OV - Compare Loops

Loops

▶sgemm.cpp: 138 - 196.18 %

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_3						Run icx_5
Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/llamafile/sgemm.cpp: 138-138 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/llamafile/sgemm.cpp: 818-846 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/llamafile/sgemm.cpp: 1038-1038 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/llamafile/sgemm.cpp: 1044-1044					Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/llamafile/sgemm.cpp: 138-138 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/llamafile/sgemm.cpp: 818-821 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/llamafile/sgemm.cpp: 827-846 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/llamafile/sgemm.cpp: 966-966 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/llamafile/sgemm.cpp: 1038-1038 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/llamafile/sgemm.cpp: 1044-1044					Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/llamafile/sgemm.cpp: 138-138 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/llamafile/sgemm.cpp: 818-846 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/llamafile/sgemm.cpp: 1038-1038 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/llamafile/sgemm.cpp: 1044-1044					Loop Source Regions
2364	8.41	6.13	65.31	0	0	2688	8.35	6.15	65.27	0	0	2437	8.36	6.13	65.60	0	0

Sum on 1 analyzed binary loop (libggml-cpu.so - 2364)						Sum on 1 analyzed binary loop (libggml-cpu.so - 2688)						Sum on 1 analyzed binary loop (libggml-cpu.so - 2437)						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: 1066 - 76.49 %

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_3		Run icx_5
Loop Source Regions		Loop Source Regions		Loop Source Regions		Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/quants.c: 1066-1073 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/simd-mappings.h: 130-130
						3755	9.12	8.05	76.49	88.89	20.14

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 - 3755)
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

▶vec.h: 508 - 1.49 %

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_3		Run icx_5
Loop Source Regions		Loop Source Regions		Loop Source Regions		Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.h: 508-509 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/./ggml-impl.h: 346-346 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/./ggml-impl.h: 355-355 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/./ggml-impl.h: 389-404
						3409	0.31	0.16	1.49	4.17	7.03

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 - 3409)
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 indirect access					1
						Vectorization Roadblocks
						Presence of indirect access					1

▶<unknown>: 0 - 1.34 %

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_3						Run icx_5
Loop Source Regions						Loop Source Regions						Loop Source Regions						Loop Source Regions
2749	0.01	0.00	0.00	0	0	2395	0.02	0.00	0.00	0	0	2659	0.01	0.00	0.00	0	0	4074	0.00	0.00	0.00	0	0
2640	0.01	0.00	0.00	0	0	2155	0.01	0.00	0.00	0	0	2756	0.00	0.00	0.00	0	0	4879	0.01	0.00	0.00	0	0
2738	0.01	0.00	0.00	0	0	2153	0.00	0.00	0.00	0	0	2660	0.00	0.00	0.00	0	0	4393	0.00	0.00	0.00	0	0
2450	0.01	0.00	0.00	0	0	2298	0.00	0.00	0.00	0	0	2650	0.00	0.00	0.00	0	0	4210	0.00	0.00	0.00	0	0
2638	0.01	0.00	0.00	0	0	1947	0.00	0.00	0.00	0	0	2759	0.00	0.00	0.00	0	0	4391	0.01	0.00	0.00	0	0
2741	0.01	0.00	0.00	0	0	1948	0.01	0.00	0.00	0	0	2511	0.00	0.00	0.00	0	0	4413	0.01	0.00	0.00	0	0
765	0.14	0.00	0.05	0	0	813	0.00	0.00	0.00	0	0	2861	0.02	0.00	0.00	0	0	3782	0.00	0.00	0.00	0	0
484	0.01	0.00	0.00	0	0	815	0.00	0.00	0.00	0	0	99	0.02	0.01	0.06	0	0	3781	0.02	0.00	0.00	0	0
90	0.01	0.00	0.02	0	0	709	0.00	0.00	0.00	0	0	357	0.01	0.00	0.01	0	0	1	0.02	0.00	0.02	0	0
398	0.07	0.00	0.03	0	0	444	0.08	0.00	0.03	0	0	1530	0.01	0.00	0.02	0	0	633	0.01	0.00	0.00	0	0
85	0.01	0.00	0.01	0	0	114	0.02	0.00	0.03	0	0	90	0.01	0.00	0.03	0	0	2819	0.02	0.01	0.07	0	0
914	0.00	0.00	0.00	0	0	124	0.02	0.01	0.07	0	0	1803	0.01	0.00	0.01	0	0	64	0.03	0.01	0.09	0	0
1709	0.02	0.00	0.04	0	0	873	0.14	0.01	0.06	0	0	397	0.06	0.00	0.02	0	0	690	0.07	0.00	0.02	0	0
1456	0.01	0.00	0.02	0	0	2099	0.02	0.00	0.03	0	0	85	0.01	0.00	0.01	0	0	75	0.01	0.00	0.02	0	0
87	0.01	0.00	0.00	0	0	2111	0.01	0.00	0.01	0	0	5	0.02	0.00	0.01	0	0	2795	0.02	0.00	0.02	0	0
1704	0.00	0.00	0.00	0	0	1418	0.03	0.00	0.01	0	0	87	0.01	0.00	0.00	0	0	93	0.02	0.01	0.05	0	0
1196	0.00	0.00	0.00	0	0	1707	0.01	0.00	0.02	0	0	1261	0.00	0.00	0.00	0	0	2465	0.02	0.00	0.01	0	0
1195	0.04	0.00	0.01	0	0	1412	0.01	0.00	0.00	0	0	1262	0.00	0.00	0.00	0	0	68	0.01	0.00	0.00	0	0
358	0.01	0.00	0.00	0	0	1120	0.01	0.00	0.00	0	0	1773	0.01	0.00	0.01	0	0	3630	0.00	0.00	0.00	0	0
99	0.02	0.01	0.05	0	0	558	0.02	0.00	0.01	0	0	1782	0.02	0.00	0.03	0	0	2463	0.00	0.00	0.00	0	0
1730	0.01	0.00	0.01	0	0	108	0.01	0.00	0.00	0	0	483	0.04	0.00	0.01	0	0	3400	0.01	0.00	0.00	0	0
5	0.02	0.00	0.01	0	0	2094	0.01	0.00	0.00	0	0	819	0.14	0.01	0.06	0	0	3404	0.00	0.00	0.00	0	0
838	0.01	0.00	0.00	0	0	401	0.00	0.00	0.01	0	0	738	0.00	0.00	0.00	0	0	76	0.02	0.00	0.04	0	0
810	0.01	0.00	0.00	0	0	1734	0.03	0.01	0.09	0	0							1622	0.11	0.00	0.03	0	0
						6	0.01	0.00	0.01	0	0							922	0.03	0.00	0.01	0	0
						113	0.01	0.00	0.01	0	0							3410	0.01	0.00	0.01	0	0
																		70	0.01	0.00	0.00	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

▶quants.c: 298 - 1.06 %

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_3						Run icx_5
Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/quants.c: 298-347 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/quants.c: 353-355 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/./ggml-impl.h: 346-346 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/./ggml-impl.h: 355-355 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/./ggml-impl.h: 389-404					Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/quants.c: 298-347 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/quants.c: 353-355 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/./ggml-impl.h: 346-346 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/./ggml-impl.h: 355-355 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/./ggml-impl.h: 389-404					Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/quants.c: 298-347 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/x86/quants.c: 353-355 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/./ggml-impl.h: 346-346 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/./ggml-impl.h: 355-355 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/./ggml-impl.h: 389-404					Loop Source Regions
2527	0.08	0.03	0.36	58.66	28.88	2918	0.07	0.03	0.35	59.66	29.26	2600	0.07	0.03	0.35	60.7	29.66

Sum on 1 analyzed binary loop (libggml-cpu.so - 2527)						Sum on 1 analyzed binary loop (libggml-cpu.so - 2918)						Sum on 1 analyzed binary loop (libggml-cpu.so - 2600)						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
Presence of expensive FP instructions					1	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					0	Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA					0
Control Flow Issues						Control Flow Issues						Control Flow Issues
Presence of 2 to 4 paths					1	Presence of 2 to 4 paths					1	Presence of 2 to 4 paths					1
Data Access Issues						Data Access Issues						Data Access Issues
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 2 to 4 paths					1	Presence of 2 to 4 paths					1	Presence of 2 to 4 paths					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

▶vec.h: 491 - 0.61 %

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_3						Run icx_5
Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.h: 491-497					Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.h: 491-497					Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.h: 491-497					Loop Source Regions
1723	0.06	0.02	0.20	100	75	2109	0.06	0.02	0.21	100	75	1796	0.07	0.02	0.20	100	75

Sum on 1 analyzed binary loop (libggml-cpu.so - 1723)						Sum on 1 analyzed binary loop (libggml-cpu.so - 2109)						Sum on 1 analyzed binary loop (libggml-cpu.so - 1796)						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: 331 - 0.57 %

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_3		Run icx_5
Loop Source Regions		Loop Source Regions		Loop Source Regions		Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.cpp: 331-332 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/simd-mappings.h: 130-130
						1610	0.19	0.06	0.57	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 - 1610)
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.cpp: 311 - 0.41 %

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_3						Run icx_5
Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.cpp: 311-316					Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.cpp: 311-316					Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.cpp: 311-316					Loop Source Regions
761	0.04	0.01	0.12	100	66.67	866	0.04	0.01	0.12	100	66.67	811	0.05	0.02	0.16	100	66.67

Sum on 1 analyzed binary loop (libggml-cpu.so - 761)						Sum on 1 analyzed binary loop (libggml-cpu.so - 866)						Sum on 1 analyzed binary loop (libggml-cpu.so - 811)						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

▶ggml-quants.c: 203 - 0.36 %

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_3		Run icx_5
Loop Source Regions		Loop Source Regions		Loop Source Regions		Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-impl.h: 346-346 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-impl.h: 355-355 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-impl.h: 389-404 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-quants.c: 203-219
						694	0.07	0.04	0.36	62.15	16.95

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 - 694)
Analysis	Count	Analysis	Count	Analysis	Count	Analysis					Count
						Loop Computation Issues
						Presence of expensive FP instructions					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
						Control Flow Issues
						Presence of calls					1
						Presence of 2 to 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 2 to 4 paths					1
						Inefficient Vectorization
						Presence of special instructions executing on a single port					1

▶ops.cpp: 6220 - 0.29 %

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	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_3						Run icx_5
Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6220-6220 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6229-6230 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6238-6245					Loop Source Regions		Loop Source Regions	/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6220-6220 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6229-6229 /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/176-976-8784/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6238-6245					Loop Source Regions
1462	0.05	0.01	0.15	6.67	7.5			1536	0.04	0.01	0.14	0	6.25

Sum on 1 analyzed binary loop (libggml-cpu.so - 1462)						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 - 1536)						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					1
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

Report Configuration

Loops

▶sgemm.cpp: 138 - 196.18 %

▶quants.c: 1066 - 76.49 %

▶vec.h: 508 - 1.49 %

▶<unknown>: 0 - 1.34 %

▶quants.c: 298 - 1.06 %

▶vec.h: 491 - 0.61 %

▶vec.cpp: 331 - 0.57 %

▶vec.cpp: 311 - 0.41 %

▶ggml-quants.c: 203 - 0.36 %

▶ops.cpp: 6220 - 0.29 %