| /dports/graphics/vapoursynth-waifu2x-ncnn-vulkan/vapoursynth-waifu2x-ncnn-vulkan-r4/deps/ncnn/src/layer/arm/ |
| H A D | convolutiondepthwise_5x5_pack4.h | 288 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
532 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
698 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
711 _sum0 = vmlaq_f32(_sum0, _k40, _r40); in convdw5x5s1_pack4_neon()
723 _sum1 = vmlaq_f32(_sum1, _k40, _r50); in convdw5x5s1_pack4_neon()
921 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
1073 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
1187 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
1436 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s2_pack4_neon() local
1593 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s2_pack4_neon() local
[all …]
|
| H A D | convolutiondepthwise_5x5_pack8_fp16s.h | 1395 float16x8_t _k40 = vld1q_f16(k0); in convdw5x5s2_pack8_fp16sa_neon() local
1402 _sum0 = vfmaq_f16(_sum0, _k40, _r40); in convdw5x5s2_pack8_fp16sa_neon()
|
| /dports/graphics/waifu2x-ncnn-vulkan/waifu2x-ncnn-vulkan-20210521/src/ncnn/src/layer/arm/ |
| H A D | convolutiondepthwise_5x5_pack4.h | 288 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
532 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
698 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
711 _sum0 = vmlaq_f32(_sum0, _k40, _r40); in convdw5x5s1_pack4_neon()
723 _sum1 = vmlaq_f32(_sum1, _k40, _r50); in convdw5x5s1_pack4_neon()
921 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
1073 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
1187 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
1436 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s2_pack4_neon() local
1593 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s2_pack4_neon() local
[all …]
|
| H A D | convolutiondepthwise_5x5_pack8_fp16s.h | 1395 float16x8_t _k40 = vld1q_f16(k0); in convdw5x5s2_pack8_fp16sa_neon() local
1402 _sum0 = vfmaq_f16(_sum0, _k40, _r40); in convdw5x5s2_pack8_fp16sa_neon()
|
| /dports/benchmarks/vkpeak/vkpeak-20210430/ncnn/src/layer/arm/ |
| H A D | convolutiondepthwise_5x5_pack4.h | 288 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
532 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
698 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
711 _sum0 = vmlaq_f32(_sum0, _k40, _r40); in convdw5x5s1_pack4_neon()
723 _sum1 = vmlaq_f32(_sum1, _k40, _r50); in convdw5x5s1_pack4_neon()
921 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
1073 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
1187 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
1436 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s2_pack4_neon() local
1593 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s2_pack4_neon() local
[all …]
|
| H A D | convolutiondepthwise_5x5_pack8_fp16s.h | 1395 float16x8_t _k40 = vld1q_f16(k0); in convdw5x5s2_pack8_fp16sa_neon() local
1402 _sum0 = vfmaq_f16(_sum0, _k40, _r40); in convdw5x5s2_pack8_fp16sa_neon()
|
| /dports/misc/ncnn/ncnn-20211208/src/layer/arm/ |
| H A D | convolutiondepthwise_5x5_pack4.h | 288 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
532 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
698 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
711 _sum0 = vmlaq_f32(_sum0, _k40, _r40); in convdw5x5s1_pack4_neon()
723 _sum1 = vmlaq_f32(_sum1, _k40, _r50); in convdw5x5s1_pack4_neon()
921 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
1073 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
1187 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
1436 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s2_pack4_neon() local
1593 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s2_pack4_neon() local
[all …]
|
| H A D | convolutiondepthwise_5x5_pack8_fp16s.h | 1395 float16x8_t _k40 = vld1q_f16(k0); in convdw5x5s2_pack8_fp16sa_neon() local
1402 _sum0 = vfmaq_f16(_sum0, _k40, _r40); in convdw5x5s2_pack8_fp16sa_neon()
|
| /dports/graphics/realsr-ncnn-vulkan/realsr-ncnn-vulkan-20210210/src/ncnn/src/layer/arm/ |
| H A D | convolutiondepthwise_5x5_pack4.h | 288 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
532 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
698 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
711 _sum0 = vmlaq_f32(_sum0, _k40, _r40); in convdw5x5s1_pack4_neon()
723 _sum1 = vmlaq_f32(_sum1, _k40, _r50); in convdw5x5s1_pack4_neon()
921 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
1073 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
1187 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s1_pack4_neon() local
1436 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s2_pack4_neon() local
1593 float32x4_t _k40 = vld1q_f32(k0); in convdw5x5s2_pack4_neon() local
[all …]
|
| H A D | convolutiondepthwise_5x5_pack8_fp16s.h | 1395 float16x8_t _k40 = vld1q_f16(k0); in convdw5x5s2_pack8_fp16sa_neon() local
1402 _sum0 = vfmaq_f16(_sum0, _k40, _r40); in convdw5x5s2_pack8_fp16sa_neon()
|
| /dports/misc/ncnn/ncnn-20211208/src/layer/mips/ |
| H A D | convolutiondepthwise_5x5_pack4.h | 171 v4f32 _k40 = (v4f32)__msa_ld_w(k0, 0); in convdw5x5s1_pack4_msa() local
178 _sum0 = __msa_fmadd_w(_sum0, _k40, _r40); in convdw5x5s1_pack4_msa()
190 _sum1 = __msa_fmadd_w(_sum1, _k40, _r50); in convdw5x5s1_pack4_msa()
317 v4f32 _k40 = (v4f32)__msa_ld_w(k0, 0); in convdw5x5s1_pack4_msa() local
324 _sum0 = __msa_fmadd_w(_sum0, _k40, _r40); in convdw5x5s1_pack4_msa()
480 v4f32 _k40 = (v4f32)__msa_ld_w(k0, 0); in convdw5x5s2_pack4_msa() local
487 _sum0 = __msa_fmadd_w(_sum0, _k40, _r40); in convdw5x5s2_pack4_msa()
|
| H A D | convolution_7x7_pack1to4.h | 292 v4f32 _k40 = (v4f32)__msa_ld_w(kptr, 0); in conv7x7s2_pack1to4_msa() local
320 _sum0 = __msa_fmadd_w(_sum0, _r40, _k40); in conv7x7s2_pack1to4_msa()
321 _sum1 = __msa_fmadd_w(_sum1, _r42, _k40); in conv7x7s2_pack1to4_msa()
322 _sum2 = __msa_fmadd_w(_sum2, _r44, _k40); in conv7x7s2_pack1to4_msa()
323 _sum3 = __msa_fmadd_w(_sum3, _r46, _k40); in conv7x7s2_pack1to4_msa()
566 v4f32 _k40 = (v4f32)__msa_ld_w(kptr, 0); in conv7x7s2_pack1to4_msa() local
579 _sum0 = __msa_fmadd_w(_sum0, (v4f32)__msa_splati_w(_r4, 0), _k40); in conv7x7s2_pack1to4_msa()
|
| /dports/graphics/vapoursynth-waifu2x-ncnn-vulkan/vapoursynth-waifu2x-ncnn-vulkan-r4/deps/ncnn/src/layer/x86/ |
| H A D | convolutiondepthwise_5x5_pack8.h | 133 __m256 _k40 = _mm256_loadu_ps(k0); in convdw5x5s1_pack8_avx() local
140 _sum0 = _mm256_fmadd_ps(_k40, _r40, _sum0); in convdw5x5s1_pack8_avx()
287 __m256 _k40 = _mm256_loadu_ps(k0); in convdw5x5s2_pack8_avx() local
294 _sum0 = _mm256_fmadd_ps(_k40, _r40, _sum0); in convdw5x5s2_pack8_avx()
|
| H A D | convolution_3x3_pack8.h | 226 __m256 _k40 = _mm256_loadu_ps(kptr); in conv3x3s1_pack8_avx() local
237 _sum00 = _mm256_fmadd_ps(_r110, _k40, _sum00); in conv3x3s1_pack8_avx()
255 _sum10 = _mm256_fmadd_ps(_r120, _k40, _sum10); in conv3x3s1_pack8_avx()
566 __m256 _k40 = _mm256_loadu_ps(kptr); in conv3x3s1_pack8_avx() local
577 _sum0 = _mm256_fmadd_ps(_r110, _k40, _sum0); in conv3x3s1_pack8_avx()
|
| /dports/graphics/waifu2x-ncnn-vulkan/waifu2x-ncnn-vulkan-20210521/src/ncnn/src/layer/x86/ |
| H A D | convolutiondepthwise_5x5_pack8.h | 133 __m256 _k40 = _mm256_loadu_ps(k0); in convdw5x5s1_pack8_avx() local
140 _sum0 = _mm256_fmadd_ps(_k40, _r40, _sum0); in convdw5x5s1_pack8_avx()
287 __m256 _k40 = _mm256_loadu_ps(k0); in convdw5x5s2_pack8_avx() local
294 _sum0 = _mm256_fmadd_ps(_k40, _r40, _sum0); in convdw5x5s2_pack8_avx()
|
| H A D | convolution_3x3_pack8.h | 226 __m256 _k40 = _mm256_loadu_ps(kptr); in conv3x3s1_pack8_avx() local
237 _sum00 = _mm256_fmadd_ps(_r110, _k40, _sum00); in conv3x3s1_pack8_avx()
255 _sum10 = _mm256_fmadd_ps(_r120, _k40, _sum10); in conv3x3s1_pack8_avx()
566 __m256 _k40 = _mm256_loadu_ps(kptr); in conv3x3s1_pack8_avx() local
577 _sum0 = _mm256_fmadd_ps(_r110, _k40, _sum0); in conv3x3s1_pack8_avx()
|
| /dports/misc/ncnn/ncnn-20211208/src/layer/x86/ |
| H A D | convolutiondepthwise_5x5_pack8.h | 133 __m256 _k40 = _mm256_loadu_ps(k0); in convdw5x5s1_pack8_avx() local
140 _sum0 = _mm256_comp_fmadd_ps(_k40, _r40, _sum0); in convdw5x5s1_pack8_avx()
287 __m256 _k40 = _mm256_loadu_ps(k0); in convdw5x5s2_pack8_avx() local
294 _sum0 = _mm256_comp_fmadd_ps(_k40, _r40, _sum0); in convdw5x5s2_pack8_avx()
|
| H A D | convolution_3x3_pack8.h | 226 __m256 _k40 = _mm256_loadu_ps(kptr); in conv3x3s1_pack8_avx() local
237 _sum00 = _mm256_comp_fmadd_ps(_r110, _k40, _sum00); in conv3x3s1_pack8_avx()
255 _sum10 = _mm256_comp_fmadd_ps(_r120, _k40, _sum10); in conv3x3s1_pack8_avx()
566 __m256 _k40 = _mm256_loadu_ps(kptr); in conv3x3s1_pack8_avx() local
577 _sum0 = _mm256_comp_fmadd_ps(_r110, _k40, _sum0); in conv3x3s1_pack8_avx()
|
| /dports/benchmarks/vkpeak/vkpeak-20210430/ncnn/src/layer/x86/ |
| H A D | convolutiondepthwise_5x5_pack8.h | 133 __m256 _k40 = _mm256_loadu_ps(k0); in convdw5x5s1_pack8_avx() local
140 _sum0 = _mm256_fmadd_ps(_k40, _r40, _sum0); in convdw5x5s1_pack8_avx()
287 __m256 _k40 = _mm256_loadu_ps(k0); in convdw5x5s2_pack8_avx() local
294 _sum0 = _mm256_fmadd_ps(_k40, _r40, _sum0); in convdw5x5s2_pack8_avx()
|
| H A D | convolution_3x3_pack8.h | 226 __m256 _k40 = _mm256_loadu_ps(kptr); in conv3x3s1_pack8_avx() local
237 _sum00 = _mm256_fmadd_ps(_r110, _k40, _sum00); in conv3x3s1_pack8_avx()
255 _sum10 = _mm256_fmadd_ps(_r120, _k40, _sum10); in conv3x3s1_pack8_avx()
566 __m256 _k40 = _mm256_loadu_ps(kptr); in conv3x3s1_pack8_avx() local
577 _sum0 = _mm256_fmadd_ps(_r110, _k40, _sum0); in conv3x3s1_pack8_avx()
|
| /dports/graphics/realsr-ncnn-vulkan/realsr-ncnn-vulkan-20210210/src/ncnn/src/layer/x86/ |
| H A D | convolutiondepthwise_5x5_pack8.h | 133 __m256 _k40 = _mm256_loadu_ps(k0); in convdw5x5s1_pack8_avx() local
140 _sum0 = _mm256_fmadd_ps(_k40, _r40, _sum0); in convdw5x5s1_pack8_avx()
287 __m256 _k40 = _mm256_loadu_ps(k0); in convdw5x5s2_pack8_avx() local
294 _sum0 = _mm256_fmadd_ps(_k40, _r40, _sum0); in convdw5x5s2_pack8_avx()
|
| /dports/misc/ncnn/ncnn-20211208/src/layer/riscv/ |
| H A D | convolutiondepthwise_5x5_packn.h | 165 vfloat32m1_t _k40 = vle32_v_f32m1(k0, vl); in convdw5x5s1_packn_rvv() local
172 _sum0 = vfmacc_vv_f32m1(_sum0, _k40, _r40, vl); in convdw5x5s1_packn_rvv()
184 _sum1 = vfmacc_vv_f32m1(_sum1, _k40, _r50, vl); in convdw5x5s1_packn_rvv()
303 vfloat32m1_t _k40 = vle32_v_f32m1(k0, vl); in convdw5x5s1_packn_rvv() local
310 _sum0 = vfmacc_vv_f32m1(_sum0, _k40, _r40, vl); in convdw5x5s1_packn_rvv()
461 vfloat32m1_t _k40 = vle32_v_f32m1(k0, vl); in convdw5x5s2_packn_rvv() local
468 _sum0 = vfmacc_vv_f32m1(_sum0, _k40, _r40, vl); in convdw5x5s2_packn_rvv()
|
| H A D | convolutiondepthwise_5x5_packn_fp16s.h | 165 vfloat16m1_t _k40 = vle16_v_f16m1(k0, vl); in convdw5x5s1_packn_fp16sa_rvv() local
172 _sum0 = vfmacc_vv_f16m1(_sum0, _k40, _r40, vl); in convdw5x5s1_packn_fp16sa_rvv()
184 _sum1 = vfmacc_vv_f16m1(_sum1, _k40, _r50, vl); in convdw5x5s1_packn_fp16sa_rvv()
303 vfloat16m1_t _k40 = vle16_v_f16m1(k0, vl); in convdw5x5s1_packn_fp16sa_rvv() local
310 _sum0 = vfmacc_vv_f16m1(_sum0, _k40, _r40, vl); in convdw5x5s1_packn_fp16sa_rvv()
461 vfloat16m1_t _k40 = vle16_v_f16m1(k0, vl); in convdw5x5s2_packn_fp16sa_rvv() local
468 _sum0 = vfmacc_vv_f16m1(_sum0, _k40, _r40, vl); in convdw5x5s2_packn_fp16sa_rvv()
|
| H A D | convolution_7x7_pack1ton.h | 339 vfloat32m1_t _k40 = vle32_v_f32m1(kptr, vl); in conv7x7s2_pack1ton_rvv() local
349 _sum0 = vfmacc_vf_f32m1(_sum0, r4[0], _k40, vl); in conv7x7s2_pack1ton_rvv()
350 _sum1 = vfmacc_vf_f32m1(_sum1, r4[2], _k40, vl); in conv7x7s2_pack1ton_rvv()
351 _sum2 = vfmacc_vf_f32m1(_sum2, r4[4], _k40, vl); in conv7x7s2_pack1ton_rvv()
352 _sum3 = vfmacc_vf_f32m1(_sum3, r4[6], _k40, vl); in conv7x7s2_pack1ton_rvv()
353 _sum4 = vfmacc_vf_f32m1(_sum4, r4[8], _k40, vl); in conv7x7s2_pack1ton_rvv()
354 _sum5 = vfmacc_vf_f32m1(_sum5, r4[10], _k40, vl); in conv7x7s2_pack1ton_rvv()
355 _sum6 = vfmacc_vf_f32m1(_sum6, r4[12], _k40, vl); in conv7x7s2_pack1ton_rvv()
356 _sum7 = vfmacc_vf_f32m1(_sum7, r4[14], _k40, vl); in conv7x7s2_pack1ton_rvv()
722 vfloat32m1_t _k40 = vle32_v_f32m1(kptr, vl); in conv7x7s2_pack1ton_rvv() local
[all …]
|
| H A D | convolution_7x7_pack1ton_fp16s.h | 339 vfloat16m1_t _k40 = vle16_v_f16m1(kptr, vl); in conv7x7s2_pack1ton_fp16sa_rvv() local
349 _sum0 = vfmacc_vf_f16m1(_sum0, r4[0], _k40, vl); in conv7x7s2_pack1ton_fp16sa_rvv()
350 _sum1 = vfmacc_vf_f16m1(_sum1, r4[2], _k40, vl); in conv7x7s2_pack1ton_fp16sa_rvv()
351 _sum2 = vfmacc_vf_f16m1(_sum2, r4[4], _k40, vl); in conv7x7s2_pack1ton_fp16sa_rvv()
352 _sum3 = vfmacc_vf_f16m1(_sum3, r4[6], _k40, vl); in conv7x7s2_pack1ton_fp16sa_rvv()
353 _sum4 = vfmacc_vf_f16m1(_sum4, r4[8], _k40, vl); in conv7x7s2_pack1ton_fp16sa_rvv()
354 _sum5 = vfmacc_vf_f16m1(_sum5, r4[10], _k40, vl); in conv7x7s2_pack1ton_fp16sa_rvv()
355 _sum6 = vfmacc_vf_f16m1(_sum6, r4[12], _k40, vl); in conv7x7s2_pack1ton_fp16sa_rvv()
356 _sum7 = vfmacc_vf_f16m1(_sum7, r4[14], _k40, vl); in conv7x7s2_pack1ton_fp16sa_rvv()
722 vfloat16m1_t _k40 = vle16_v_f16m1(kptr, vl); in conv7x7s2_pack1ton_fp16sa_rvv() local
[all …]
|