| Loop id | Source Location | Source Function | Level | Max Thread Time / Walltime armclang_4 (%) | Exclusive Coverage armclang_4 (%) | Inclusive Coverage armclang_4 (%) | Max Exclusive Time Over Threads armclang_4 (s) | Max Inclusive Time Over Threads armclang_4 (s) | Exclusive Time w.r.t. Wall Time armclang_4 (s) | Inclusive Time w.r.t. Wall Time armclang_4 (s) | Nb Threads armclang_4 | Vectorization Ratio (%) | Vector Length Use (%) | Speedup If No Scalar Integer | Speedup If FP Vectorized | Speedup If Fully Vectorized | Speedup If Perfect Load Balancing armclang_4 | Stride 0 | Stride 1 | Stride n | Stride Unknown | Stride Indirect | Array Access Efficiency |
|---|
| 2479 | 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 | 33.82 | 54.94 | 54.94 | 3.30 | 3.30 | 3.73 | 3.73 | 64 | 28.14 | 27.69 | 1 | 1.18 | 4.53 | 1.06 | 0 | 0 | 5 | 0 | 0 | 75.00 |
| 2376 | libggml-cpu.so - quants.c:2506-2590 [...] | ggml_vec_dot_q6_K_q8_K | Innermost | 8.85 | 14.14 | 14.14 | 0.87 | 0.87 | 0.96 | 0.96 | 64 | 50.56 | 33.98 | 1 | 1 | 2.57 | 1.08 | 2 | 0 | 1 | 19 | 0 | 55.68 |
| 2377 | libggml-cpu.so - quants.c:2492-2660 [...] | ggml_vec_dot_q6_K_q8_K | Outermost | 1.84 | 2.18 | 16.32 | 0.18 | 0.98 | 0.15 | 1.11 | 64 | 18.52 | 21.41 | 1 | 1.05 | 3.4 | 1.45 | 0 | 0 | 0 | 22 | 0 | 50.00 |
| 1840 | libggml-cpu.so - vec.h:375-751 [...] | ggml_compute_forward_flash_attn_ext | InBetween | 0.87 | 0.82 | 1.31 | 0.09 | 0.12 | 0.06 | 0.09 | 64 | 0 | 18.8 | 2.25 | 1.89 | 6.13 | 1.82 | NA | NA | NA | NA | NA | 0.00 |
| 1536 | libggml-cpu.so - ops.cpp:6220-6245 [...] | ggml_compute_forward_rope_f32(ggml_compute_params const*, ggml_tensor*, bool) | Innermost | 0.61 | 0.66 | 0.66 | 0.06 | 0.06 | 0.04 | 0.04 | 64 | 0 | 13.07 | 1 | 2 | 8.39 | 1.61 | 2 | 0 | 0 | 1 | 0 | 83.33 |
| 857 | libggml-cpu.so - vec.cpp:311-316 | ggml_vec_dot_f16 | Single | 0.72 | 0.54 | 0.54 | 0.07 | 0.07 | 0.04 | 0.04 | 64 | 50 | 50 | 1 | 1.47 | 2 | 2.27 | 2 | 0 | 0 | 0 | 0 | 100.00 |
| 1852 | libggml-cpu.so - vec.h:491-497 | ggml_compute_forward_flash_attn_ext | Innermost | 0.56 | 0.49 | 0.49 | 0.05 | 0.05 | 0.03 | 0.03 | 64 | 40 | 50 | 1 | 1 | 2 | 1.99 | 2 | 0 | 0 | 0 | 0 | 100.00 |
| 865 | libggml-cpu.so - vec.h:1045-1072 [...] | ggml_vec_swiglu_f32 | Single | 1.18 | 0.35 | 0.35 | 0.12 | 0.12 | 0.02 | 0.02 | 16 | 76.67 | 46.35 | 1.03 | 1.38 | 1.66 | 1.43 | 2 | 0 | 1 | 0 | 0 | 91.67 |
| 2480 | 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 | 0.41 | 0.28 | 55.22 | 0.04 | 3.33 | 0.02 | 3.75 | 62 | 27.59 | 27.86 | 1 | 1 | 4 | 2.43 | NA | NA | NA | NA | NA | 0.00 |
| 2447 | libggml-cpu.so - kai_lhs_quant_pack_qsi8d32p4x8sb_f32_neon.c:96-258 [...] | kai_run_lhs_quant_pack_qsi8d32p4x8sb_f32_neon | Innermost | 1.74 | 0.17 | 0.17 | 0.17 | 0.17 | 0.01 | 0.01 | 4 | 77.23 | 48.22 | 1 | 1.12 | 2.08 | 1.09 | 1 | 0 | 1 | 3 | 0 | 65.00 |
| 75 | libggml-cpu.so - ggml-cpu.c:533-2897 [...] | ggml_graph_compute_thread | Outermost | 0.36 | 0.16 | 0.16 | 0.03 | 0.03 | 0.01 | 0.01 | 50 | 0 | 19.42 | 1 | 1 | 10.67 | 2.94 | NA | NA | NA | NA | NA | 0.00 |
| 0 | libggml-cpu.so - ggml-cpu.c:3228-3229 [...] | ggml_cpu_fp32_to_fp16 | Single | 0.31 | 0.12 | 0.12 | 0.03 | 0.03 | 0.01 | 0.01 | 46 | 91.67 | 49.31 | 1 | 1.03 | 2 | 3.07 | 2 | 0 | 0 | 0 | 0 | 100.00 |
| 2462 | libggml-cpu.so - kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c:127-139 [...] | kai_run_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0 | Innermost | 4.14 | 0.11 | 0.11 | 0.41 | 0.41 | 0.01 | 0.01 | 1 | 0 | 25 | 1 | 1 | 4 | 1 | 1 | 0 | 0 | 0 | 1 | 50.00 |
| 1531 | libggml-cpu.so - ops.cpp:6446-6456 [...] | ggml_compute_forward_rope_f32(ggml_compute_params const*, ggml_tensor*, bool) | Innermost | 0.46 | 0.11 | 0.11 | 0.05 | 0.05 | 0.01 | 0.01 | 45 | 45.45 | 22.44 | 1.07 | 1.61 | 4.71 | 5.13 | 0 | 0 | 1 | 2 | 0 | 58.33 |
| 484 | libggml-cpu.so - binary-ops.cpp:18-32 [...] | ggml_compute_forward_mul | Innermost | 0.41 | 0.09 | 0.09 | 0.04 | 0.04 | 0.01 | 0.01 | 16 | 0 | 11.84 | 1 | 1 | 10.67 | 1.94 | 0 | 0 | 0 | 3 | 0 | 50.00 |
| 380 | libggml-cpu.so - binary-ops.cpp:10-32 [...] | ggml_compute_forward_add_non_quantized | Innermost | 0.46 | 0.09 | 0.09 | 0.04 | 0.04 | 0.01 | 0.01 | 16 | 0 | 12.5 | 1 | 1 | 10.67 | 2.18 | 0 | 0 | 0 | 3 | 0 | 50.00 |
| 2463 | libggml-cpu.so - kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c:115-118 | kai_run_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0 | Innermost | 2.46 | 0.07 | 0.07 | 0.24 | 0.24 | 0.00 | 0.00 | 1 | 0 | 15.63 | 1 | 1 | 10.35 | 1 | 0 | 0 | 0 | 1 | 1 | 25.00 |
| 1527 | libggml-cpu.so - ops.cpp:6210-6462 [...] | ggml_compute_forward_rope_f32(ggml_compute_params const*, ggml_tensor*, bool) | InBetween | 0.15 | 0.04 | 0.70 | 0.01 | 0.07 | 0.00 | 0.05 | 28 | 0 | 19.86 | 5.92 | 1.47 | 6.07 | 2.62 | NA | NA | NA | NA | NA | 0.00 |
| 58 | libggml-cpu.so - ggml-cpu.c:1163-1198 [...] | ggml_compute_forward_mul_mat | InBetween | 0.15 | 0.04 | 0.08 | 0.02 | 0.02 | 0.00 | 0.01 | 26 | 0 | 24.36 | 1 | 1 | 2 | 2.44 | NA | NA | NA | NA | NA | 0.00 |
| 54 | libggml-cpu.so - ggml-cpu.c:1183-1194 [...] | ggml_compute_forward_mul_mat | Innermost | 0.10 | 0.03 | 0.03 | 0.01 | 0.01 | 0.00 | 0.00 | 21 | 0 | 25 | 1 | 1 | 4 | 1.83 | 1 | 0 | 0 | 0 | 0 | 100.00 |
| 1294 | libggml-cpu.so - ops.cpp:4325-4326 | ggml_compute_forward_rms_norm | Innermost | 0.15 | 0.03 | 0.03 | 0.02 | 0.02 | 0.00 | 0.00 | 11 | 16.67 | 27.08 | 1 | 1.25 | 4 | 1.57 | 0 | 0 | 1 | 0 | 0 | 75.00 |
| 1858 | libggml-cpu.so - vec.h:677-682 | ggml_compute_forward_flash_attn_ext | Innermost | 0.10 | 0.03 | 0.03 | 0.01 | 0.01 | 0.00 | 0.00 | 16 | 50 | 50 | 1 | 1.13 | 2 | 1.52 | 1 | 0 | 0 | 0 | 0 | 100.00 |
| 1837 | libggml-cpu.so - vec.h:375-751 [...] | ggml_compute_forward_flash_attn_ext | InBetween | 0.10 | 0.02 | 1.36 | 0.01 | 0.13 | 0.00 | 0.09 | 13 | 0 | 24.68 | 2.01 | 1.34 | 1.34 | 1.86 | NA | NA | NA | NA | NA | 0.00 |
| 1528 | libggml-cpu.so - ops.cpp:6210-6484 [...] | ggml_compute_forward_rope_f32(ggml_compute_params const*, ggml_tensor*, bool) | Outermost | 0.05 | 0.02 | 0.83 | 0.01 | 0.09 | 0.00 | 0.06 | 14 | 0 | 20.94 | 1 | 1 | 5.05 | 1 | NA | NA | NA | NA | NA | 0.00 |
| 53 | libggml-cpu.so - ggml-cpu.c:1125-1395 [...] | ggml_compute_forward_mul_mat | Innermost | 0.10 | 0.02 | 0.02 | 0.01 | 0.01 | 0.00 | 0.00 | 12 | 0 | 23.86 | 1 | 1 | 2 | 1.71 | NA | NA | NA | NA | NA | 0.00 |
| 2461 | libggml-cpu.so - kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c:107-154 [...] | kai_run_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0 | Outermost | 0.56 | 0.02 | 0.19 | 0.06 | 0.70 | 0.00 | 0.01 | 1 | 0 | 23.13 | 1 | 1 | 2 | 1 | NA | NA | NA | NA | NA | 0.00 |
| 347 | libggml-cpu.so - traits.cpp:13-17 [...] | ggml_cpu_extra_compute_forward | Single | 0.15 | 0.01 | 0.01 | 0.01 | 0.01 | 0.00 | 0.00 | 6 | 0 | 25 | 1 | 1 | 4 | 2.25 | 0 | 0 | 1 | 2 | 0 | 62.19 |
| 1295 | libggml-cpu.so - vec.h:677-682 | ggml_compute_forward_rms_norm | Innermost | 0.10 | 0.01 | 0.01 | 0.01 | 0.01 | 0.00 | 0.00 | 7 | 50 | 50 | 1 | 1.13 | 2 | 1.75 | 1 | 0 | 0 | 0 | 0 | 100.00 |