OV - Compare Loops

Loops

▶repack.cpp: 125 - 127.26 %

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 (%)	ASM Loop ID	Vector Length Use (%)
Run orig_default						Run icx_default						Run aocc_7						Run icx_8
Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 125-125 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 137-140 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 755-779 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 822-842 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 869-908 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 943-968 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 1011-1031					Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 125-125 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 137-140 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 869-869 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 892-908 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 943-943 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 949-949 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 955-955 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 961-968 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 1011-1031 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 755-760 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 766-766 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 772-779 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 822-842					Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 125-125 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 137-140 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 755-779 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 822-842 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 869-908 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 943-968 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 1011-1031					Loop Source Regions
2594	0.04	0.02	0.61	99.51	83.19	2985	1.64	1.40	40.82	99.19	85.93	2218	0.05	0.02	0.58	99.51	83.19
2591	1.64	1.41	41.58	99.23	80.75	2990	0.04	0.02	0.65	99.48	85.43	2215	2.10	1.63	43.02	99.23	80.75

Sum on 2 analyzed binary loops (libggml-cpu.so - 2594, libggml-cpu.so - 2591)						Sum on 2 analyzed binary loops (libggml-cpu.so - 2985, libggml-cpu.so - 2990)						Sum on 2 analyzed binary loops (libggml-cpu.so - 2218, libggml-cpu.so - 2215)						No Optimizer analysis found for any assembly loop. More loops can be analyzed using option --optimizer-loop-count.
Analysis					Count	Analysis					Count	Analysis					Count	Analysis	Count
Data Access Issues						Data Access Issues						Data Access Issues
More than 10% of the vector loads instructions are unaligned					1	More than 10% of the vector loads instructions are unaligned					0	More than 10% of the vector loads instructions are unaligned					1
Presence of special instructions executing on a single port					1	Presence of special instructions executing on a single port					1	Presence of special instructions executing on a single port					1
More than 20% of the loads are accessing the stack					1	More than 20% of the loads are accessing the stack					1	More than 20% of the loads are accessing the stack					1
Inefficient Vectorization						Inefficient Vectorization						Inefficient Vectorization
Presence of special instructions executing on a single port					1	Presence of special instructions executing on a single port					1	Presence of special instructions executing on a single port					1
Use of masked instructions					1	Use of masked instructions					1	Use of masked instructions					1

▶quants.c: 682 - 63.80 %

ASM Loop ID	Vector Length Use (%)	ASM Loop ID	Vector Length Use (%)	ASM Loop ID	Vector Length Use (%)	ASM Loop ID	Max Time Over Threads (s)	Time w.r.t. Wall Time (s)	Cov (%)	Vect. Ratio (%)	Vector Length Use (%)
Run orig_default		Run icx_default		Run aocc_7		Run icx_8
Loop Source Regions		Loop Source Regions		Loop Source Regions		Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/quants.c: 682-682 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/quants.c: 690-695
						3699	5.42	4.89	63.80	93.02	21.37

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.		Sum on 1 analyzed binary loop (libggml-cpu.so - 3699)
Analysis	Count	Analysis	Count	Analysis	Count	Analysis					Count
						Loop Computation Issues
						Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA					1
						Data Access Issues
						Presence of indirect access					1
						Presence of special instructions executing on a single port					1
						Vectorization Roadblocks
						Presence of indirect access					1
						Inefficient Vectorization
						Presence of special instructions executing on a single port					1

▶quants.c: 2150 - 37.51 %

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 (%)	ASM Loop ID	Vector Length Use (%)
Run orig_default						Run icx_default						Run aocc_7						Run icx_8
Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/quants.c: 2150-2152 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/quants.c: 2158-2158 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/quants.c: 2166-2216					Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/quants.c: 2150-2152 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/quants.c: 2158-2158 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/quants.c: 2166-2216					Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/quants.c: 2150-2152 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/quants.c: 2158-2158 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/quants.c: 2166-2216 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/simd-mappings.h: 130-130					Loop Source Regions
2545	0.48	0.43	12.78	97.09	43.87	2936	0.49	0.46	13.35	96.81	44.08	2166	0.49	0.43	11.38	97.09	43.87

Sum on 1 analyzed binary loop (libggml-cpu.so - 2545)						Sum on 1 analyzed binary loop (libggml-cpu.so - 2936)						Sum on 1 analyzed binary loop (libggml-cpu.so - 2166)						No Optimizer analysis found for any assembly loop. More loops can be analyzed using option --optimizer-loop-count.
Analysis					Count	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
Presence of indirect access					1	Presence of indirect access					1	Presence of indirect access					1
Presence of special instructions executing on a single port					1	Presence of special instructions executing on a single port					1	Presence of special instructions executing on a single port					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
Presence of indirect access					1	Presence of indirect access					1	Presence of indirect access					1
Inefficient Vectorization						Inefficient Vectorization						Inefficient Vectorization
Presence of special instructions executing on a single port					1	Presence of special instructions executing on a single port					1	Presence of special instructions executing on a single port					1

▶repack.cpp: 153 - 12.79 %

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 (%)	ASM Loop ID	Vector Length Use (%)
Run orig_default						Run icx_default						Run aocc_7						Run icx_8
Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 153-153 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 170-172 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 1227-1251 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 1285-1301 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 1344-1363					Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 153-153 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 170-172 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 1227-1251 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 1285-1285 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 1291-1291 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 1297-1297 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 1344-1363					Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 153-153 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 170-172 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 1227-1251 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 1285-1301 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/repack.cpp: 1344-1363					Loop Source Regions
2584	0.28	0.15	4.37	0	0	2975	0.25	0.15	4.28	0	0	2208	0.28	0.16	4.14	0	0

Sum on 1 analyzed binary loop (libggml-cpu.so - 2584)						Sum on 1 analyzed binary loop (libggml-cpu.so - 2975)						Sum on 1 analyzed binary loop (libggml-cpu.so - 2208)						No Optimizer analysis found for any assembly loop. More loops can be analyzed using option --optimizer-loop-count.
Analysis					Count	Analysis					Count	Analysis					Count	Analysis	Count

▶quants.c: 716 - 11.40 %

ASM Loop ID	Vector Length Use (%)	ASM Loop ID	Vector Length Use (%)	ASM Loop ID	Vector Length Use (%)	ASM Loop ID	Max Time Over Threads (s)	Time w.r.t. Wall Time (s)	Cov (%)	Vect. Ratio (%)	Vector Length Use (%)
Run orig_default		Run icx_default		Run aocc_7		Run icx_8
Loop Source Regions		Loop Source Regions		Loop Source Regions		Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/simd-mappings.h: 130-130 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/quants.c: 716-716 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/quants.c: 725-725 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/quants.c: 733-736 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/quants.c: 745-754
						494	0.90	0.87	11.40	46.92	13.2

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.		Sum on 1 analyzed binary loop (libggml-cpu.so - 494)
Analysis	Count	Analysis	Count	Analysis	Count	Analysis					Count
						Loop Computation Issues
						Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA					1
						Presence of a large number of scalar integer instructions					1
						Data Access Issues
						Presence of constant non-unit stride data access					1
						Presence of indirect access					1
						Presence of special instructions executing on a single port					1
						More than 20% of the loads are accessing the stack					1
						Vectorization Roadblocks
						Presence of constant non-unit stride data access					1
						Presence of indirect access					1
						Inefficient Vectorization
						Presence of special instructions executing on a single port					1

▶<unknown>: 0 - 2.54 %

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 (%)	ASM Loop ID	Max Time Over Threads (s)	Time w.r.t. Wall Time (s)	Cov (%)	Vect. Ratio (%)	Vector Length Use (%)
Run orig_default						Run icx_default						Run aocc_7						Run icx_8
Loop Source Regions						Loop Source Regions						Loop Source Regions						Loop Source Regions
2485	0.02	0.00	0.01	0	0	2395	0.02	0.00	0.01	0	0	2426	0.00	0.00	0.00	0	0	4113	0.00	0.00	0.00	0	0
1282	0.00	0.00	0.00	0	0	2155	0.00	0.00	0.00	0	0	2417	0.01	0.00	0.00	0	0	441	0.00	0.00	0.00	0	0
1126	0.00	0.00	0.00	0	0	2189	0.00	0.00	0.00	0	0	2197	0.00	0.00	0.00	0	0	4116	0.01	0.00	0.00	0	0
2741	0.02	0.00	0.01	0	0	2298	0.02	0.00	0.01	0	0	2330	0.00	0.00	0.00	0	0	4128	0.00	0.00	0.00	0	0
2738	0.00	0.00	0.00	0	0	444	0.06	0.00	0.08	0	0	2517	0.01	0.00	0.00	0	0	4122	0.00	0.00	0.00	0	0
0	0.01	0.00	0.00	0	0	106	0.01	0.00	0.00	0	0	2414	0.00	0.00	0.00	0	0	867	0.01	0.00	0.00	0	0
99	0.01	0.00	0.07	0	0	124	0.01	0.00	0.09	0	0	315	0.01	0.00	0.01	0	0	3515	0.00	0.00	0.00	0	0
485	0.02	0.00	0.02	0	0	1412	0.00	0.00	0.00	0	0	766	0.09	0.00	0.08	0	0	3514	0.01	0.00	0.00	0	0
2563	0.08	0.00	0.03	0	0	1120	0.00	0.00	0.00	0	0	201	0.22	0.00	0.07	0	0	2855	0.04	0.01	0.08	0	0
399	0.08	0.00	0.09	0	0	873	0.05	0.00	0.07	0	0	1271	0.03	0.00	0.05	0	0	105	0.02	0.01	0.07	0	0
83	0.01	0.00	0.00	0	0	2099	0.01	0.00	0.07	0	0	353	0.08	0.00	0.08	0	0	114	0.01	0.00	0.01	0	0
85	0.00	0.00	0.00	0	0	6	0.01	0.00	0.04	0	0	63	0.01	0.00	0.02	0	0	631	0.06	0.00	0.02	0	0
1731	0.00	0.00	0.00	0	0	2980	0.02	0.00	0.09	0	0	433	0.02	0.00	0.02	0	0	2844	0.02	0.00	0.03	0	0
1197	0.00	0.00	0.00	0	0	2094	0.01	0.00	0.00	0	0	1094	0.01	0.00	0.00	0	0	1635	0.13	0.00	0.06	0	0
1710	0.02	0.00	0.09	0	0	1418	0.02	0.00	0.02	0	0	5	0.01	0.00	0.03	0	0	116	0.01	0.00	0.04	0	0
1457	0.02	0.00	0.04	0	0	1707	0.02	0.00	0.05	0	0	1095	0.00	0.00	0.00	0	0	3292	0.01	0.00	0.01	0	0
359	0.01	0.00	0.01	0	0	401	0.01	0.00	0.01	0	0	1466	0.00	0.00	0.00	0	0	109	0.01	0.00	0.03	0	0
1705	0.01	0.00	0.01	0	0	102	0.01	0.00	0.02	0	0	1440	0.00	0.00	0.00	0	0	2442	0.00	0.00	0.00	0	0
1196	0.04	0.00	0.03	0	0	558	0.03	0.00	0.03	0	0	2186	0.15	0.00	0.05	0	0	2689	0.00	0.00	0.00	0	0
5	0.01	0.00	0.02	0	0	2111	0.01	0.00	0.01	0	0	78	0.02	0.00	0.07	0	0	2444	0.02	0.00	0.01	0	0
82	0.01	0.00	0.03	0	0	257	0.27	0.00	0.09	0	0	2211	0.01	0.00	0.08	0	0	3296	0.01	0.00	0.00	0	0
766	0.07	0.00	0.08	0	0	2954	0.11	0.00	0.04	0	0	444	0.01	0.00	0.00	0	0	111	0.01	0.00	0.00	0	0
233	0.17	0.00	0.06	0	0	1947	0.01	0.00	0.00	0	0							525	0.00	0.00	0.00	0	0
2587	0.02	0.00	0.09	0	0	815	0.01	0.00	0.00	0	0							132	0.03	0.01	0.09	0	0
313	0.00	0.00	0.00	0	0													3	0.03	0.00	0.04	0	0
																		3309	0.01	0.00	0.02	0	0
																		893	0.03	0.00	0.01	0	0

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.						No Optimizer analysis found for any assembly loop. More loops can be analyzed using option --optimizer-loop-count.
Analysis					Count	Analysis					Count	Analysis					Count	Analysis					Count

▶vec.h: 491 - 1.49 %

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 (%)	ASM Loop ID	Vector Length Use (%)
Run orig_default						Run icx_default						Run aocc_7						Run icx_8
Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.h: 491-497					Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.h: 491-497					Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.h: 491-497					Loop Source Regions
1724	0.05	0.02	0.49	100	75	2109	0.05	0.02	0.52	100	75	1459	0.04	0.02	0.48	100	75

Sum on 1 analyzed binary loop (libggml-cpu.so - 1724)						Sum on 1 analyzed binary loop (libggml-cpu.so - 2109)						Sum on 1 analyzed binary loop (libggml-cpu.so - 1459)						No Optimizer analysis found for any assembly loop. More loops can be analyzed using option --optimizer-loop-count.
Analysis					Count	Analysis					Count	Analysis					Count	Analysis	Count

▶vec.cpp: 311 - 0.98 %

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 (%)	ASM Loop ID	Vector Length Use (%)
Run orig_default						Run icx_default						Run aocc_7						Run icx_8
Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.cpp: 311-316					Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.cpp: 311-316					Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.cpp: 311-316					Loop Source Regions
762	0.04	0.01	0.34	100	66.67	866	0.04	0.01	0.36	100	66.67	758	0.04	0.01	0.28	100	66.67

Sum on 1 analyzed binary loop (libggml-cpu.so - 762)						Sum on 1 analyzed binary loop (libggml-cpu.so - 866)						Sum on 1 analyzed binary loop (libggml-cpu.so - 758)						No Optimizer analysis found for any assembly loop. More loops can be analyzed using option --optimizer-loop-count.
Analysis					Count	Analysis					Count	Analysis					Count	Analysis	Count
Loop Computation Issues						Loop Computation Issues						Loop Computation Issues
Low iteration count						Low iteration count					1	Low iteration count
Control Flow Issues						Control Flow Issues						Control Flow Issues
Low iteration count						Low iteration count					1	Low iteration count

▶vec.cpp: 331 - 0.94 %

ASM Loop ID	Vector Length Use (%)	ASM Loop ID	Vector Length Use (%)	ASM Loop ID	Vector Length Use (%)	ASM Loop ID	Max Time Over Threads (s)	Time w.r.t. Wall Time (s)	Cov (%)	Vect. Ratio (%)	Vector Length Use (%)
Run orig_default		Run icx_default		Run aocc_7		Run icx_8
Loop Source Regions		Loop Source Regions		Loop Source Regions		Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/simd-mappings.h: 130-130 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.cpp: 331-332
						1621	0.20	0.07	0.94	5.88	8.82

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.		Sum on 1 analyzed binary loop (libggml-cpu.so - 1621)
Analysis	Count	Analysis	Count	Analysis	Count	Analysis					Count
						Loop Computation Issues
						Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA					1
						Data Access Issues
						Presence of indirect access					1
						Vectorization Roadblocks
						Presence of indirect access					1

▶vec.h: 508 - 0.91 %

ASM Loop ID	Vector Length Use (%)	ASM Loop ID	Vector Length Use (%)	ASM Loop ID	Vector Length Use (%)	ASM Loop ID	Max Time Over Threads (s)	Time w.r.t. Wall Time (s)	Cov (%)	Vect. Ratio (%)	Vector Length Use (%)
Run orig_default		Run icx_default		Run aocc_7		Run icx_8
Loop Source Regions		Loop Source Regions		Loop Source Regions		Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.h: 508-509 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/./ggml-impl.h: 346-346 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/./ggml-impl.h: 389-389 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/./ggml-impl.h: 399-404 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/simd-mappings.h: 130-130
						3306	0.18	0.07	0.91	85.05	20.79

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.		Sum on 1 analyzed binary loop (libggml-cpu.so - 3306)
Analysis	Count	Analysis	Count	Analysis	Count	Analysis					Count
						Loop Computation Issues
						Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA					1
						Data Access Issues
						Presence of indirect access					1
						Presence of special instructions executing on a single port					1
						Vectorization Roadblocks
						Presence of indirect access					1
						Inefficient Vectorization
						Presence of special instructions executing on a single port					1

▶ops.cpp: 6220 - 0.53 %

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	Vector Length Use (%)	ASM Loop ID	Vector Length Use (%)
Run orig_default						Run icx_default						Run aocc_7		Run icx_8
Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6220-6220 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6229-6230 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6238-6245					Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6220-6220 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6229-6230 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6238-6245					Loop Source Regions		Loop Source Regions
1463	0.04	0.01	0.28	6.67	7.5	1734	0.03	0.01	0.25	15.56	9.17

Sum on 1 analyzed binary loop (libggml-cpu.so - 1463)						Sum on 1 analyzed binary loop (libggml-cpu.so - 1734)						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	Analysis	Count
Loop Computation Issues						Loop Computation Issues
Presence of expensive FP instructions					1	Presence of expensive FP instructions					1
Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA					1	Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA					1
Presence of a large number of scalar integer instructions					1	Presence of a large number of scalar integer instructions					0
Control Flow Issues						Control Flow Issues
Presence of calls					1	Presence of calls					1
Data Access Issues						Data Access Issues
More than 20% of the loads are accessing the stack					1	More than 20% of the loads are accessing the stack					1
Vectorization Roadblocks						Vectorization Roadblocks
Presence of calls					1	Presence of calls					1

▶ops.cpp: 6210 - 0.34 %

ASM Loop ID	Vector Length Use (%)	ASM Loop ID	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	Vector Length Use (%)
Run orig_default		Run icx_default		Run aocc_7						Run icx_8
Loop Source Regions		Loop Source Regions		Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6210-6211 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6220-6230 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6238-6245					Loop Source Regions
				1274	0.04	0.01	0.34	5.56	7.29

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.		Sum on 1 analyzed binary loop (libggml-cpu.so - 1274)						No Optimizer analysis found for any assembly loop. More loops can be analyzed using option --optimizer-loop-count.
Analysis	Count	Analysis	Count	Analysis					Count	Analysis	Count
				Loop Computation Issues
				Presence of expensive FP instructions					1
				Presence of a large number of scalar integer instructions					1
				Control Flow Issues
				Presence of calls					1
				Presence of more than 4 paths					1
				Data Access Issues
				Presence of special instructions executing on a single port					1
				More than 20% of the loads are accessing the stack					1
				Vectorization Roadblocks
				Presence of calls					1
				Presence of more than 4 paths					1
				Inefficient Vectorization
				Presence of special instructions executing on a single port					1
				Use of masked instructions					1

▶ggml-quants.c: 216 - 0.12 %

ASM Loop ID	Vector Length Use (%)	ASM Loop ID	Vector Length Use (%)	ASM Loop ID	Vector Length Use (%)	ASM Loop ID	Max Time Over Threads (s)	Time w.r.t. Wall Time (s)	Cov (%)	Vect. Ratio (%)	Vector Length Use (%)
Run orig_default		Run icx_default		Run aocc_7		Run icx_8
Loop Source Regions		Loop Source Regions		Loop Source Regions		Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-quants.c: 216-219
						638	0.04	0.01	0.12	95	22.5

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.		Sum on 1 analyzed binary loop (libggml-base.so - 638)
Analysis	Count	Analysis	Count	Analysis	Count	Analysis					Count
						Loop Computation Issues
						Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA					1
						Presence of a large number of scalar integer instructions					1
						Control Flow Issues
						Presence of calls					1
						Data Access Issues
						More than 10% of the vector loads instructions are unaligned					1
						Presence of special instructions executing on a single port					1
						More than 20% of the loads are accessing the stack					1
						Vectorization Roadblocks
						Presence of calls					1
						Inefficient Vectorization
						Presence of special instructions executing on a single port					1

▶ops.cpp: 8885 - 0.11 %

ASM Loop ID	Vector Length Use (%)	ASM Loop ID	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	Vector Length Use (%)
Run orig_default		Run icx_default		Run aocc_7						Run icx_8
Loop Source Regions		Loop Source Regions		Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 8885-8886 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz17.benchmarkcenter.megware.com/176-989-3417/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/simd-mappings.h: 130-130					Loop Source Regions
				1445	0.03	0.00	0.11	0	6.25

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.		Sum on 1 analyzed binary loop (libggml-cpu.so - 1445)						No Optimizer analysis found for any assembly loop. More loops can be analyzed using option --optimizer-loop-count.
Analysis	Count	Analysis	Count	Analysis					Count	Analysis	Count
				Data Access Issues
				Presence of indirect access					1
				More than 20% of the loads are accessing the stack					1
				Vectorization Roadblocks
				Presence of indirect access					1

Report Configuration

Loops

▶repack.cpp: 125 - 127.26 %

▶quants.c: 682 - 63.80 %

▶quants.c: 2150 - 37.51 %

▶repack.cpp: 153 - 12.79 %

▶quants.c: 716 - 11.40 %

▶<unknown>: 0 - 2.54 %

▶vec.h: 491 - 1.49 %

▶vec.cpp: 311 - 0.98 %

▶vec.cpp: 331 - 0.94 %

▶vec.h: 508 - 0.91 %

▶ops.cpp: 6220 - 0.53 %

▶ops.cpp: 6210 - 0.34 %

▶ggml-quants.c: 216 - 0.12 %

▶ops.cpp: 8885 - 0.11 %