/external/XNNPACK/src/qs8-gemm/gen/ |
D | 2x8c8-minmax-neon-mull-padal.c | 126 const int32x4_t vsum1x67 = vpaddq_s32(vacc1x6, vacc1x7); in xnn_qs8_gemm_minmax_ukernel_2x8c8__neon_mull_padal() local 158 const int32x2_t vsum1x67 = vpadd_s32(vpsum1x6, vpsum1x7); in xnn_qs8_gemm_minmax_ukernel_2x8c8__neon_mull_padal() local
|
D | 2x8c16-minmax-neon-mlal-padal.c | 142 const int32x4_t vsum1x67 = vpaddq_s32(vacc1x6, vacc1x7); in xnn_qs8_gemm_minmax_ukernel_2x8c16__neon_mlal_padal() local 174 const int32x2_t vsum1x67 = vpadd_s32(vpsum1x6, vpsum1x7); in xnn_qs8_gemm_minmax_ukernel_2x8c16__neon_mlal_padal() local
|
D | 2x8c8-minmax-neon-mlal-padal.c | 201 const int32x4_t vsum1x67 = vpaddq_s32(vacc1x6, vacc1x7); in xnn_qs8_gemm_minmax_ukernel_2x8c8__neon_mlal_padal() local 233 const int32x2_t vsum1x67 = vpadd_s32(vpsum1x6, vpsum1x7); in xnn_qs8_gemm_minmax_ukernel_2x8c8__neon_mlal_padal() local
|
D | 3x8c16-minmax-neon-mlal-padal.c | 181 const int32x4_t vsum1x67 = vpaddq_s32(vacc1x6, vacc1x7); in xnn_qs8_gemm_minmax_ukernel_3x8c16__neon_mlal_padal() local 219 const int32x2_t vsum1x67 = vpadd_s32(vpsum1x6, vpsum1x7); in xnn_qs8_gemm_minmax_ukernel_3x8c16__neon_mlal_padal() local
|
D | 3x8c8-minmax-neon-mull-padal.c | 157 const int32x4_t vsum1x67 = vpaddq_s32(vacc1x6, vacc1x7); in xnn_qs8_gemm_minmax_ukernel_3x8c8__neon_mull_padal() local 195 const int32x2_t vsum1x67 = vpadd_s32(vpsum1x6, vpsum1x7); in xnn_qs8_gemm_minmax_ukernel_3x8c8__neon_mull_padal() local
|
D | 2x16c16-minmax-neon-mlal-padal.c | 218 const int32x4_t vsum1x67 = vpaddq_s32(vacc1x6, vacc1x7); in xnn_qs8_gemm_minmax_ukernel_2x16c16__neon_mlal_padal() local 272 const int32x2_t vsum1x67 = vpadd_s32(vpsum1x6, vpsum1x7); in xnn_qs8_gemm_minmax_ukernel_2x16c16__neon_mlal_padal() local
|
D | 3x8c8-minmax-neon-mlal-padal.c | 258 const int32x4_t vsum1x67 = vpaddq_s32(vacc1x6, vacc1x7); in xnn_qs8_gemm_minmax_ukernel_3x8c8__neon_mlal_padal() local 296 const int32x2_t vsum1x67 = vpadd_s32(vpsum1x6, vpsum1x7); in xnn_qs8_gemm_minmax_ukernel_3x8c8__neon_mlal_padal() local
|
D | 4x8c8-minmax-neon-mull-padal.c | 188 const int32x4_t vsum1x67 = vpaddq_s32(vacc1x6, vacc1x7); in xnn_qs8_gemm_minmax_ukernel_4x8c8__neon_mull_padal() local 232 const int32x2_t vsum1x67 = vpadd_s32(vpsum1x6, vpsum1x7); in xnn_qs8_gemm_minmax_ukernel_4x8c8__neon_mull_padal() local
|
D | 2x16c8-minmax-neon-mull-padal.c | 186 const int32x4_t vsum1x67 = vpaddq_s32(vacc1x6, vacc1x7); in xnn_qs8_gemm_minmax_ukernel_2x16c8__neon_mull_padal() local 240 const int32x2_t vsum1x67 = vpadd_s32(vpsum1x6, vpsum1x7); in xnn_qs8_gemm_minmax_ukernel_2x16c8__neon_mull_padal() local
|
D | 4x8c16-minmax-neon-mlal-padal.c | 220 const int32x4_t vsum1x67 = vpaddq_s32(vacc1x6, vacc1x7); in xnn_qs8_gemm_minmax_ukernel_4x8c16__neon_mlal_padal() local 264 const int32x2_t vsum1x67 = vpadd_s32(vpsum1x6, vpsum1x7); in xnn_qs8_gemm_minmax_ukernel_4x8c16__neon_mlal_padal() local
|
D | 4x8c8-minmax-neon-mlal-padal.c | 315 const int32x4_t vsum1x67 = vpaddq_s32(vacc1x6, vacc1x7); in xnn_qs8_gemm_minmax_ukernel_4x8c8__neon_mlal_padal() local 359 const int32x2_t vsum1x67 = vpadd_s32(vpsum1x6, vpsum1x7); in xnn_qs8_gemm_minmax_ukernel_4x8c8__neon_mlal_padal() local
|
D | 3x16c8-minmax-neon-mull-padal.c | 241 const int32x4_t vsum1x67 = vpaddq_s32(vacc1x6, vacc1x7); in xnn_qs8_gemm_minmax_ukernel_3x16c8__neon_mull_padal() local 307 const int32x2_t vsum1x67 = vpadd_s32(vpsum1x6, vpsum1x7); in xnn_qs8_gemm_minmax_ukernel_3x16c8__neon_mull_padal() local
|
D | 2x16c8-minmax-neon-mlal-padal.c | 325 const int32x4_t vsum1x67 = vpaddq_s32(vacc1x6, vacc1x7); in xnn_qs8_gemm_minmax_ukernel_2x16c8__neon_mlal_padal() local 379 const int32x2_t vsum1x67 = vpadd_s32(vpsum1x6, vpsum1x7); in xnn_qs8_gemm_minmax_ukernel_2x16c8__neon_mlal_padal() local
|
/external/XNNPACK/src/qs8-igemm/gen/ |
D | 2x8c16-minmax-neon-mlal-padal.c | 158 const int32x4_t vsum1x67 = vpaddq_s32(vacc1x6, vacc1x7); in xnn_qs8_igemm_minmax_ukernel_2x8c16__neon_mlal_padal() local 190 const int32x2_t vsum1x67 = vpadd_s32(vpsum1x6, vpsum1x7); in xnn_qs8_igemm_minmax_ukernel_2x8c16__neon_mlal_padal() local
|
D | 2x8c8-minmax-neon-mull-padal.c | 142 const int32x4_t vsum1x67 = vpaddq_s32(vacc1x6, vacc1x7); in xnn_qs8_igemm_minmax_ukernel_2x8c8__neon_mull_padal() local 174 const int32x2_t vsum1x67 = vpadd_s32(vpsum1x6, vpsum1x7); in xnn_qs8_igemm_minmax_ukernel_2x8c8__neon_mull_padal() local
|
D | 2x8c8-minmax-neon-mlal-padal.c | 217 const int32x4_t vsum1x67 = vpaddq_s32(vacc1x6, vacc1x7); in xnn_qs8_igemm_minmax_ukernel_2x8c8__neon_mlal_padal() local 249 const int32x2_t vsum1x67 = vpadd_s32(vpsum1x6, vpsum1x7); in xnn_qs8_igemm_minmax_ukernel_2x8c8__neon_mlal_padal() local
|
D | 3x8c8-minmax-neon-mull-padal.c | 175 const int32x4_t vsum1x67 = vpaddq_s32(vacc1x6, vacc1x7); in xnn_qs8_igemm_minmax_ukernel_3x8c8__neon_mull_padal() local 213 const int32x2_t vsum1x67 = vpadd_s32(vpsum1x6, vpsum1x7); in xnn_qs8_igemm_minmax_ukernel_3x8c8__neon_mull_padal() local
|
D | 3x8c16-minmax-neon-mlal-padal.c | 199 const int32x4_t vsum1x67 = vpaddq_s32(vacc1x6, vacc1x7); in xnn_qs8_igemm_minmax_ukernel_3x8c16__neon_mlal_padal() local 237 const int32x2_t vsum1x67 = vpadd_s32(vpsum1x6, vpsum1x7); in xnn_qs8_igemm_minmax_ukernel_3x8c16__neon_mlal_padal() local
|
D | 4x8c8-minmax-neon-mull-padal.c | 208 const int32x4_t vsum1x67 = vpaddq_s32(vacc1x6, vacc1x7); in xnn_qs8_igemm_minmax_ukernel_4x8c8__neon_mull_padal() local 252 const int32x2_t vsum1x67 = vpadd_s32(vpsum1x6, vpsum1x7); in xnn_qs8_igemm_minmax_ukernel_4x8c8__neon_mull_padal() local
|
D | 4x8c16-minmax-neon-mlal-padal.c | 240 const int32x4_t vsum1x67 = vpaddq_s32(vacc1x6, vacc1x7); in xnn_qs8_igemm_minmax_ukernel_4x8c16__neon_mlal_padal() local 284 const int32x2_t vsum1x67 = vpadd_s32(vpsum1x6, vpsum1x7); in xnn_qs8_igemm_minmax_ukernel_4x8c16__neon_mlal_padal() local
|
D | 3x8c8-minmax-neon-mlal-padal.c | 276 const int32x4_t vsum1x67 = vpaddq_s32(vacc1x6, vacc1x7); in xnn_qs8_igemm_minmax_ukernel_3x8c8__neon_mlal_padal() local 314 const int32x2_t vsum1x67 = vpadd_s32(vpsum1x6, vpsum1x7); in xnn_qs8_igemm_minmax_ukernel_3x8c8__neon_mlal_padal() local
|
D | 2x16c8-minmax-neon-mull-padal.c | 202 const int32x4_t vsum1x67 = vpaddq_s32(vacc1x6, vacc1x7); in xnn_qs8_igemm_minmax_ukernel_2x16c8__neon_mull_padal() local 256 const int32x2_t vsum1x67 = vpadd_s32(vpsum1x6, vpsum1x7); in xnn_qs8_igemm_minmax_ukernel_2x16c8__neon_mull_padal() local
|
D | 2x16c16-minmax-neon-mlal-padal.c | 234 const int32x4_t vsum1x67 = vpaddq_s32(vacc1x6, vacc1x7); in xnn_qs8_igemm_minmax_ukernel_2x16c16__neon_mlal_padal() local 288 const int32x2_t vsum1x67 = vpadd_s32(vpsum1x6, vpsum1x7); in xnn_qs8_igemm_minmax_ukernel_2x16c16__neon_mlal_padal() local
|
D | 4x8c8-minmax-neon-mlal-padal.c | 335 const int32x4_t vsum1x67 = vpaddq_s32(vacc1x6, vacc1x7); in xnn_qs8_igemm_minmax_ukernel_4x8c8__neon_mlal_padal() local 379 const int32x2_t vsum1x67 = vpadd_s32(vpsum1x6, vpsum1x7); in xnn_qs8_igemm_minmax_ukernel_4x8c8__neon_mlal_padal() local
|
D | 2x16c8-minmax-neon-mlal-padal.c | 341 const int32x4_t vsum1x67 = vpaddq_s32(vacc1x6, vacc1x7); in xnn_qs8_igemm_minmax_ukernel_2x16c8__neon_mlal_padal() local 395 const int32x2_t vsum1x67 = vpadd_s32(vpsum1x6, vpsum1x7); in xnn_qs8_igemm_minmax_ukernel_2x16c8__neon_mlal_padal() local
|