Loops
kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c: 131 - 166.98 %
| Run orig_default | Run gcc_default | Run gcc_4 | |||||||||||||||
| Loop Source Regions |
| Loop Source Regions |
| Loop Source Regions |
| ||||||||||||
| ASM Loop ID | Max Time Over Threads (s) | Time w.r.t. Wall Time (s) | Cov (%) | Vect. Ratio (%) | Vector Length Use (%) | ASM Loop ID | Max Time Over Threads (s) | Time w.r.t. Wall Time (s) | Cov (%) | Vect. Ratio (%) | Vector Length Use (%) | ASM Loop ID | Max Time Over Threads (s) | Time w.r.t. Wall Time (s) | Cov (%) | Vect. Ratio (%) | Vector Length Use (%) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2423 | 1.93 | 2.32 | 53.70 | 32.79 | 30.39 | 2089 | 1.94 | 2.49 | 55.43 | 32.79 | 30.39 | 2101 | 1.97 | 2.56 | 55.61 | 32.79 | 30.39 |
| 2426 | 0.04 | 0.03 | 0.72 | 28.14 | 27.69 | 2092 | 0.04 | 0.03 | 0.77 | 28.14 | 27.69 | 2104 | 0.04 | 0.03 | 0.74 | 28.14 | 27.69 |
| Sum on 2 analyzed binary loops (libggml-cpu.so - 2423, libggml-cpu.so - 2426) | Sum on 2 analyzed binary loops (libggml-cpu.so - 2089, libggml-cpu.so - 2092) | Sum on 2 analyzed binary loops (libggml-cpu.so - 2101, libggml-cpu.so - 2104) | |||||||||||||||
| Analysis | Count | Analysis | Count | Analysis | Count | ||||||||||||
| Data Access Issues | Data Access Issues | Data Access Issues | |||||||||||||||
| Presence of constant non-unit stride data access | 1 | Presence of constant non-unit stride data access | 1 | Presence of constant non-unit stride data access | 1 | ||||||||||||
| Vectorization Roadblocks | Vectorization Roadblocks | Vectorization Roadblocks | |||||||||||||||
| Presence of constant non-unit stride data access | 1 | Presence of constant non-unit stride data access | 1 | Presence of constant non-unit stride data access | 1 | ||||||||||||
quants.c: 2506 - 20.21 %
| Run orig_default | Run gcc_default | Run gcc_4 | |||||||||||||||
| Loop Source Regions |
| Loop Source Regions |
| Loop Source Regions |
| ||||||||||||
| ASM Loop ID | Max Time Over Threads (s) | Time w.r.t. Wall Time (s) | Cov (%) | Vect. Ratio (%) | Vector Length Use (%) | ASM Loop ID | Max Time Over Threads (s) | Time w.r.t. Wall Time (s) | Cov (%) | Vect. Ratio (%) | Vector Length Use (%) | ASM Loop ID | Max Time Over Threads (s) | Time w.r.t. Wall Time (s) | Cov (%) | Vect. Ratio (%) | Vector Length Use (%) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2308 | 0.33 | 0.29 | 6.67 | 50.56 | 33.98 | 1968 | 0.25 | 0.29 | 6.56 | 49.18 | 36.04 | 1986 | 0.28 | 0.32 | 6.98 | 47.34 | 35.41 |
| Sum on 1 analyzed binary loop (libggml-cpu.so - 2308) | Sum on 1 analyzed binary loop (libggml-cpu.so - 1968) | Sum on 1 analyzed binary loop (libggml-cpu.so - 1986) | |||||||||||||||
| Analysis | Count | Analysis | Count | Analysis | Count | ||||||||||||
| Data Access Issues | Data Access Issues | Data Access Issues | |||||||||||||||
| Presence of constant non-unit stride data access | 1 | Presence of constant non-unit stride data access | 1 | Presence of constant non-unit stride data access | 1 | ||||||||||||
| Vectorization Roadblocks | Vectorization Roadblocks | Vectorization Roadblocks | |||||||||||||||
| Presence of constant non-unit stride data access | 1 | Presence of constant non-unit stride data access | 1 | Presence of constant non-unit stride data access | 1 | ||||||||||||
kai_lhs_quant_pack_qsi8d32p4x8sb_f32_neon.c: 271 - 0.28 %
| Run orig_default | Run gcc_default | Run gcc_4 | |||||||||||||||
| Loop Source Regions |
| Loop Source Regions |
| Loop Source Regions |
| ||||||||||||
| ASM Loop ID | Max Time Over Threads (s) | Time w.r.t. Wall Time (s) | Cov (%) | Vect. Ratio (%) | Vector Length Use (%) | ASM Loop ID | Max Time Over Threads (s) | Time w.r.t. Wall Time (s) | Cov (%) | Vect. Ratio (%) | Vector Length Use (%) | ASM Loop ID | Max Time Over Threads (s) | Time w.r.t. Wall Time (s) | Cov (%) | Vect. Ratio (%) | Vector Length Use (%) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2387 | 0.16 | 0.00 | 0.08 | 64.41 | 42.29 | 2057 | 0.20 | 0.00 | 0.10 | 59.85 | 43.48 | 2070 | 0.22 | 0.00 | 0.11 | 59.85 | 43.48 |
| No Optimizer analysis found for any assembly loop. More loops can be analyzed using option --optimizer-loop-count. | No Optimizer analysis found for any assembly loop. More loops can be analyzed using option --optimizer-loop-count. | No Optimizer analysis found for any assembly loop. More loops can be analyzed using option --optimizer-loop-count. | |||||||||||||||
| Analysis | Count | Analysis | Count | Analysis | Count | ||||||||||||
ggml-cpu.c: 1125 - 0.14 %
| Run orig_default | Run gcc_default | Run gcc_4 | |||||||||||||||
| Loop Source Regions |
| Loop Source Regions |
| Loop Source Regions |
| ||||||||||||
| ASM Loop ID | Max Time Over Threads (s) | Time w.r.t. Wall Time (s) | Cov (%) | Vect. Ratio (%) | Vector Length Use (%) | ASM Loop ID | Max Time Over Threads (s) | Time w.r.t. Wall Time (s) | Cov (%) | Vect. Ratio (%) | Vector Length Use (%) | ASM Loop ID | Max Time Over Threads (s) | Time w.r.t. Wall Time (s) | Cov (%) | Vect. Ratio (%) | Vector Length Use (%) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 65 | 0.01 | 0.00 | 0.05 | 0 | 23.77 | 48 | 0.01 | 0.00 | 0.04 | 0 | 24.36 | 55 | 0.01 | 0.00 | 0.05 | 0 | 23.99 |
| No Optimizer analysis found for any assembly loop. More loops can be analyzed using option --optimizer-loop-count. | No Optimizer analysis found for any assembly loop. More loops can be analyzed using option --optimizer-loop-count. | No Optimizer analysis found for any assembly loop. More loops can be analyzed using option --optimizer-loop-count. | |||||||||||||||
| Analysis | Count | Analysis | Count | Analysis | Count | ||||||||||||
ggml-cpu.c: 3228 - 0.10 %
| Run orig_default | Run gcc_default | Run gcc_4 | |||||||||||||||
| Loop Source Regions |
| Loop Source Regions |
| Loop Source Regions |
| ||||||||||||
| ASM Loop ID | Max Time Over Threads (s) | Time w.r.t. Wall Time (s) | Cov (%) | Vect. Ratio (%) | Vector Length Use (%) | ASM Loop ID | Max Time Over Threads (s) | Time w.r.t. Wall Time (s) | Cov (%) | Vect. Ratio (%) | Vector Length Use (%) | ASM Loop ID | Max Time Over Threads (s) | Time w.r.t. Wall Time (s) | Cov (%) | Vect. Ratio (%) | Vector Length Use (%) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 0.02 | 0.00 | 0.02 | 86.36 | 95.61 | 4 | 0.01 | 0.00 | 0.03 | 93.18 | 94.03 | 1 | 0.01 | 0.00 | 0.05 | 72.6 | 41.78 |
| No Optimizer analysis found for any assembly loop. More loops can be analyzed using option --optimizer-loop-count. | No Optimizer analysis found for any assembly loop. More loops can be analyzed using option --optimizer-loop-count. | No Optimizer analysis found for any assembly loop. More loops can be analyzed using option --optimizer-loop-count. | |||||||||||||||
| Analysis | Count | Analysis | Count | Analysis | Count | ||||||||||||
ops.cpp: 6446 - 0.08 %
| Run orig_default | Run gcc_default | Run gcc_4 | |||||||||||||||
| Loop Source Regions |
| Loop Source Regions |
| Loop Source Regions |
| ||||||||||||
| ASM Loop ID | Max Time Over Threads (s) | Time w.r.t. Wall Time (s) | Cov (%) | Vect. Ratio (%) | Vector Length Use (%) | ASM Loop ID | Max Time Over Threads (s) | Time w.r.t. Wall Time (s) | Cov (%) | Vect. Ratio (%) | Vector Length Use (%) | ASM Loop ID | Max Time Over Threads (s) | Time w.r.t. Wall Time (s) | Cov (%) | Vect. Ratio (%) | Vector Length Use (%) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1437 | 0.01 | 0.00 | 0.04 | 0 | 12.5 | 799 | 0.01 | 0.00 | 0.03 | 37.5 | 40.63 | 848 | 0.01 | 0.00 | 0.01 | 37.5 | 40.63 |
| No Optimizer analysis found for any assembly loop. More loops can be analyzed using option --optimizer-loop-count. | No Optimizer analysis found for any assembly loop. More loops can be analyzed using option --optimizer-loop-count. | No Optimizer analysis found for any assembly loop. More loops can be analyzed using option --optimizer-loop-count. | |||||||||||||||
| Analysis | Count | Analysis | Count | Analysis | Count | ||||||||||||