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

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

Global Metrics

Metric		r0	r1	r2	r3	r4	r5	r6	r7	r8	r9	r10
Total Time (s)		178.45	97.06	50.63	28.44	21.42	19.75	18.38	17.41	16.80	17.08	17.04
Max (Thread Active Time) (s)		177.65	96.25	49.78	27.62	20.47	18.77	17.60	16.58	15.99	16.18	16.11
Average Active Time (s)		177.65	96.06	49.56	27.24	20.14	18.36	17.12	16.17	15.52	15.74	15.71
Activity Ratio (%)		99.6	99.4	99.0	98.1	97.3	96.6	97.1	97.1	96.8	96.5	96.6
Average number of active threads		0.996	1.979	3.916	7.662	15.044	22.313	29.796	37.157	44.343	51.620	59.006
Affinity Stability (%)		36.1	99.6	99.5	99.4	99.4	99.5	99.4	99.4	99.4	99.4	99.3
Time in analyzed loops (%)		98.7	98.0	96.3	92.2	88.1	85.2	85.8	84.7	83.6	81.9	82.9
Time in analyzed innermost loops (%)		96.8	96.4	94.7	90.8	87.2	84.4	85.1	84.0	82.9	81.2	82.2
Time in user code (%)		99.4	98.6	97.0	92.9	88.5	85.6	86.1	85.1	84.0	82.3	83.4
Compilation Options Score (%)		100	100.0	99.9	99.8	99.7	99.7	99.7	99.6	99.6	99.6	99.6
Array Access Efficiency (%)		87.3	86.6	87.5	86.1	85.8	83.5	85.0	85.5	84.5	83.4	81.5
Potential Speedups
Perfect Flow Complexity		1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00
Perfect OpenMP/MPI/Pthread/TBB		1.00	1.00	1.01	1.03	1.08	1.11	1.08	1.12	1.09	1.15	1.11
Perfect OpenMP/MPI/Pthread/TBB + Perfect Load Distribution		1.00	1.01	1.03	1.09	1.14	1.19	1.19	1.20	1.22	1.24	1.23
Scalability - Gap		1.00	1.09	1.13	1.28	1.92	2.66	3.30	3.90	4.52	5.36	6.11
No Scalar Integer	Potential Speedup	1.20	1.19	1.19	1.18	1.17	1.17	1.17	1.16	1.16	1.16	1.16
	Nb Loops to get 80%	1	1	1	1	1	1	1	1	1	1	1
FP Vectorised	Potential Speedup	1.20	1.20	1.19	1.18	1.17	1.17	1.17	1.16	1.16	1.16	1.16
	Nb Loops to get 80%	1	1	1	1	1	1	1	1	1	1	1
Fully Vectorised	Potential Speedup	3.36	3.29	3.17	2.90	2.68	2.54	2.56	2.51	2.46	2.39	2.43
	Nb Loops to get 80%	1	1	1	1	1	1	1	1	1	1	1
Only FP Arithmetic	Potential Speedup	1.99	1.97	1.94	1.86	1.79	1.75	1.75	1.74	1.72	1.70	1.71
	Nb Loops to get 80%	1	1	1	1	1	1	1	1	1	1	1

Scalability Speedup

Cumulated Speedup If No Scalar Integer

Cumulated Speedup If FP Vectorized

Cumulated Speedup If Fully Vectorized

Cumulated Speedup If Only FP Arithmetic

Loop Based Profiles

Innermost / Single Loops

Inbetween Loops

Outermost Loops

Cumulated Coverage With All Loops

Innermost Loop Based Profiles

Coverage

Count

Application Categorization

Time

Coverage

Compilation Options

r0r1r2r3r4r5r6r7r8r9r10

Source Object	Issue
▼libllama.so–
▼hashtable.h–
○
▼llama-model.cpp–
○
▼llama-context.cpp–
○
▼llama-batch.cpp–
○
▼llama-kv-cache.cpp–
○
▼llama-vocab.cpp–
○
▼llama-sampling.cpp–
○
▼stl_uninitialized.h–
○
▼stl_vector.h–
○
▼llama-kv-cells.h–
○
▼unicode.cpp–
○
▼hashtable_policy.h–
○
▼llama-impl.cpp–
○
▼libggml-cpu.so–
▼binary-ops.cpp–
○
▼traits.cpp–
○
▼repack.cpp–
○
▼ggml-cpu.cpp–
○
▼ops.cpp–
○
▼vec.cpp–
○
▼ggml-cpu.c–
○
▼quants.c–
○
▼libggml-base.so–
▼ggml.c–
○
▼stl_construct.h–
○
▼ggml-alloc.c–
○
▼gguf.cpp–
○
▼vector.tcc–
○
▼ggml-backend.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.
▼exec–
▼regex_executor.tcc–
○
▼common.cpp–
○
▼sampling.cpp–
○
▼vector.tcc–
○
▼invoke.h–
○
▼main.cpp–
○

Source Object	Issue
▼libllama.so–
▼hashtable.h–
○
▼llama-model.cpp–
○
▼llama-context.cpp–
○
▼llama-batch.cpp–
○
▼llama-kv-cache.cpp–
○
▼llama-vocab.cpp–
○
▼llama-sampling.cpp–
○
▼stl_uninitialized.h–
○
▼stl_vector.h–
○
▼llama-kv-cells.h–
○
▼unicode.cpp–
○
▼hashtable_policy.h–
○
▼llama-impl.cpp–
○
▼libggml-cpu.so–
▼binary-ops.cpp–
○
▼traits.cpp–
○
▼repack.cpp–
○
▼ggml-cpu.cpp–
○
▼ops.cpp–
○
▼vec.cpp–
○
▼ggml-cpu.c–
○
▼quants.c–
○
▼libggml-base.so–
▼ggml.c–
○
▼stl_construct.h–
○
▼ggml-alloc.c–
○
▼gguf.cpp–
○
▼vector.tcc–
○
▼ggml-backend.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.
▼exec–
▼regex_executor.tcc–
○
▼common.cpp–
○
▼sampling.cpp–
○
▼vector.tcc–
○
▼invoke.h–
○
▼main.cpp–
○

Source Object	Issue
▼libllama.so–
▼hashtable.h–
○
▼llama-model.cpp–
○
▼llama-context.cpp–
○
▼llama-batch.cpp–
○
▼llama-kv-cache.cpp–
○
▼llama-vocab.cpp–
○
▼llama-sampling.cpp–
○
▼stl_uninitialized.h–
○
▼stl_vector.h–
○
▼llama-kv-cells.h–
○
▼unicode.cpp–
○
▼hashtable_policy.h–
○
▼llama-impl.cpp–
○
▼libggml-cpu.so–
▼binary-ops.cpp–
○
▼traits.cpp–
○
▼repack.cpp–
○
▼ggml-cpu.cpp–
○
▼ops.cpp–
○
▼vec.cpp–
○
▼ggml-cpu.c–
○
▼quants.c–
○
▼libggml-base.so–
▼ggml.c–
○
▼stl_construct.h–
○
▼ggml-alloc.c–
○
▼gguf.cpp–
○
▼vector.tcc–
○
▼ggml-backend.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.
▼exec–
▼regex_executor.tcc–
○
▼common.cpp–
○
▼sampling.cpp–
○
▼vector.tcc–
○
▼invoke.h–
○
▼main.cpp–
○

Source Object	Issue
▼libllama.so–
▼hashtable.h–
○
▼llama-model.cpp–
○
▼llama-context.cpp–
○
▼llama-batch.cpp–
○
▼llama-kv-cache.cpp–
○
▼llama-vocab.cpp–
○
▼llama-sampling.cpp–
○
▼stl_uninitialized.h–
○
▼stl_vector.h–
○
▼llama-kv-cells.h–
○
▼unicode.cpp–
○
▼hashtable_policy.h–
○
▼llama-impl.cpp–
○
▼libggml-cpu.so–
▼binary-ops.cpp–
○
▼traits.cpp–
○
▼repack.cpp–
○
▼ggml-cpu.cpp–
○
▼ops.cpp–
○
▼vec.cpp–
○
▼ggml-cpu.c–
○
▼quants.c–
○
▼libggml-base.so–
▼ggml.c–
○
▼stl_construct.h–
○
▼ggml-alloc.c–
○
▼gguf.cpp–
○
▼vector.tcc–
○
▼ggml-backend.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.
▼exec–
▼regex_executor.tcc–
○
▼common.cpp–
○
▼sampling.cpp–
○
▼vector.tcc–
○
▼invoke.h–
○
▼main.cpp–
○

Source Object	Issue
▼libllama.so–
▼hashtable.h–
○
▼llama-model.cpp–
○
▼llama-context.cpp–
○
▼llama-batch.cpp–
○
▼llama-kv-cache.cpp–
○
▼llama-vocab.cpp–
○
▼llama-sampling.cpp–
○
▼stl_uninitialized.h–
○
▼stl_vector.h–
○
▼llama-kv-cells.h–
○
▼unicode.cpp–
○
▼hashtable_policy.h–
○
▼llama-impl.cpp–
○
▼libggml-cpu.so–
▼binary-ops.cpp–
○
▼traits.cpp–
○
▼repack.cpp–
○
▼ggml-cpu.cpp–
○
▼ops.cpp–
○
▼vec.cpp–
○
▼ggml-cpu.c–
○
▼quants.c–
○
▼libggml-base.so–
▼ggml.c–
○
▼stl_construct.h–
○
▼ggml-alloc.c–
○
▼gguf.cpp–
○
▼vector.tcc–
○
▼ggml-backend.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.
▼exec–
▼regex_executor.tcc–
○
▼common.cpp–
○
▼sampling.cpp–
○
▼vector.tcc–
○
▼invoke.h–
○
▼main.cpp–
○

Source Object	Issue
▼libllama.so–
▼hashtable.h–
○
▼llama-model.cpp–
○
▼llama-context.cpp–
○
▼llama-batch.cpp–
○
▼llama-kv-cache.cpp–
○
▼llama-vocab.cpp–
○
▼llama-sampling.cpp–
○
▼stl_uninitialized.h–
○
▼stl_vector.h–
○
▼llama-kv-cells.h–
○
▼unicode.cpp–
○
▼hashtable_policy.h–
○
▼llama-impl.cpp–
○
▼libggml-cpu.so–
▼binary-ops.cpp–
○
▼traits.cpp–
○
▼repack.cpp–
○
▼ggml-cpu.cpp–
○
▼ops.cpp–
○
▼vec.cpp–
○
▼ggml-cpu.c–
○
▼quants.c–
○
▼libggml-base.so–
▼ggml.c–
○
▼stl_construct.h–
○
▼ggml-alloc.c–
○
▼gguf.cpp–
○
▼vector.tcc–
○
▼ggml-backend.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.
▼exec–
▼regex_executor.tcc–
○
▼common.cpp–
○
▼sampling.cpp–
○
▼vector.tcc–
○
▼invoke.h–
○
▼main.cpp–
○

Source Object	Issue
▼libllama.so–
▼hashtable.h–
○
▼llama-model.cpp–
○
▼llama-context.cpp–
○
▼llama-batch.cpp–
○
▼llama-kv-cache.cpp–
○
▼llama-vocab.cpp–
○
▼llama-sampling.cpp–
○
▼stl_uninitialized.h–
○
▼stl_vector.h–
○
▼llama-kv-cells.h–
○
▼unicode.cpp–
○
▼hashtable_policy.h–
○
▼llama-impl.cpp–
○
▼libggml-cpu.so–
▼binary-ops.cpp–
○
▼traits.cpp–
○
▼repack.cpp–
○
▼ggml-cpu.cpp–
○
▼ops.cpp–
○
▼vec.cpp–
○
▼ggml-cpu.c–
○
▼quants.c–
○
▼libggml-base.so–
▼ggml.c–
○
▼stl_construct.h–
○
▼ggml-alloc.c–
○
▼gguf.cpp–
○
▼vector.tcc–
○
▼ggml-backend.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.
▼exec–
▼regex_executor.tcc–
○
▼common.cpp–
○
▼sampling.cpp–
○
▼vector.tcc–
○
▼invoke.h–
○
▼main.cpp–
○

Source Object	Issue
▼libllama.so–
▼hashtable.h–
○
▼llama-model.cpp–
○
▼llama-context.cpp–
○
▼llama-batch.cpp–
○
▼llama-kv-cache.cpp–
○
▼llama-vocab.cpp–
○
▼llama-sampling.cpp–
○
▼stl_uninitialized.h–
○
▼stl_vector.h–
○
▼llama-kv-cells.h–
○
▼unicode.cpp–
○
▼hashtable_policy.h–
○
▼llama-impl.cpp–
○
▼libggml-cpu.so–
▼binary-ops.cpp–
○
▼traits.cpp–
○
▼repack.cpp–
○
▼ggml-cpu.cpp–
○
▼ops.cpp–
○
▼vec.cpp–
○
▼ggml-cpu.c–
○
▼quants.c–
○
▼libggml-base.so–
▼ggml.c–
○
▼stl_construct.h–
○
▼ggml-alloc.c–
○
▼gguf.cpp–
○
▼vector.tcc–
○
▼ggml-backend.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.
▼exec–
▼regex_executor.tcc–
○
▼common.cpp–
○
▼sampling.cpp–
○
▼vector.tcc–
○
▼invoke.h–
○
▼main.cpp–
○

Source Object	Issue
▼libllama.so–
▼hashtable.h–
○
▼llama-model.cpp–
○
▼llama-context.cpp–
○
▼llama-batch.cpp–
○
▼llama-kv-cache.cpp–
○
▼llama-vocab.cpp–
○
▼llama-sampling.cpp–
○
▼stl_uninitialized.h–
○
▼stl_vector.h–
○
▼llama-kv-cells.h–
○
▼unicode.cpp–
○
▼hashtable_policy.h–
○
▼llama-impl.cpp–
○
▼libggml-cpu.so–
▼binary-ops.cpp–
○
▼traits.cpp–
○
▼repack.cpp–
○
▼ggml-cpu.cpp–
○
▼ops.cpp–
○
▼vec.cpp–
○
▼ggml-cpu.c–
○
▼quants.c–
○
▼libggml-base.so–
▼ggml.c–
○
▼stl_construct.h–
○
▼ggml-alloc.c–
○
▼gguf.cpp–
○
▼vector.tcc–
○
▼ggml-backend.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.
▼exec–
▼regex_executor.tcc–
○
▼common.cpp–
○
▼sampling.cpp–
○
▼vector.tcc–
○
▼invoke.h–
○
▼main.cpp–
○

Source Object	Issue
▼libllama.so–
▼hashtable.h–
○
▼llama-model.cpp–
○
▼llama-context.cpp–
○
▼llama-batch.cpp–
○
▼llama-kv-cache.cpp–
○
▼llama-vocab.cpp–
○
▼llama-sampling.cpp–
○
▼stl_uninitialized.h–
○
▼stl_vector.h–
○
▼llama-kv-cells.h–
○
▼unicode.cpp–
○
▼hashtable_policy.h–
○
▼llama-impl.cpp–
○
▼libggml-cpu.so–
▼binary-ops.cpp–
○
▼traits.cpp–
○
▼repack.cpp–
○
▼ggml-cpu.cpp–
○
▼ops.cpp–
○
▼vec.cpp–
○
▼ggml-cpu.c–
○
▼quants.c–
○
▼libggml-base.so–
▼ggml.c–
○
▼stl_construct.h–
○
▼ggml-alloc.c–
○
▼gguf.cpp–
○
▼vector.tcc–
○
▼ggml-backend.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.
▼exec–
▼regex_executor.tcc–
○
▼common.cpp–
○
▼sampling.cpp–
○
▼vector.tcc–
○
▼invoke.h–
○
▼main.cpp–
○

Source Object	Issue
▼libllama.so–
▼hashtable.h–
○
▼llama-model.cpp–
○
▼llama-context.cpp–
○
▼llama-batch.cpp–
○
▼llama-kv-cache.cpp–
○
▼llama-vocab.cpp–
○
▼llama-sampling.cpp–
○
▼stl_uninitialized.h–
○
▼stl_vector.h–
○
▼llama-kv-cells.h–
○
▼unicode.cpp–
○
▼hashtable_policy.h–
○
▼llama-impl.cpp–
○
▼libggml-cpu.so–
▼binary-ops.cpp–
○
▼traits.cpp–
○
▼repack.cpp–
○
▼ggml-cpu.cpp–
○
▼ops.cpp–
○
▼vec.cpp–
○
▼ggml-cpu.c–
○
▼quants.c–
○
▼libggml-base.so–
▼ggml.c–
○
▼stl_construct.h–
○
▼ggml-alloc.c–
○
▼gguf.cpp–
○
▼vector.tcc–
○
▼ggml-backend.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.
▼exec–
▼regex_executor.tcc–
○
▼common.cpp–
○
▼sampling.cpp–
○
▼vector.tcc–
○
▼invoke.h–
○
▼main.cpp–
○

Path Count Profiles

Coverage

Count

Low Iteration Count Profiles

Coverage

Count

Average Number of Active Threads

Run 1 - 1x1

Run 2 - 1x2

Run 3 - 1x4

Run 4 - 1x8

Run 5 - 1x16

Run 6 - 1x24

Run 7 - 1x32

Run 8 - 1x40

Run 9 - 1x48

Run 10 - 1x56

Run 11 - 1x64