Home
last modified time | relevance | path

Searched refs:_k21 (Results 1 – 25 of 97) sorted by relevance

1234

/dports/graphics/vapoursynth-waifu2x-ncnn-vulkan/vapoursynth-waifu2x-ncnn-vulkan-r4/deps/ncnn/src/layer/arm/
H A Dconvolutiondepthwise_3x3_pack4.h213 "w"(_k21), // %19 in convdw3x3s1_pack4_neon()
313 "w"(_k21), // %19 in convdw3x3s1_pack4_neon()
393 "w"(_k21), // %19 in convdw3x3s1_pack4_neon()
507 "w"(_k21), // %15 in convdw3x3s1_pack4_neon()
608 "w"(_k21), // %15 in convdw3x3s1_pack4_neon()
686 "w"(_k21), // %15 in convdw3x3s1_pack4_neon()
759 "w"(_k21), // %15 in convdw3x3s1_pack4_neon()
948 "w"(_k21), // %15 in convdw3x3s2_pack4_neon()
1083 "w"(_k21), // %15 in convdw3x3s2_pack4_neon()
1164 "w"(_k21), // %15 in convdw3x3s2_pack4_neon()
[all …]
H A Dconvolutiondepthwise_3x3_pack4_bf16s.h285 "w"(_k21), // %19 in convdw3x3s1_pack4_bf16s_neon()
431 "w"(_k21), // %19 in convdw3x3s1_pack4_bf16s_neon()
536 "w"(_k21), // %19 in convdw3x3s1_pack4_bf16s_neon()
688 "w"(_k21), // %15 in convdw3x3s1_pack4_bf16s_neon()
821 "w"(_k21), // %15 in convdw3x3s1_pack4_bf16s_neon()
917 "w"(_k21), // %15 in convdw3x3s1_pack4_bf16s_neon()
1008 "w"(_k21), // %15 in convdw3x3s1_pack4_bf16s_neon()
1035 _sum0 = vmlaq_f32(_sum0, _k21, _r21); in convdw3x3s1_pack4_bf16s_neon()
1250 "w"(_k21), // %15 in convdw3x3s2_pack4_bf16s_neon()
1358 "w"(_k21), // %15 in convdw3x3s2_pack4_bf16s_neon()
[all …]
H A Dconvolutiondepthwise_3x3_pack8_fp16s.h52 float16x8_t _k21 = vld1q_f16(k0 + 56); in convdw3x3s1_pack8_fp16sa_neon() local
209 "w"(_k21), // %19 in convdw3x3s1_pack8_fp16sa_neon()
304 "w"(_k21), // %19 in convdw3x3s1_pack8_fp16sa_neon()
383 "w"(_k21), // %19 in convdw3x3s1_pack8_fp16sa_neon()
488 "w"(_k21), // %15 in convdw3x3s1_pack8_fp16sa_neon()
555 "w"(_k21), // %15 in convdw3x3s1_pack8_fp16sa_neon()
612 "w"(_k21), // %15 in convdw3x3s1_pack8_fp16sa_neon()
662 float16x8_t _k21 = vld1q_f16(k0 + 56); in convdw3x3s2_pack8_fp16sa_neon() local
766 "w"(_k21), // %15 in convdw3x3s2_pack8_fp16sa_neon()
841 "w"(_k21), // %15 in convdw3x3s2_pack8_fp16sa_neon()
[all …]
H A Dconvolution_3x3_pack1to4_fp16s.h52 float16x4_t _k21 = vld1_f16(k0 + 28); in conv3x3s1_pack1to4_fp16sa_neon() local
181 "w"(_k21), // %15 in conv3x3s1_pack1to4_fp16sa_neon()
266 "w"(_k21), // %15 in conv3x3s1_pack1to4_fp16sa_neon()
327 "w"(_k21), // %15 in conv3x3s1_pack1to4_fp16sa_neon()
379 "w"(_k21), // %15 in conv3x3s1_pack1to4_fp16sa_neon()
434 float16x4_t _k21 = vld1_f16(k0 + 28); in conv3x3s2_pack1to4_fp16sa_neon() local
521 "w"(_k21), // %15 in conv3x3s2_pack1to4_fp16sa_neon()
581 "w"(_k21), // %15 in conv3x3s2_pack1to4_fp16sa_neon()
633 "w"(_k21), // %15 in conv3x3s2_pack1to4_fp16sa_neon()
H A Dconvolution_3x3_pack1to8_fp16s.h52 float16x8_t _k21 = vld1q_f16(k0 + 56); in conv3x3s1_pack1to8_fp16sa_neon() local
181 "w"(_k21), // %15 in conv3x3s1_pack1to8_fp16sa_neon()
266 "w"(_k21), // %15 in conv3x3s1_pack1to8_fp16sa_neon()
327 "w"(_k21), // %15 in conv3x3s1_pack1to8_fp16sa_neon()
379 "w"(_k21), // %15 in conv3x3s1_pack1to8_fp16sa_neon()
434 float16x8_t _k21 = vld1q_f16(k0 + 56); in conv3x3s2_pack1to8_fp16sa_neon() local
521 "w"(_k21), // %15 in conv3x3s2_pack1to8_fp16sa_neon()
581 "w"(_k21), // %15 in conv3x3s2_pack1to8_fp16sa_neon()
633 "w"(_k21), // %15 in conv3x3s2_pack1to8_fp16sa_neon()
/dports/graphics/waifu2x-ncnn-vulkan/waifu2x-ncnn-vulkan-20210521/src/ncnn/src/layer/arm/
H A Dconvolutiondepthwise_3x3_pack4.h213 "w"(_k21), // %19 in convdw3x3s1_pack4_neon()
313 "w"(_k21), // %19 in convdw3x3s1_pack4_neon()
393 "w"(_k21), // %19 in convdw3x3s1_pack4_neon()
507 "w"(_k21), // %15 in convdw3x3s1_pack4_neon()
608 "w"(_k21), // %15 in convdw3x3s1_pack4_neon()
686 "w"(_k21), // %15 in convdw3x3s1_pack4_neon()
759 "w"(_k21), // %15 in convdw3x3s1_pack4_neon()
948 "w"(_k21), // %15 in convdw3x3s2_pack4_neon()
1083 "w"(_k21), // %15 in convdw3x3s2_pack4_neon()
1164 "w"(_k21), // %15 in convdw3x3s2_pack4_neon()
[all …]
H A Dconvolutiondepthwise_3x3_pack4_bf16s.h285 "w"(_k21), // %19 in convdw3x3s1_pack4_bf16s_neon()
431 "w"(_k21), // %19 in convdw3x3s1_pack4_bf16s_neon()
536 "w"(_k21), // %19 in convdw3x3s1_pack4_bf16s_neon()
688 "w"(_k21), // %15 in convdw3x3s1_pack4_bf16s_neon()
821 "w"(_k21), // %15 in convdw3x3s1_pack4_bf16s_neon()
917 "w"(_k21), // %15 in convdw3x3s1_pack4_bf16s_neon()
1008 "w"(_k21), // %15 in convdw3x3s1_pack4_bf16s_neon()
1035 _sum0 = vmlaq_f32(_sum0, _k21, _r21); in convdw3x3s1_pack4_bf16s_neon()
1250 "w"(_k21), // %15 in convdw3x3s2_pack4_bf16s_neon()
1358 "w"(_k21), // %15 in convdw3x3s2_pack4_bf16s_neon()
[all …]
H A Dconvolutiondepthwise_3x3_pack8_fp16s.h52 float16x8_t _k21 = vld1q_f16(k0 + 56); in convdw3x3s1_pack8_fp16sa_neon() local
209 "w"(_k21), // %19 in convdw3x3s1_pack8_fp16sa_neon()
304 "w"(_k21), // %19 in convdw3x3s1_pack8_fp16sa_neon()
383 "w"(_k21), // %19 in convdw3x3s1_pack8_fp16sa_neon()
488 "w"(_k21), // %15 in convdw3x3s1_pack8_fp16sa_neon()
555 "w"(_k21), // %15 in convdw3x3s1_pack8_fp16sa_neon()
612 "w"(_k21), // %15 in convdw3x3s1_pack8_fp16sa_neon()
662 float16x8_t _k21 = vld1q_f16(k0 + 56); in convdw3x3s2_pack8_fp16sa_neon() local
766 "w"(_k21), // %15 in convdw3x3s2_pack8_fp16sa_neon()
841 "w"(_k21), // %15 in convdw3x3s2_pack8_fp16sa_neon()
[all …]
H A Dconvolution_3x3_pack1to8_fp16s.h52 float16x8_t _k21 = vld1q_f16(k0 + 56); in conv3x3s1_pack1to8_fp16sa_neon() local
181 "w"(_k21), // %15 in conv3x3s1_pack1to8_fp16sa_neon()
266 "w"(_k21), // %15 in conv3x3s1_pack1to8_fp16sa_neon()
327 "w"(_k21), // %15 in conv3x3s1_pack1to8_fp16sa_neon()
379 "w"(_k21), // %15 in conv3x3s1_pack1to8_fp16sa_neon()
434 float16x8_t _k21 = vld1q_f16(k0 + 56); in conv3x3s2_pack1to8_fp16sa_neon() local
521 "w"(_k21), // %15 in conv3x3s2_pack1to8_fp16sa_neon()
581 "w"(_k21), // %15 in conv3x3s2_pack1to8_fp16sa_neon()
633 "w"(_k21), // %15 in conv3x3s2_pack1to8_fp16sa_neon()
H A Dconvolution_3x3_pack1to4_fp16s.h52 float16x4_t _k21 = vld1_f16(k0 + 28); in conv3x3s1_pack1to4_fp16sa_neon() local
181 "w"(_k21), // %15 in conv3x3s1_pack1to4_fp16sa_neon()
266 "w"(_k21), // %15 in conv3x3s1_pack1to4_fp16sa_neon()
327 "w"(_k21), // %15 in conv3x3s1_pack1to4_fp16sa_neon()
379 "w"(_k21), // %15 in conv3x3s1_pack1to4_fp16sa_neon()
434 float16x4_t _k21 = vld1_f16(k0 + 28); in conv3x3s2_pack1to4_fp16sa_neon() local
521 "w"(_k21), // %15 in conv3x3s2_pack1to4_fp16sa_neon()
581 "w"(_k21), // %15 in conv3x3s2_pack1to4_fp16sa_neon()
633 "w"(_k21), // %15 in conv3x3s2_pack1to4_fp16sa_neon()
/dports/benchmarks/vkpeak/vkpeak-20210430/ncnn/src/layer/arm/
H A Dconvolutiondepthwise_3x3_pack4.h213 "w"(_k21), // %19 in convdw3x3s1_pack4_neon()
313 "w"(_k21), // %19 in convdw3x3s1_pack4_neon()
393 "w"(_k21), // %19 in convdw3x3s1_pack4_neon()
507 "w"(_k21), // %15 in convdw3x3s1_pack4_neon()
608 "w"(_k21), // %15 in convdw3x3s1_pack4_neon()
686 "w"(_k21), // %15 in convdw3x3s1_pack4_neon()
759 "w"(_k21), // %15 in convdw3x3s1_pack4_neon()
948 "w"(_k21), // %15 in convdw3x3s2_pack4_neon()
1083 "w"(_k21), // %15 in convdw3x3s2_pack4_neon()
1164 "w"(_k21), // %15 in convdw3x3s2_pack4_neon()
[all …]
H A Dconvolutiondepthwise_3x3_pack4_bf16s.h285 "w"(_k21), // %19 in convdw3x3s1_pack4_bf16s_neon()
431 "w"(_k21), // %19 in convdw3x3s1_pack4_bf16s_neon()
536 "w"(_k21), // %19 in convdw3x3s1_pack4_bf16s_neon()
688 "w"(_k21), // %15 in convdw3x3s1_pack4_bf16s_neon()
821 "w"(_k21), // %15 in convdw3x3s1_pack4_bf16s_neon()
917 "w"(_k21), // %15 in convdw3x3s1_pack4_bf16s_neon()
1008 "w"(_k21), // %15 in convdw3x3s1_pack4_bf16s_neon()
1035 _sum0 = vmlaq_f32(_sum0, _k21, _r21); in convdw3x3s1_pack4_bf16s_neon()
1250 "w"(_k21), // %15 in convdw3x3s2_pack4_bf16s_neon()
1358 "w"(_k21), // %15 in convdw3x3s2_pack4_bf16s_neon()
[all …]
H A Dconvolutiondepthwise_3x3_pack8_fp16s.h52 float16x8_t _k21 = vld1q_f16(k0 + 56); in convdw3x3s1_pack8_fp16sa_neon() local
209 "w"(_k21), // %19 in convdw3x3s1_pack8_fp16sa_neon()
304 "w"(_k21), // %19 in convdw3x3s1_pack8_fp16sa_neon()
383 "w"(_k21), // %19 in convdw3x3s1_pack8_fp16sa_neon()
488 "w"(_k21), // %15 in convdw3x3s1_pack8_fp16sa_neon()
555 "w"(_k21), // %15 in convdw3x3s1_pack8_fp16sa_neon()
612 "w"(_k21), // %15 in convdw3x3s1_pack8_fp16sa_neon()
662 float16x8_t _k21 = vld1q_f16(k0 + 56); in convdw3x3s2_pack8_fp16sa_neon() local
766 "w"(_k21), // %15 in convdw3x3s2_pack8_fp16sa_neon()
841 "w"(_k21), // %15 in convdw3x3s2_pack8_fp16sa_neon()
[all …]
H A Dconvolution_3x3_pack1to4_fp16s.h52 float16x4_t _k21 = vld1_f16(k0 + 28); in conv3x3s1_pack1to4_fp16sa_neon() local
181 "w"(_k21), // %15 in conv3x3s1_pack1to4_fp16sa_neon()
266 "w"(_k21), // %15 in conv3x3s1_pack1to4_fp16sa_neon()
327 "w"(_k21), // %15 in conv3x3s1_pack1to4_fp16sa_neon()
379 "w"(_k21), // %15 in conv3x3s1_pack1to4_fp16sa_neon()
434 float16x4_t _k21 = vld1_f16(k0 + 28); in conv3x3s2_pack1to4_fp16sa_neon() local
521 "w"(_k21), // %15 in conv3x3s2_pack1to4_fp16sa_neon()
581 "w"(_k21), // %15 in conv3x3s2_pack1to4_fp16sa_neon()
633 "w"(_k21), // %15 in conv3x3s2_pack1to4_fp16sa_neon()
H A Dconvolution_3x3_pack1to8_fp16s.h52 float16x8_t _k21 = vld1q_f16(k0 + 56); in conv3x3s1_pack1to8_fp16sa_neon() local
181 "w"(_k21), // %15 in conv3x3s1_pack1to8_fp16sa_neon()
266 "w"(_k21), // %15 in conv3x3s1_pack1to8_fp16sa_neon()
327 "w"(_k21), // %15 in conv3x3s1_pack1to8_fp16sa_neon()
379 "w"(_k21), // %15 in conv3x3s1_pack1to8_fp16sa_neon()
434 float16x8_t _k21 = vld1q_f16(k0 + 56); in conv3x3s2_pack1to8_fp16sa_neon() local
521 "w"(_k21), // %15 in conv3x3s2_pack1to8_fp16sa_neon()
581 "w"(_k21), // %15 in conv3x3s2_pack1to8_fp16sa_neon()
633 "w"(_k21), // %15 in conv3x3s2_pack1to8_fp16sa_neon()
/dports/misc/ncnn/ncnn-20211208/src/layer/arm/
H A Dconvolutiondepthwise_3x3_pack4.h213 "w"(_k21), // %19 in convdw3x3s1_pack4_neon()
313 "w"(_k21), // %19 in convdw3x3s1_pack4_neon()
393 "w"(_k21), // %19 in convdw3x3s1_pack4_neon()
507 "w"(_k21), // %15 in convdw3x3s1_pack4_neon()
608 "w"(_k21), // %15 in convdw3x3s1_pack4_neon()
686 "w"(_k21), // %15 in convdw3x3s1_pack4_neon()
759 "w"(_k21), // %15 in convdw3x3s1_pack4_neon()
948 "w"(_k21), // %15 in convdw3x3s2_pack4_neon()
1083 "w"(_k21), // %15 in convdw3x3s2_pack4_neon()
1164 "w"(_k21), // %15 in convdw3x3s2_pack4_neon()
[all …]
H A Dconvolutiondepthwise_3x3_pack4_bf16s.h285 "w"(_k21), // %19 in convdw3x3s1_pack4_bf16s_neon()
431 "w"(_k21), // %19 in convdw3x3s1_pack4_bf16s_neon()
536 "w"(_k21), // %19 in convdw3x3s1_pack4_bf16s_neon()
688 "w"(_k21), // %15 in convdw3x3s1_pack4_bf16s_neon()
821 "w"(_k21), // %15 in convdw3x3s1_pack4_bf16s_neon()
917 "w"(_k21), // %15 in convdw3x3s1_pack4_bf16s_neon()
1008 "w"(_k21), // %15 in convdw3x3s1_pack4_bf16s_neon()
1035 _sum0 = vmlaq_f32(_sum0, _k21, _r21); in convdw3x3s1_pack4_bf16s_neon()
1250 "w"(_k21), // %15 in convdw3x3s2_pack4_bf16s_neon()
1358 "w"(_k21), // %15 in convdw3x3s2_pack4_bf16s_neon()
[all …]
H A Dconvolutiondepthwise_3x3_pack8_fp16s.h52 float16x8_t _k21 = vld1q_f16(k0 + 56); in convdw3x3s1_pack8_fp16sa_neon() local
209 "w"(_k21), // %19 in convdw3x3s1_pack8_fp16sa_neon()
304 "w"(_k21), // %19 in convdw3x3s1_pack8_fp16sa_neon()
383 "w"(_k21), // %19 in convdw3x3s1_pack8_fp16sa_neon()
488 "w"(_k21), // %15 in convdw3x3s1_pack8_fp16sa_neon()
555 "w"(_k21), // %15 in convdw3x3s1_pack8_fp16sa_neon()
612 "w"(_k21), // %15 in convdw3x3s1_pack8_fp16sa_neon()
662 float16x8_t _k21 = vld1q_f16(k0 + 56); in convdw3x3s2_pack8_fp16sa_neon() local
766 "w"(_k21), // %15 in convdw3x3s2_pack8_fp16sa_neon()
841 "w"(_k21), // %15 in convdw3x3s2_pack8_fp16sa_neon()
[all …]
H A Dconvolution_3x3_pack1to4_fp16s.h52 float16x4_t _k21 = vld1_f16(k0 + 28); in conv3x3s1_pack1to4_fp16sa_neon() local
181 "w"(_k21), // %15 in conv3x3s1_pack1to4_fp16sa_neon()
266 "w"(_k21), // %15 in conv3x3s1_pack1to4_fp16sa_neon()
327 "w"(_k21), // %15 in conv3x3s1_pack1to4_fp16sa_neon()
379 "w"(_k21), // %15 in conv3x3s1_pack1to4_fp16sa_neon()
434 float16x4_t _k21 = vld1_f16(k0 + 28); in conv3x3s2_pack1to4_fp16sa_neon() local
521 "w"(_k21), // %15 in conv3x3s2_pack1to4_fp16sa_neon()
581 "w"(_k21), // %15 in conv3x3s2_pack1to4_fp16sa_neon()
633 "w"(_k21), // %15 in conv3x3s2_pack1to4_fp16sa_neon()
H A Dconvolution_3x3_pack1to8_fp16s.h52 float16x8_t _k21 = vld1q_f16(k0 + 56); in conv3x3s1_pack1to8_fp16sa_neon() local
181 "w"(_k21), // %15 in conv3x3s1_pack1to8_fp16sa_neon()
266 "w"(_k21), // %15 in conv3x3s1_pack1to8_fp16sa_neon()
327 "w"(_k21), // %15 in conv3x3s1_pack1to8_fp16sa_neon()
379 "w"(_k21), // %15 in conv3x3s1_pack1to8_fp16sa_neon()
434 float16x8_t _k21 = vld1q_f16(k0 + 56); in conv3x3s2_pack1to8_fp16sa_neon() local
521 "w"(_k21), // %15 in conv3x3s2_pack1to8_fp16sa_neon()
581 "w"(_k21), // %15 in conv3x3s2_pack1to8_fp16sa_neon()
633 "w"(_k21), // %15 in conv3x3s2_pack1to8_fp16sa_neon()
/dports/graphics/realsr-ncnn-vulkan/realsr-ncnn-vulkan-20210210/src/ncnn/src/layer/arm/
H A Dconvolutiondepthwise_3x3_pack4.h213 "w"(_k21), // %19 in convdw3x3s1_pack4_neon()
313 "w"(_k21), // %19 in convdw3x3s1_pack4_neon()
393 "w"(_k21), // %19 in convdw3x3s1_pack4_neon()
507 "w"(_k21), // %15 in convdw3x3s1_pack4_neon()
608 "w"(_k21), // %15 in convdw3x3s1_pack4_neon()
686 "w"(_k21), // %15 in convdw3x3s1_pack4_neon()
759 "w"(_k21), // %15 in convdw3x3s1_pack4_neon()
948 "w"(_k21), // %15 in convdw3x3s2_pack4_neon()
1083 "w"(_k21), // %15 in convdw3x3s2_pack4_neon()
1164 "w"(_k21), // %15 in convdw3x3s2_pack4_neon()
[all …]
H A Dconvolutiondepthwise_3x3_pack4_bf16s.h285 "w"(_k21), // %19 in convdw3x3s1_pack4_bf16s_neon()
431 "w"(_k21), // %19 in convdw3x3s1_pack4_bf16s_neon()
536 "w"(_k21), // %19 in convdw3x3s1_pack4_bf16s_neon()
688 "w"(_k21), // %15 in convdw3x3s1_pack4_bf16s_neon()
821 "w"(_k21), // %15 in convdw3x3s1_pack4_bf16s_neon()
917 "w"(_k21), // %15 in convdw3x3s1_pack4_bf16s_neon()
1008 "w"(_k21), // %15 in convdw3x3s1_pack4_bf16s_neon()
1035 _sum0 = vmlaq_f32(_sum0, _k21, _r21); in convdw3x3s1_pack4_bf16s_neon()
1250 "w"(_k21), // %15 in convdw3x3s2_pack4_bf16s_neon()
1358 "w"(_k21), // %15 in convdw3x3s2_pack4_bf16s_neon()
[all …]
H A Dconvolutiondepthwise_3x3_pack8_fp16s.h52 float16x8_t _k21 = vld1q_f16(k0 + 56); in convdw3x3s1_pack8_fp16sa_neon() local
209 "w"(_k21), // %19 in convdw3x3s1_pack8_fp16sa_neon()
304 "w"(_k21), // %19 in convdw3x3s1_pack8_fp16sa_neon()
383 "w"(_k21), // %19 in convdw3x3s1_pack8_fp16sa_neon()
488 "w"(_k21), // %15 in convdw3x3s1_pack8_fp16sa_neon()
555 "w"(_k21), // %15 in convdw3x3s1_pack8_fp16sa_neon()
612 "w"(_k21), // %15 in convdw3x3s1_pack8_fp16sa_neon()
662 float16x8_t _k21 = vld1q_f16(k0 + 56); in convdw3x3s2_pack8_fp16sa_neon() local
766 "w"(_k21), // %15 in convdw3x3s2_pack8_fp16sa_neon()
841 "w"(_k21), // %15 in convdw3x3s2_pack8_fp16sa_neon()
[all …]
H A Dconvolution_3x3_pack1to8_fp16s.h52 float16x8_t _k21 = vld1q_f16(k0 + 56); in conv3x3s1_pack1to8_fp16sa_neon() local
181 "w"(_k21), // %15 in conv3x3s1_pack1to8_fp16sa_neon()
266 "w"(_k21), // %15 in conv3x3s1_pack1to8_fp16sa_neon()
327 "w"(_k21), // %15 in conv3x3s1_pack1to8_fp16sa_neon()
379 "w"(_k21), // %15 in conv3x3s1_pack1to8_fp16sa_neon()
434 float16x8_t _k21 = vld1q_f16(k0 + 56); in conv3x3s2_pack1to8_fp16sa_neon() local
521 "w"(_k21), // %15 in conv3x3s2_pack1to8_fp16sa_neon()
581 "w"(_k21), // %15 in conv3x3s2_pack1to8_fp16sa_neon()
633 "w"(_k21), // %15 in conv3x3s2_pack1to8_fp16sa_neon()
/dports/misc/ncnn/ncnn-20211208/src/layer/riscv/
H A Dconvolution_3x3_pack1ton.h55 vfloat32m1_t _k21 = vle32_v_f32m1(k0 + packn * 7, vl); in conv3x3s1_pack1ton_rvv() local
131 _sum0 = vfmacc_vf_f32m1(_sum0, r2[1], _k21, vl); in conv3x3s1_pack1ton_rvv()
132 _sum1 = vfmacc_vf_f32m1(_sum1, r2[2], _k21, vl); in conv3x3s1_pack1ton_rvv()
133 _sum2 = vfmacc_vf_f32m1(_sum2, r2[3], _k21, vl); in conv3x3s1_pack1ton_rvv()
134 _sum3 = vfmacc_vf_f32m1(_sum3, r2[4], _k21, vl); in conv3x3s1_pack1ton_rvv()
135 _sum4 = vfmacc_vf_f32m1(_sum4, r2[5], _k21, vl); in conv3x3s1_pack1ton_rvv()
136 _sum5 = vfmacc_vf_f32m1(_sum5, r2[6], _k21, vl); in conv3x3s1_pack1ton_rvv()
137 _sum6 = vfmacc_vf_f32m1(_sum6, r2[7], _k21, vl); in conv3x3s1_pack1ton_rvv()
138 _sum7 = vfmacc_vf_f32m1(_sum7, r2[8], _k21, vl); in conv3x3s1_pack1ton_rvv()
200 _sum0 = vfmacc_vf_f32m1(_sum0, r2[1], _k21, vl); in conv3x3s1_pack1ton_rvv()
[all …]

1234