| Loop id | Source Location | Source Function | Level | Max Thread Time / Walltime gcc_5 (%) | Exclusive Coverage gcc_5 (%) | Inclusive Coverage gcc_5 (%) | Max Exclusive Time Over Threads gcc_5 (s) | Max Inclusive Time Over Threads gcc_5 (s) | Exclusive Time w.r.t. Wall Time gcc_5 (s) | Inclusive Time w.r.t. Wall Time gcc_5 (s) | Nb Threads gcc_5 | Vectorization Ratio (%) | Vector Length Use (%) | Speedup If No Scalar Integer | Speedup If FP Vectorized | Speedup If Fully Vectorized | Speedup If Perfect Load Balancing gcc_5 | Stride 0 | Stride 1 | Stride n | Stride Unknown | Stride Indirect | Array Access Efficiency |
|---|
| 2097 | libggml-cpu.so - kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c:131-131 | kai_run_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm | Innermost | 29.12 | 47.63 | 47.63 | 2.27 | 2.27 | 2.42 | 2.42 | 95 | 28.14 | 55.37 | 1 | 1 | 1.85 | 1.11 | 0 | 0 | 5 | 0 | 0 | 75.00 |
| 1960 | libggml-cpu.so - quants.c:2506-2575 [...] | ggml_vec_dot_q6_K_q8_K | Innermost | 6.73 | 11.14 | 11.14 | 0.53 | 0.53 | 0.57 | 0.57 | 96 | 49.44 | 66.92 | 1.06 | 1 | 1.26 | 1.11 | 3 | 4 | 2 | 0 | 0 | 94.44 |
| 1961 | libggml-cpu.so - quants.c:2492-2654 [...] | ggml_vec_dot_q6_K_q8_K | InBetween | 1.35 | 1.64 | 12.78 | 0.11 | 0.58 | 0.08 | 0.65 | 96 | 59.32 | 72.09 | 1.05 | 1 | 1.17 | 1.51 | 1 | 4 | 2 | 2 | 0 | 83.33 |
| 726 | libggml-cpu.so - ops.cpp:6238-6245 | ggml_rope_cache_init(float, float, float const*, float*, long, float, float, float*, float, float) | Innermost | 0.64 | 0.60 | 0.60 | 0.05 | 0.05 | 0.03 | 0.03 | 94 | 0 | 29.85 | 1 | 2.22 | 1.73 | 1.92 | 0 | 0 | 0 | 1 | 0.75 | 31.25 |
| 1493 | libggml-cpu.so - ops.cpp:8817-8881 [...] | ggml_compute_forward_flash_attn_ext | InBetween | 0.71 | 0.56 | 0.89 | 0.06 | 0.08 | 0.03 | 0.05 | 86 | 6.63 | 38.53 | 2.96 | 1.06 | 1.29 | 2.05 | NA | NA | NA | NA | NA | 0.00 |
| 712 | libggml-cpu.so - vec.cpp:231-262 | ggml_vec_dot_f16 | Single | 0.58 | 0.44 | 0.44 | 0.05 | 0.05 | 0.02 | 0.02 | 86 | 100 | 100 | 1 | 1 | 1 | 2.13 | 0 | 0 | 0 | 16 | 0 | 50.00 |
| 725 | libggml-cpu.so - ops.cpp:6211-6231 [...] | rope_yarn(float, float, float*, long, float, float, float*, float*) | Single | 0.51 | 0.41 | 0.41 | 0.04 | 0.04 | 0.02 | 0.02 | 92 | 0 | 24.77 | 2.71 | 2.3 | 4.75 | 2.22 | NA | NA | NA | NA | NA | 0.00 |
| 1495 | libggml-cpu.so - vec.h:411-458 | ggml_compute_forward_flash_attn_ext | Innermost | 0.51 | 0.33 | 0.33 | 0.04 | 0.04 | 0.02 | 0.02 | 83 | 96.97 | 96.99 | 1.24 | 1 | 1 | 2.48 | 1 | 0 | 0 | 16 | 0 | 52.94 |
| 714 | libggml-cpu.so - vec.cpp:385-387 [...] | ggml_vec_swiglu_f32 | Single | 1.73 | 0.29 | 0.29 | 0.13 | 0.13 | 0.01 | 0.01 | 16 | 91.43 | 97.34 | 1 | 1 | 1.03 | 1.8 | 0 | 0 | 0 | 3 | 0 | 50.00 |
| 2066 | libggml-cpu.so - kai_lhs_quant_pack_qsi8d32p4x8sb_f32_neon.c:96-262 [...] | kai_run_lhs_quant_pack_qsi8d32p4x8sb_f32_neon | Innermost | 2.63 | 0.18 | 0.18 | 0.20 | 0.20 | 0.01 | 0.01 | 4 | 76.21 | 96.75 | 1 | 1 | 1.03 | 1.12 | 0 | 0 | 6 | 0 | 0 | 75.00 |
| 2098 | libggml-cpu.so - kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c:131-131 | kai_run_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm | InBetween | 0.32 | 0.16 | 47.78 | 0.03 | 2.25 | 0.01 | 2.43 | 66 | 27.59 | 55.71 | 1 | 1 | 1.62 | 2.58 | NA | NA | NA | NA | NA | 0.00 |
| 362 | libggml-cpu.so - binary-ops.cpp:10-32 [...] | ggml_compute_forward_add_non_quantized | Innermost | 0.58 | 0.11 | 0.11 | 0.04 | 0.04 | 0.01 | 0.01 | 16 | 0 | 18.75 | 1 | 1 | 5.33 | 1.67 | 0 | 0 | 3 | 0 | 0 | 75.00 |
| 440 | libggml-cpu.so - binary-ops.cpp:18-32 [...] | ggml_compute_forward_mul | Innermost | 0.51 | 0.08 | 0.08 | 0.04 | 0.04 | 0.00 | 0.00 | 16 | 0 | 18.75 | 1 | 1 | 5.33 | 2.06 | 0 | 0 | 3 | 0 | 0 | 75.00 |
| 1959 | libggml-cpu.so - quants.c:2486-2923 [...] | ggml_vec_dot_q6_K_q8_K | Outermost | 0.13 | 0.07 | 12.85 | 0.01 | 0.58 | 0.00 | 0.65 | 45 | 4.55 | 42.13 | 1.69 | 1 | 1.72 | 1.5 | NA | NA | NA | NA | NA | 0.00 |
| 2079 | libggml-cpu.so - kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c:127-139 [...] | kai_run_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0 | Innermost | 3.78 | 0.07 | 0.07 | 0.29 | 0.29 | 0.00 | 0.00 | 1 | 0 | 50 | 1 | 1 | 2 | 1 | 0 | 0 | 1 | 1 | 0 | 62.50 |
| 4 | libggml-cpu.so - ggml-cpu.c:3228-3229 [...] | ggml_cpu_fp32_to_fp16 | Single | 0.19 | 0.07 | 0.07 | 0.02 | 0.02 | 0.00 | 0.00 | 42 | 82.05 | 93.27 | 1 | 1 | 1 | 2.17 | 0 | 0 | 0 | 3 | 0 | 50.00 |
| 732 | libggml-cpu.so - ops.cpp:6389-6475 [...] | ggml_compute_forward_rope_f32(ggml_compute_params const*, ggml_tensor*, bool) | InBetween | 0.26 | 0.06 | 0.07 | 0.02 | 0.02 | 0.00 | 0.00 | 34 | 0 | 25.53 | 1.71 | 1.27 | 4.43 | 2.78 | NA | NA | NA | NA | NA | 0.00 |
| 60 | libggml-cpu.so - ggml-cpu.c:1664-2898 [...] | ggml_graph_compute_thread | Outermost | 0.19 | 0.06 | 0.07 | 0.02 | 0.02 | 0.00 | 0.00 | 35 | 0 | 35.8 | 1 | 1 | 4 | 2.28 | NA | NA | NA | NA | NA | 0.00 |
| 2075 | libggml-cpu.so - kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c:107-153 [...] | kai_run_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0 | InBetween | 2.82 | 0.05 | 0.05 | 0.22 | 0.22 | 0.00 | 0.00 | 1 | 0 | 44.33 | 1 | 1 | 2.14 | 1 | NA | NA | NA | NA | NA | 0.00 |
| 1124 | libggml-cpu.so - ops.cpp:4325-4326 | ggml_compute_forward_rms_norm | Innermost | 0.32 | 0.05 | 0.05 | 0.03 | 0.03 | 0.00 | 0.00 | 16 | 100 | 82.5 | 1 | 1 | 1.16 | 2 | 0 | 0 | 0 | 1 | 0 | 50.00 |
| 742 | libggml-cpu.so - ops.cpp:6446-6457 | ggml_compute_forward_rope_f32(ggml_compute_params const*, ggml_tensor*, bool) | Innermost | 0.19 | 0.05 | 0.05 | 0.02 | 0.02 | 0.00 | 0.00 | 28 | 0 | 18.91 | 1 | 1 | 5.22 | 2.21 | 0 | 0 | 4 | 4 | 0 | 62.50 |
| 1499 | libggml-cpu.so - vec.h:646-653 | ggml_compute_forward_flash_attn_ext | Innermost | 0.19 | 0.04 | 0.04 | 0.01 | 0.01 | 0.00 | 0.00 | 25 | 100 | 100 | 1 | 1 | 1 | 2.59 | 0 | 0 | 0 | 2 | 0 | 50.00 |
| 53 | libggml-cpu.so - ggml-cpu.c:1193-1194 | ggml_compute_forward_mul_mat | Innermost | 0.13 | 0.03 | 0.03 | 0.01 | 0.01 | 0.00 | 0.00 | 20 | 0 | 44.55 | 1 | 1 | 2.73 | 1.82 | 1 | 0 | 0 | 0 | 0 | 100.00 |
| 2074 | libggml-cpu.so - kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c:126-154 [...] | kai_run_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0 | InBetween | 1.28 | 0.02 | 0.10 | 0.10 | 0.39 | 0.00 | 0.00 | 1 | 0 | 40.42 | 1 | 1 | 2.92 | 1 | NA | NA | NA | NA | NA | 0.00 |
| 710 | libggml-cpu.so - vec.cpp:224-337 [...] | ggml_vec_dot_f16 | Outermost | 0.13 | 0.02 | 0.02 | 0.01 | 0.01 | 0.00 | 0.00 | 16 | 41.67 | 63.12 | 1.55 | 1 | 1.35 | 1.78 | NA | NA | NA | NA | NA | 0.00 |
| 1128 | libggml-cpu.so - vec.h:646-653 | ggml_compute_forward_rms_norm | Innermost | 0.19 | 0.01 | 0.01 | 0.02 | 0.02 | 0.00 | 0.00 | 10 | 100 | 100 | 1 | 1 | 1 | 2.5 | 0 | 0 | 0 | 2 | 0 | 50.00 |
| 1491 | libggml-cpu.so - simd-mappings.h:51-51 [...] | ggml_compute_forward_flash_attn_ext | Single | 0.06 | 0.01 | 0.01 | 0.01 | 0.01 | 0.00 | 0.00 | 12 | 0 | 34.72 | 1.25 | 1.19 | 3.08 | 1 | 0 | 0 | 0 | 2 | 0 | 50.00 |
| 735 | libggml-cpu.so - ops.cpp:6389-6490 [...] | ggml_compute_forward_rope_f32(ggml_compute_params const*, ggml_tensor*, bool) | Outermost | 0.06 | 0.01 | 0.13 | 0.01 | 0.02 | 0.00 | 0.01 | 12 | 0 | 25.52 | 1.42 | 1.21 | 4.12 | 1 | NA | NA | NA | NA | NA | 0.00 |
| 51 | libggml-cpu.so - ggml-cpu.c:1125-1395 [...] | ggml_compute_forward_mul_mat | Innermost | 0.06 | 0.01 | 0.01 | 0.01 | 0.01 | 0.00 | 0.00 | 11 | 0 | 47.55 | 1 | 1 | 2 | 1 | NA | NA | NA | NA | NA | 0.00 |
| 1490 | libggml-cpu.so - ops.cpp:8778-8939 [...] | ggml_compute_forward_flash_attn_ext | InBetween | 0.06 | 0.01 | 0.06 | 0.01 | 0.02 | 0.00 | 0.00 | 11 | 5.84 | 41.94 | 40.28 | 1.01 | 1.61 | 1 | NA | NA | NA | NA | NA | 0.00 |
| 47 | libggml-cpu.so - ggml-cpu.c:1164-1198 [...] | ggml_compute_forward_mul_mat | InBetween | 0.06 | 0.01 | 0.04 | 0.01 | 0.01 | 0.00 | 0.00 | 11 | 0 | 47.74 | 1 | 1 | 2 | 1 | NA | NA | NA | NA | NA | 0.00 |
| 2037 | libggml-cpu.so - kleidiai.cpp:535-547 [...] | ggml::cpu::kleidiai::extra_buffer_type::get_tensor_traits(ggml_tensor const*) | Single | 0.13 | 0.01 | 0.01 | 0.01 | 0.01 | 0.00 | 0.00 | 9 | 0 | 50 | 1 | 1 | 2 | 1.8 | NA | NA | NA | NA | NA | 0.00 |