rw-r--r--sve_smallK_hybrid_s8s32_dot_8x1VL.hpp2512logplain
d---------sve_smallK_hybrid_s8s32_dot_8x1VL39logplain
-rw-r--r--sve_smallK_hybrid_u8u32_dot_8x1VL.hpp2523logplain
d---------sve_smallK_hybrid_u8u32_dot_8x1VL39logplain