| /external/XNNPACK/src/qs8-gemm/gen/ |
| D | 1x8c8-minmax-neon-mull-padal.c | 93 int32x4_t vacc0x4567 = vpaddq_s32(vsum0x45, vsum0x67); in xnn_qs8_gemm_minmax_ukernel_1x8c8__neon_mull_padal() local
108 int32x4_t vacc0x4567 = vcombine_s32(vsum0x45, vsum0x67 ); in xnn_qs8_gemm_minmax_ukernel_1x8c8__neon_mull_padal() local
|
| D | 1x8c16-minmax-neon-mlal-padal.c | 101 int32x4_t vacc0x4567 = vpaddq_s32(vsum0x45, vsum0x67); in xnn_qs8_gemm_minmax_ukernel_1x8c16__neon_mlal_padal() local
116 int32x4_t vacc0x4567 = vcombine_s32(vsum0x45, vsum0x67 ); in xnn_qs8_gemm_minmax_ukernel_1x8c16__neon_mlal_padal() local
|
| D | 1x8c8-minmax-neon-mlal-padal.c | 142 int32x4_t vacc0x4567 = vpaddq_s32(vsum0x45, vsum0x67); in xnn_qs8_gemm_minmax_ukernel_1x8c8__neon_mlal_padal() local
157 int32x4_t vacc0x4567 = vcombine_s32(vsum0x45, vsum0x67 ); in xnn_qs8_gemm_minmax_ukernel_1x8c8__neon_mlal_padal() local
|
| /external/XNNPACK/src/qs8-igemm/gen/ |
| D | 1x8c8-minmax-neon-mull-padal.c | 107 int32x4_t vacc0x4567 = vpaddq_s32(vsum0x45, vsum0x67); in xnn_qs8_igemm_minmax_ukernel_1x8c8__neon_mull_padal() local
122 int32x4_t vacc0x4567 = vcombine_s32(vsum0x45, vsum0x67 ); in xnn_qs8_igemm_minmax_ukernel_1x8c8__neon_mull_padal() local
|
| D | 1x8c16-minmax-neon-mlal-padal.c | 115 int32x4_t vacc0x4567 = vpaddq_s32(vsum0x45, vsum0x67); in xnn_qs8_igemm_minmax_ukernel_1x8c16__neon_mlal_padal() local
130 int32x4_t vacc0x4567 = vcombine_s32(vsum0x45, vsum0x67 ); in xnn_qs8_igemm_minmax_ukernel_1x8c16__neon_mlal_padal() local
|
| /external/XNNPACK/src/f32-gemm/gen/ |
| D | 1x8-minmax-sse-load1.c | 43 __m128 vacc0x4567 = _mm_load_ps(w + 4); in xnn_f32_gemm_minmax_ukernel_1x8__sse_load1() local
|
| D | 1x8-minmax-wasmsimd-arm-loadsplat.c | 45 v128_t vacc0x4567 = wasm_v128_load(w + 4); in xnn_f32_gemm_minmax_ukernel_1x8__wasmsimd_arm_loadsplat() local
|
| D | 1x8-minmax-wasmsimd-x86-loadsplat.c | 43 v128_t vacc0x4567 = wasm_v128_load(w + 4); in xnn_f32_gemm_minmax_ukernel_1x8__wasmsimd_x86_loadsplat() local
|
| D | 1x8-minmax-neon-lane-ld64.c | 44 float32x4_t vacc0x4567 = vld1q_f32(w); w += 4; in xnn_f32_gemm_minmax_ukernel_1x8__neon_lane_ld64() local
|
| D | 1x8-minmax-neonfma-lane-ld64.c | 44 float32x4_t vacc0x4567 = vld1q_f32(w); w += 4; in xnn_f32_gemm_minmax_ukernel_1x8__neonfma_lane_ld64() local
|
| D | 1x8-minmax-neonfma-dup-ld64.c | 44 float32x4_t vacc0x4567 = vld1q_f32(w); w += 4; in xnn_f32_gemm_minmax_ukernel_1x8__neonfma_dup_ld64() local
|
| D | 1x8-minmax-neon-dup-ld64.c | 44 float32x4_t vacc0x4567 = vld1q_f32(w); w += 4; in xnn_f32_gemm_minmax_ukernel_1x8__neon_dup_ld64() local
|
| D | 1x8s4-minmax-wasmsimd-x86.c | 43 v128_t vacc0x4567 = wasm_v128_load(w + 4); in xnn_f32_gemm_minmax_ukernel_1x8s4__wasmsimd_x86() local
|
| /external/XNNPACK/src/f32-gemm/gen-inc/ |
| D | 1x8inc-minmax-wasmsimd-arm-loadsplat.c | 47 v128_t vacc0x4567 = wasm_v128_load(acc + 4); in xnn_f32_gemminc_minmax_ukernel_1x8__wasmsimd_arm_loadsplat() local
|
| D | 1x8inc-minmax-wasmsimd-x86-loadsplat.c | 45 v128_t vacc0x4567 = wasm_v128_load(acc + 4); in xnn_f32_gemminc_minmax_ukernel_1x8__wasmsimd_x86_loadsplat() local
|
| D | 1x8inc-minmax-sse-load1.c | 45 __m128 vacc0x4567 = _mm_load_ps(acc + 4); in xnn_f32_gemminc_minmax_ukernel_1x8__sse_load1() local
|
| D | 1x8inc-minmax-neonfma-lane-ld64.c | 46 float32x4_t vacc0x4567 = vld1q_f32(acc); acc += 4; in xnn_f32_gemminc_minmax_ukernel_1x8__neonfma_lane_ld64() local
|
| D | 1x8inc-minmax-neon-lane-ld64.c | 46 float32x4_t vacc0x4567 = vld1q_f32(acc); acc += 4; in xnn_f32_gemminc_minmax_ukernel_1x8__neon_lane_ld64() local
|
| D | 1x8inc-minmax-neon-dup-ld64.c | 46 float32x4_t vacc0x4567 = vld1q_f32(acc); acc += 4; in xnn_f32_gemminc_minmax_ukernel_1x8__neon_dup_ld64() local
|
| D | 1x8inc-minmax-neonfma-dup-ld64.c | 46 float32x4_t vacc0x4567 = vld1q_f32(acc); acc += 4; in xnn_f32_gemminc_minmax_ukernel_1x8__neonfma_dup_ld64() local
|
| /external/XNNPACK/src/f32-igemm/gen/ |
| D | 1x8-minmax-wasmsimd-x86-loadsplat.c | 47 v128_t vacc0x4567 = wasm_v128_load(w + 4); in xnn_f32_igemm_minmax_ukernel_1x8__wasmsimd_x86_loadsplat() local
|
| D | 1x8-minmax-sse-load1.c | 47 __m128 vacc0x4567 = _mm_load_ps(w + 4); in xnn_f32_igemm_minmax_ukernel_1x8__sse_load1() local
|
| D | 1x8-minmax-wasmsimd-arm-loadsplat.c | 49 v128_t vacc0x4567 = wasm_v128_load(w + 4); in xnn_f32_igemm_minmax_ukernel_1x8__wasmsimd_arm_loadsplat() local
|
| D | 1x8-minmax-neon-lane-ld64.c | 48 float32x4_t vacc0x4567 = vld1q_f32(w); w += 4; in xnn_f32_igemm_minmax_ukernel_1x8__neon_lane_ld64() local
|
| /external/XNNPACK/src/f32-prelu/gen/ |
| D | wasmsimd-minmax-1x8.c | 52 v128_t vacc0x4567 = wasm_i32x4_max(vi0x4567, vzero); in xnn_f32_prelu_ukernel__wasmsimd_minmax_1x8() local
|