• Global
  • Configurations
  • Environment Variables
  • Outputs
  • Logs
  • Tools Status
• Summary
• Application
• Functions
• Loops
• Topology
  • lstopo
  • lstopo_PU
  • lstopo_threads

exec - 2025-11-24 14:47:56 - MAQAO 2025.1.3

Help is available by moving the cursor above any symbol or by checking MAQAO website.

Global Metrics

Total Time (s)		2.60
Max (Thread Active Time) (s)		1.10
Average Active Time (s)		0.45
Activity Ratio (%)		57.2
Average number of active threads		16.790
Affinity Stability (%)		65.9
Time in analyzed loops (%)		53.4
Time in analyzed innermost loops (%)		52.1
Time in user code (%)		53.7
Compilation Options Score (%)		99.0
Array Access Efficiency (%)		74.1
Potential Speedups
Perfect Flow Complexity		1.00
Perfect OpenMP/MPI/Pthread/TBB		1.06
Perfect OpenMP/MPI/Pthread/TBB + Perfect Load Distribution		4.41
No Scalar Integer	Potential Speedup	1.01
	Nb Loops to get 80%	4
FP Vectorised	Potential Speedup	1.00
	Nb Loops to get 80%	2
Fully Vectorised	Potential Speedup	1.32
	Nb Loops to get 80%	1
FP Arithmetic Only	Potential Speedup	1.70
	Nb Loops to get 80%	1

CQA Potential Speedups Summary

Average Active Threads Count⏎

Loop Based Profile⏎

Innermost Loop Based Profile⏎

Application Categorization⏎

Compilation Options⏎

Source Object	Issue
▼libllama.so–
○hashtable.h	-mcpu=native is missing.
○stl_pair.h	-mcpu=native is missing.
○unicode.cpp	-mcpu=native is missing.
○hashtable_policy.h	-mcpu=native is missing.
○llama-vocab.cpp	-mcpu=native is missing.
▼libggml-cpu.so–
○vec.cpp
○binary-ops.cpp
○kleidiai.cpp
○kai_lhs_quant_pack_qsi8d32p4x8sb_f32_neon.c
○kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c
○ggml-cpu.c
○kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c
○ops.cpp
▼[vdso]–
▼–
○	-g is missing for some functions (possibly ones added by the compiler), it is needed to have more accurate reports. Other recommended flags are: -O2/-O3, -march=(target)
○	-O2, -O3 or -Ofast is missing.
○	-mcpu=native is missing.

Loop Path Count Profile⏎

Cumulated Speedup If No Scalar Integer⏎

Cumulated Speedup If FP Vectorized⏎

Cumulated Speedup If Fully Vectorized⏎

Cumulated Speedup If FP Arithmetic Only⏎