OV - Compare Loops

Loops

▶kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c: 131 - 188.47 %

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 gcc_default						Run armclang_3						Run gcc_5
Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p/kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c: 131-131					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/gcc/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p/kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c: 131-131					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/armclang_3/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p/kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c: 131-131					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/gcc_5/_deps/kleidiai_download-src/kai/ukernels/matmul/matmul_clamp_f32_qsi8d32p_qsi4c32p/kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c: 131-131
2354	2.26	2.45	46.25	28.14	55.37	2189	2.25	2.47	47.25	28.14	55.37	2528	2.25	2.38	47.35	28.14	55.37	2097	2.27	2.42	47.63	28.14	55.37

Sum on 1 analyzed binary loop (libggml-cpu.so - 2354)						Sum on 1 analyzed binary loop (libggml-cpu.so - 2189)						Sum on 1 analyzed binary loop (libggml-cpu.so - 2528)						Sum on 1 analyzed binary loop (libggml-cpu.so - 2097)
Analysis					Count	Analysis					Count	Analysis					Count	Analysis					Count
Data Access Issues						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 constant non-unit stride data access					1
Vectorization Roadblocks						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 constant non-unit stride data access					1

▶quants.c: 2506 - 46.48 %

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 gcc_default						Run armclang_3						Run gcc_5
Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/arm/quants.c: 2506-2558 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/arm/quants.c: 2565-2590					Loop Source Regions	/opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 1122-1122 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 1234-1234 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 4144-4144 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 12569-12569 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 12710-12710 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 12893-12893 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 21097-21097 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 21435-21435 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 25379-25379 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 25654-25654 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 28759-28759 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/arm/quants.c: 2506-2506 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/arm/quants.c: 2569-2575					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/arm/quants.c: 2506-2558 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/arm/quants.c: 2565-2590					Loop Source Regions	/opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 1122-1122 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 1234-1234 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 4144-4144 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 12569-12569 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 12710-12710 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 12893-12893 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 21097-21097 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 21435-21435 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 25379-25379 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 25654-25654 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 28759-28759 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/arm/quants.c: 2506-2506 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/arch/arm/quants.c: 2572-2575
2236	0.57	0.64	12.10	50.56	65.8	2056	0.54	0.58	11.15	49.72	67.11	2425	0.55	0.61	12.09	50.56	65.8	1960	0.53	0.57	11.14	49.44	66.92

Sum on 1 analyzed binary loop (libggml-cpu.so - 2236)						Sum on 1 analyzed binary loop (libggml-cpu.so - 2056)						Sum on 1 analyzed binary loop (libggml-cpu.so - 2425)						Sum on 1 analyzed binary loop (libggml-cpu.so - 1960)
Analysis					Count	Analysis					Count	Analysis					Count	Analysis					Count
Data Access Issues						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 constant non-unit stride data access					1
Vectorization Roadblocks						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 constant non-unit stride data access					1

▶vec.cpp: 385 - 1.44 %

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	Max Time Over Threads (s)	Time w.r.t. Wall Time (s)	Cov (%)	Vect. Ratio (%)	Vector Length Use (%)
Run orig_default						Run gcc_default						Run armclang_3		Run gcc_5
Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.cpp: 385-387 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.h: 1008-1034					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.cpp: 385-387 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.h: 1009-1023 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.h: 1031-1034					Loop Source Regions		Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.cpp: 385-387 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.h: 1009-1023 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.h: 1031-1034
886	0.26	0.04	0.84	70.45	83.38	756	0.11	0.02	0.31	80	97.68			714	0.13	0.01	0.29	91.43	97.34

Sum on 1 analyzed binary loop (libggml-cpu.so - 886)						Sum on 1 analyzed binary loop (libggml-cpu.so - 756)						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 - 714)
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
Data Access Issues						Data Access Issues								Data Access Issues
Presence of constant non-unit stride data access						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						Presence of constant non-unit stride data access					1			Presence of constant non-unit stride data access					1

▶ops.cpp: 6220 - 1.38 %

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 gcc_default		Run armclang_3						Run gcc_5
Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6220-6220 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6229-6230 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6238-6245					Loop Source Regions		Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6220-6220 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6229-6229 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6238-6245					Loop Source Regions
1408	0.08	0.04	0.70	0	24.52			1596	0.06	0.03	0.68	0	26.14

Sum on 1 analyzed binary loop (libggml-cpu.so - 1408)						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 - 1596)						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
Control Flow Issues								Control Flow Issues
Presence of calls					1			Presence of calls					1
Data Access Issues								Data Access Issues
Presence of constant non-unit stride data access					1			Presence of constant non-unit stride data access					1
Vectorization Roadblocks								Vectorization Roadblocks
Presence of calls					1			Presence of calls					1
Presence of constant non-unit stride data access					1			Presence of constant non-unit stride data access					1

▶vec.cpp: 231 - 1.25 %

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	Max Time Over Threads (s)	Time w.r.t. Wall Time (s)	Cov (%)	Vect. Ratio (%)	Vector Length Use (%)
Run orig_default						Run gcc_default						Run armclang_3		Run gcc_5
Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.cpp: 231-262					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.cpp: 231-262					Loop Source Regions		Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.cpp: 231-262
879	0.04	0.02	0.38	100	100	754	0.05	0.02	0.43	100	100			712	0.05	0.02	0.44	100	100

Sum on 1 analyzed binary loop (libggml-cpu.so - 879)						Sum on 1 analyzed binary loop (libggml-cpu.so - 754)						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 - 712)
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
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

▶ops.cpp: 6210 - 1.19 %

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 (%)	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 gcc_default						Run armclang_3		Run gcc_5
Loop Source Regions		Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6210-6211 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6220-6230 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6238-6245					Loop Source Regions		Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6211-6211 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6223-6231
		775	0.07	0.04	0.78	1.83	20.42			725	0.04	0.02	0.41	0	24.77

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 - 775)						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 - 725)
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
		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
		Presence of 2 to 4 paths					1			Presence of 2 to 4 paths					0
		Data Access Issues								Data Access Issues
		Presence of indirect access					1			Presence of indirect access
		Vectorization Roadblocks								Vectorization Roadblocks
		Presence of calls					1			Presence of calls					1
		Presence of 2 to 4 paths					1			Presence of 2 to 4 paths					0
		Presence of more than 4 paths					0			Presence of more than 4 paths					1
		Presence of indirect access					1			Presence of indirect access					0

▶vec.h: 411 - 0.97 %

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	Max Time Over Threads (s)	Time w.r.t. Wall Time (s)	Cov (%)	Vect. Ratio (%)	Vector Length Use (%)
Run orig_default						Run gcc_default						Run armclang_3		Run gcc_5
Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.h: 411-458					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.h: 411-458					Loop Source Regions		Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.h: 411-458
1718	0.04	0.02	0.32	100	100	1525	0.04	0.02	0.32	96.97	96.99			1495	0.04	0.02	0.33	96.97	96.99

Sum on 1 analyzed binary loop (libggml-cpu.so - 1718)						Sum on 1 analyzed binary loop (libggml-cpu.so - 1525)						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 - 1495)
Analysis					Count	Analysis					Count	Analysis	Count	Analysis					Count
Loop Computation Issues						Loop Computation Issues								Loop Computation Issues
Presence of a large number of scalar integer instructions						Presence of a large number of scalar integer instructions					1			Presence of a large number of scalar integer instructions					1
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: 96 - 0.70 %

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 gcc_default						Run armclang_3						Run gcc_5
Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/pack/kai_lhs_quant_pack_qsi8d32p4x8sb_f32_neon.c: 96-96 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/pack/kai_lhs_quant_pack_qsi8d32p4x8sb_f32_neon.c: 111-142 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/pack/kai_lhs_quant_pack_qsi8d32p4x8sb_f32_neon.c: 151-258					Loop Source Regions	/opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 898-898 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 2943-2943 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 6853-6853 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 7154-7154 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 9945-9945 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 10568-10568 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 10805-10805 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 12531-12531 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 15594-15594 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 16191-16191 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 17374-17374 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 17537-17537 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 22069-22069 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 22134-22134 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 24797-24797 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 24809-24809 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/gcc/_deps/kleidiai_download-src/kai/ukernels/matmul/pack/kai_lhs_quant_pack_qsi8d32p4x8sb_f32_neon.c: 96-96 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/gcc/_deps/kleidiai_download-src/kai/ukernels/matmul/pack/kai_lhs_quant_pack_qsi8d32p4x8sb_f32_neon.c: 262-262					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/armclang_3/_deps/kleidiai_download-src/kai/ukernels/matmul/pack/kai_lhs_quant_pack_qsi8d32p4x8sb_f32_neon.c: 96-96 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/armclang_3/_deps/kleidiai_download-src/kai/ukernels/matmul/pack/kai_lhs_quant_pack_qsi8d32p4x8sb_f32_neon.c: 111-142 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/armclang_3/_deps/kleidiai_download-src/kai/ukernels/matmul/pack/kai_lhs_quant_pack_qsi8d32p4x8sb_f32_neon.c: 151-258					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/gcc_5/_deps/kleidiai_download-src/kai/ukernels/matmul/pack/kai_lhs_quant_pack_qsi8d32p4x8sb_f32_neon.c: 96-96 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/gcc_5/_deps/kleidiai_download-src/kai/ukernels/matmul/pack/kai_lhs_quant_pack_qsi8d32p4x8sb_f32_neon.c: 262-262 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 898-898 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 2943-2943 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 6853-6853 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 7154-7154 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 9945-9945 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 10568-10568 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 10805-10805 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 12531-12531 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 15594-15594 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 16191-16191 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 17374-17374 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 17537-17537 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 22069-22069 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 22134-22134 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 24797-24797 /opt/arm/gcc-14.2.0_AmazonLinux-2023/lib/gcc/aarch64-linux-gnu/14.2.0/include/arm_neon.h: 24809-24809
2317	0.23	0.01	0.18	77.23	96.44	2158	0.18	0.01	0.16	76.21	96.75	2496	0.21	0.01	0.18	77.23	96.44	2066	0.20	0.01	0.18	76.21	96.75

Sum on 1 analyzed binary loop (libggml-cpu.so - 2317)						Sum on 1 analyzed binary loop (libggml-cpu.so - 2158)						Sum on 1 analyzed binary loop (libggml-cpu.so - 2496)						Sum on 1 analyzed binary loop (libggml-cpu.so - 2066)
Analysis					Count	Analysis					Count	Analysis					Count	Analysis					Count
Loop Computation Issues						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	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	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
Data Access Issues						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 constant non-unit stride data access					1
Vectorization Roadblocks						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 constant non-unit stride data access					1

▶ops.cpp: 6238 - 0.60 %

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 gcc_default		Run armclang_3		Run gcc_5
Loop Source Regions		Loop Source Regions		Loop Source Regions		Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6238-6245
						726	0.05	0.03	0.60	0	29.85

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 - 726)
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
						Control Flow Issues
						Presence of calls					1
						Presence of 2 to 4 paths					1
						Data Access Issues
						Presence of constant non-unit stride data access					1
						Presence of indirect access					1
						Vectorization Roadblocks
						Presence of calls					1
						Presence of 2 to 4 paths					1
						Presence of constant non-unit stride data access					1
						Presence of indirect access					1

▶vec.cpp: 311 - 0.57 %

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 gcc_default		Run armclang_3						Run gcc_5
Loop Source Regions		Loop Source Regions		Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.cpp: 311-316					Loop Source Regions
				908	0.06	0.03	0.57	50	100

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 - 908)						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

▶ggml-cpu.c: 3228 - 0.36 %

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 gcc_default						Run armclang_3						Run gcc_5
Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 3228-3229 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/./ggml-impl.h: 389-404					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 3228-3229 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/./ggml-impl.h: 389-404					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 3228-3229 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/./ggml-impl.h: 389-404					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 3228-3229 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/./ggml-impl.h: 389-404
0	0.02	0.00	0.09	92.5	98.75	5	0.03	0.01	0.10	88.37	93.9	0	0.03	0.00	0.10	91.67	98.61	4	0.02	0.00	0.07	82.05	93.27

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

▶binary-ops.cpp: 10 - 0.32 %

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 gcc_default						Run armclang_3						Run gcc_5
Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/binary-ops.cpp: 10-10 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/binary-ops.cpp: 31-32					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/binary-ops.cpp: 10-10 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/binary-ops.cpp: 31-32					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/binary-ops.cpp: 10-10 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/binary-ops.cpp: 31-32					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/binary-ops.cpp: 10-10 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/binary-ops.cpp: 31-32
413	0.04	0.00	0.07	0	25	419	0.04	0.00	0.05	25	100	437	0.05	0.00	0.08	0	25	362	0.04	0.01	0.11	0	18.75

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 - 0.32 %

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 gcc_default		Run armclang_3						Run gcc_5
Loop Source Regions		Loop Source Regions		Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.h: 491-497					Loop Source Regions
				1908	0.04	0.02	0.32	40	100

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 - 1908)						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

▶binary-ops.cpp: 18 - 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	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 gcc_default						Run armclang_3						Run gcc_5
Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/binary-ops.cpp: 18-18 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/binary-ops.cpp: 31-32					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/binary-ops.cpp: 18-18 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/binary-ops.cpp: 31-32					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/binary-ops.cpp: 18-18 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/binary-ops.cpp: 31-32					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/binary-ops.cpp: 18-18 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/binary-ops.cpp: 31-32
520	0.05	0.01	0.09	0	23.68	493	0.03	0.00	0.03	25	100	541	0.04	0.00	0.09	0	23.68	440	0.04	0.00	0.08	0	18.75

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

▶kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c: 127 - 0.27 %

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 gcc_default						Run armclang_3						Run gcc_5
Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/pack/kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c: 127-128 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/pack/kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c: 137-139					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/gcc/_deps/kleidiai_download-src/kai/ukernels/matmul/pack/kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c: 127-132 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/gcc/_deps/kleidiai_download-src/kai/ukernels/matmul/pack/kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c: 138-139					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/armclang_3/_deps/kleidiai_download-src/kai/ukernels/matmul/pack/kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c: 127-128 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/armclang_3/_deps/kleidiai_download-src/kai/ukernels/matmul/pack/kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c: 137-139					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/gcc_5/_deps/kleidiai_download-src/kai/ukernels/matmul/pack/kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c: 127-132 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/gcc_5/_deps/kleidiai_download-src/kai/ukernels/matmul/pack/kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c: 138-139
2337	0.31	0.00	0.08	0	50	2172	0.24	0.00	0.06	0	50	2511	0.27	0.00	0.07	0	50	2079	0.29	0.00	0.07	0	50

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: 1045 - 0.25 %

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 gcc_default		Run armclang_3						Run gcc_5
Loop Source Regions		Loop Source Regions		Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.cpp: 390-391 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.h: 1045-1058 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.h: 1072-1072					Loop Source Regions
				916	0.10	0.01	0.25	73.33	91.67

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 - 916)						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
				Control Flow Issues
				Vectorization Roadblocks
				Presence of more than 4 paths					1

▶ops.cpp: 4325 - 0.25 %

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 gcc_default						Run armclang_3						Run gcc_5
Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 4325-4326					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 4325-4326					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 4325-4326					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 4325-4326
1249	0.05	0.00	0.09	94.12	97.06	1148	0.04	0.00	0.09	0	26.56	1353	0.02	0.00	0.02	16.67	54.17	1124	0.03	0.00	0.05	100	82.5

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.						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

▶ops.cpp: 6446 - 0.22 %

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 gcc_default						Run armclang_3						Run gcc_5
Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6446-6447 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6453-6456					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6446-6447 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6453-6457					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6446-6447 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6453-6456					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 6446-6457
1403	0.03	0.00	0.06	35.29	39.71	785	0.02	0.00	0.05	42.86	53.57	1591	0.02	0.00	0.07	55.56	43.75	742	0.02	0.00	0.05	0	18.91

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

▶<unknown>: 0 - 0.18 %

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	Max Time Over Threads (s)	Time w.r.t. Wall Time (s)	Cov (%)	Vect. Ratio (%)	Vector Length Use (%)
Run orig_default						Run gcc_default						Run armclang_3						Run gcc_5
Loop Source Regions						Loop Source Regions						Loop Source Regions						Loop Source Regions
1356	0.00	0.00	0.00	0	0	3762	0.01	0.00	0.00	0	0	2872	0.00	0.00	0.00	0	0	3597	0.01	0.00	0.00	0	0
2724	0.00	0.00	0.00	0	0	1648	0.01	0.00	0.00	0	0	3078	0.00	0.00	0.00	0	0	3388	0.00	0.00	0.00	0	0
2600	0.00	0.00	0.00	0	0	4530	0.00	0.00	0.00	0	0	3222	0.01	0.00	0.00	0	0	3903	0.01	0.00	0.00	0	0
2975	0.00	0.00	0.00	0	0	3828	0.01	0.00	0.00	0	0	1510	0.00	0.00	0.00	0	0	1733	0.00	0.00	0.00	0	0
1398	0.01	0.00	0.00	0	0	4132	0.00	0.00	0.00	0	0	1554	0.00	0.00	0.00	0	0	3491	0.01	0.00	0.00	0	0
2558	0.00	0.00	0.00	0	0	1856	0.02	0.00	0.00	0	0	3080	0.00	0.00	0.00	0	0	3745	0.01	0.00	0.00	0	0
2741	0.00	0.00	0.00	0	0	3977	0.00	0.00	0.00	0	0	2862	0.00	0.00	0.00	0	0	461	0.00	0.00	0.00	0	0
2855	0.00	0.00	0.00	0	0	3983	0.00	0.00	0.00	0	0	3369	0.01	0.00	0.00	0	0	588	0.01	0.00	0.00	0	0
2858	0.01	0.00	0.00	0	0	4130	0.00	0.00	0.00	0	0	3245	0.00	0.00	0.00	0	0	22	0.01	0.00	0.00	0	0
2565	0.00	0.00	0.00	0	0	4159	0.00	0.00	0.00	0	0	705	0.01	0.00	0.01	0	0	33	0.00	0.00	0.00	0	0
2740	0.01	0.00	0.00	0	0	2115	0.01	0.00	0.00	0	0	72	0.00	0.00	0.00	0	0	2019	0.01	0.00	0.00	0	0
1326	0.01	0.00	0.00	0	0	1522	0.01	0.00	0.00	0	0	2470	0.00	0.00	0.00	0	0	49	0.01	0.00	0.00	0	0
365	0.01	0.00	0.00	0	0	70	0.01	0.00	0.01	0	0	95	0.00	0.00	0.00	0	0	2036	0.01	0.00	0.00	0	0
57	0.00	0.00	0.00	0	0	63	0.00	0.00	0.00	0	0	1911	0.01	0.00	0.00	0	0	1185	0.01	0.00	0.00	0	0
1719	0.01	0.00	0.00	0	0	2118	0.01	0.00	0.00	0	0	407	0.01	0.00	0.01	0	0	1496	0.00	0.00	0.00	0	0
1329	0.01	0.00	0.01	0	0	1526	0.01	0.00	0.01	0	0							59	0.01	0.00	0.01	0	0
1724	0.01	0.00	0.00	0	0	57	0.01	0.00	0.00	0	0							733	0.01	0.00	0.01	0	0
305	0.01	0.00	0.00	0	0	603	0.01	0.00	0.00	0	0							1501	0.01	0.00	0.01	0	0
531	0.01	0.00	0.00	0	0													86	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

▶vec.h: 646 - 0.15 %

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	Max Time Over Threads (s)	Time w.r.t. Wall Time (s)	Cov (%)	Vect. Ratio (%)	Vector Length Use (%)
Run orig_default						Run gcc_default						Run armclang_3		Run gcc_5
Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.h: 646-653					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.h: 646-653					Loop Source Regions		Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.h: 646-653
1252	0.01	0.00	0.01	100	100	1147	0.01	0.00	0.01	100	100			1128	0.02	0.00	0.01	100	100
1722	0.02	0.00	0.05	100	100	1520	0.01	0.00	0.03	100	100			1499	0.01	0.00	0.04	100	100

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

▶kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c: 115 - 0.14 %

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 gcc_default						Run armclang_3						Run gcc_5
Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/build/_deps/kleidiai_download-src/kai/ukernels/matmul/pack/kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c: 115-118					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/gcc/_deps/kleidiai_download-src/kai/ukernels/matmul/pack/kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c: 115-118					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/armclang_3/_deps/kleidiai_download-src/kai/ukernels/matmul/pack/kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c: 115-118					Loop Source Regions
2338	0.16	0.00	0.04	0	31.25	2170	0.28	0.00	0.07	0	31.25	2512	0.13	0.00	0.03	0	31.25

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.						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

▶ops.cpp: 8817 - 0.08 %

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 (%)	ASM Loop ID	Vector Length Use (%)
Run orig_default		Run gcc_default						Run armclang_3		Run gcc_5
Loop Source Regions		Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 8817-8819 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/simd-mappings.h: 51-51					Loop Source Regions		Loop Source Regions
		1516	0.02	0.00	0.08	0	27.08

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

▶ggml-cpu.c: 1125 - 0.08 %

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 gcc_default						Run armclang_3						Run gcc_5
Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 1125-1130 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 1142-1142 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 1198-1198 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 1371-1379 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 1386-1395					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 1125-1130 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 1142-1142 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 1371-1395					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 1125-1130 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 1142-1142 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 1198-1198 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 1371-1379 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 1386-1395					Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 1125-1130 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 1142-1142 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 1371-1395
54	0.01	0.00	0.02	0	46.88	62	0.01	0.00	0.02	0	48.05	54	0.01	0.00	0.02	0	46.88	51	0.01	0.00	0.01	0	47.55

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

▶ggml-cpu.c: 1183 - 0.07 %

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 gcc_default		Run armclang_3						Run gcc_5
Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 1183-1187 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 1193-1194					Loop Source Regions		Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 1183-1187 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 1193-1194					Loop Source Regions
53	0.01	0.00	0.05	0	50			53	0.01	0.00	0.02	0	50

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: 677 - 0.06 %

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 gcc_default		Run armclang_3						Run gcc_5
Loop Source Regions		Loop Source Regions		Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/vec.h: 677-682					Loop Source Regions
				1914	0.01	0.00	0.04	50	100
				1354	0.01	0.00	0.02	50	100

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

▶ggml-cpu.c: 1193 - 0.04 %

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 (%)	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 gcc_default						Run armclang_3		Run gcc_5
Loop Source Regions		Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 1193-1194					Loop Source Regions		Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ggml-cpu.c: 1193-1194
		59	0.01	0.00	0.02	0	43.06			53	0.01	0.00	0.03	0	44.55

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

▶ops.cpp: 8825 - 0.04 %

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 (%)	ASM Loop ID	Vector Length Use (%)
Run orig_default		Run gcc_default						Run armclang_3		Run gcc_5
Loop Source Regions		Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 8825-8826					Loop Source Regions		Loop Source Regions
		1523	0.02	0.00	0.04	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

▶kleidiai.cpp: 535 - 0.02 %

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 (%)	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 gcc_default						Run armclang_3		Run gcc_5
Loop Source Regions		Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/kleidiai/kleidiai.cpp: 535-540 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/kleidiai/kleidiai.cpp: 547-547					Loop Source Regions		Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/kleidiai/kleidiai.cpp: 535-540 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/kleidiai/kleidiai.cpp: 547-547
		2128	0.01	0.00	0.01	0	50			2037	0.01	0.00	0.01	0	50

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

▶traits.cpp: 13 - 0.02 %

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 (%)	ASM Loop ID	Vector Length Use (%)
Run orig_default		Run gcc_default						Run armclang_3		Run gcc_5
Loop Source Regions		Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/traits.cpp: 13-17					Loop Source Regions		Loop Source Regions
		382	0.01	0.00	0.02	0	47.4

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

▶simd-mappings.h: 51 - 0.01 %

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 gcc_default		Run armclang_3		Run gcc_5
Loop Source Regions		Loop Source Regions		Loop Source Regions		Loop Source Regions	/home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/ops.cpp: 8817-8819 /home/eoseret/Tools/QaaS/qaas_runs/ip-172-31-47-249.ec2.internal/176-138-2040/llama.cpp/build/llama.cpp/ggml/src/ggml-cpu/simd-mappings.h: 51-51
						1491	0.01	0.00	0.01	0	34.72

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

Report Configuration

Loops

▶kai_matmul_clamp_f32_qsi8d32p4x8_qsi4c32p4x8_16x4_neon_i8mm.c: 131 - 188.47 %

▶quants.c: 2506 - 46.48 %

▶vec.cpp: 385 - 1.44 %

▶ops.cpp: 6220 - 1.38 %

▶vec.cpp: 231 - 1.25 %

▶ops.cpp: 6210 - 1.19 %

▶vec.h: 411 - 0.97 %

▶kai_lhs_quant_pack_qsi8d32p4x8sb_f32_neon.c: 96 - 0.70 %

▶ops.cpp: 6238 - 0.60 %

▶vec.cpp: 311 - 0.57 %

▶ggml-cpu.c: 3228 - 0.36 %

▶binary-ops.cpp: 10 - 0.32 %

▶vec.h: 491 - 0.32 %

▶binary-ops.cpp: 18 - 0.29 %

▶kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c: 127 - 0.27 %

▶vec.h: 1045 - 0.25 %

▶ops.cpp: 4325 - 0.25 %

▶ops.cpp: 6446 - 0.22 %

▶<unknown>: 0 - 0.18 %

▶vec.h: 646 - 0.15 %

▶kai_rhs_pack_nxk_qsi4c32pscalef16_qsu4c32s16s0.c: 115 - 0.14 %

▶ops.cpp: 8817 - 0.08 %

▶ggml-cpu.c: 1125 - 0.08 %

▶ggml-cpu.c: 1183 - 0.07 %

▶vec.h: 677 - 0.06 %

▶ggml-cpu.c: 1193 - 0.04 %

▶ops.cpp: 8825 - 0.04 %

▶kleidiai.cpp: 535 - 0.02 %

▶traits.cpp: 13 - 0.02 %

▶simd-mappings.h: 51 - 0.01 %