OV - - Loops

MAQAO

options

Loops Index

35 loops have been discarded from the report because their ratio ((Max Inclusive Time Over Threads * 100) / Max Thread Active Time) is lower than the threshold set by object_coverage_threshold (0.1%). It represents about 0.05% of the application. To include them, change the value of object_coverage_threshold in the experiment directory configuration file, then rerun the command with the additionnal parameter --force-static-analysis

▶Filters

Loop id	Source Location	Source Function	Level	Max Thread Time / Walltime icx_3 (%)	Exclusive Coverage icx_3 (%)	Inclusive Coverage icx_3 (%)	Max Exclusive Time Over Threads icx_3 (s)	Max Inclusive Time Over Threads icx_3 (s)	Exclusive Time w.r.t. Wall Time icx_3 (s)	Inclusive Time w.r.t. Wall Time icx_3 (s)	Nb Threads icx_3	GFLOPS icx_3	Vectorization Ratio (%)	Vector Length Use (%)	Speedup If No Scalar Integer	Speedup If FP Vectorized	Speedup If Fully Vectorized	Speedup If Perfect Load Balancing icx_3	Stride 0	Stride 1	Stride n	Stride Unknown	Stride Indirect	Array Access Efficiency
5766	libggml-cpu.so - quants.c:108-1042 [...]	ggml_vec_dot_q8_0_q8_0	Single	82.58	61.71	61.71	56.87	56.87	37.54	37.54	192	53.96	NA	NA	NA	NA	NA	1.53	NA	NA	NA	NA	NA	0.00
236	libggml-cpu.so - ggml-cpu.c:1164-1198 [...]	ggml_compute_forward_mul_mat	InBetween	0.25	0.12	0.13	0.17	0.19	0.07	0.08	192	45.49	0	11.14	1	1	9.6	2.43	NA	NA	NA	NA	NA	0.00
5326	libggml-cpu.so - sgemm.cpp:138-1044 [...]	_ZN12_GLOBAL__N_115tinyBLAS_Q0_AVXI10block_q8_0S1_fE7gemm4xNILi2EEEvllll.A	Innermost	0.19	0.11	0.11	0.13	0.13	0.07	0.07	189	103.91	NA	NA	NA	NA	NA	1.91	NA	NA	NA	NA	NA	0.00
5300	libggml-cpu.so - sgemm.cpp:138-1044 [...]	_ZN12_GLOBAL__N_115tinyBLAS_Q0_AVXI10block_q8_0S1_fE7gemm4xNILi4EEEvllll.A	Innermost	0.19	0.11	0.11	0.13	0.13	0.07	0.07	189	224.01	NA	NA	NA	NA	NA	1.93	NA	NA	NA	NA	NA	0.00
1779	libggml-cpu.so - vec.cpp:311-316	ggml_vec_dot_f16	Single	0.35	0.04	0.04	0.24	0.24	0.02	0.02	33	700.95	100	66.67	1	1	1.57	1.88	0	2	0	0	0	100.00
4189	libggml-cpu.so - vec.h:491-497	ggml_compute_forward_flash_attn_ext	Innermost	0.32	0.03	0.03	0.22	0.22	0.02	0.02	33	1486.96	100	75	1	1	1.45	1.83	0	2	0	0	0	100.00
5745	libggml-cpu.so - quants.c:298-355 [...]	quantize_row_q8_0	Single	0.13	0.03	0.03	0.09	0.09	0.02	0.02	151	0.50	60.7	29.66	1	1.11	2.67	3.55	0	2	0	0	0	100.00
262	libggml-cpu.so - ggml-cpu.c:533-2891 [...]	ggml_graph_compute_thread	Single	0.12	0.02	0.02	0.08	0.08	0.01	0.01	173	50.77	0	9.82	1	1	14.64	4.94	NA	NA	NA	NA	NA	0.00
235	libggml-cpu.so - ggml-cpu.c:1163-1198 [...]	ggml_compute_forward_mul_mat	InBetween	0.12	0.02	0.15	0.08	0.21	0.01	0.09	186	60.92	0	12.5	1	1	8	5.51	NA	NA	NA	NA	NA	0.00
3433	libggml-cpu.so - ops.cpp:6220-6245 [...]	ggml_compute_forward_rope_f32	Innermost	0.10	0.02	0.02	0.07	0.07	0.01	0.01	192	288.08	15.56	9.17	1	1.32	11.46	6.96	1	1	0	0	0	100.00
233	libggml-cpu.so - ggml-cpu.c:1193-1194	ggml_compute_forward_mul_mat	Innermost	0.10	0.01	0.01	0.07	0.07	0.01	0.01	164	123.73	0	12.5	1	1	8	7.87	1	0	0	0	0	100.00
4180	libggml-cpu.so - ops.cpp:8759-8927 [...]	ggml_compute_forward_flash_attn_ext	InBetween	0.12	0.01	0.04	0.08	0.23	0.01	0.03	33	4020.85	53.38	48.02	1.41	1.01	1.36	2.56	NA	NA	NA	NA	NA	0.00
1793	libggml-cpu.so - vec.h:1084-1116 [...]	ggml_vec_swiglu_f32	Single	0.33	0.00	0.00	0.23	0.23	0.00	0.00	7	3681.88	100	100	1	1	1	6.19	0	3	0	0	0	100.00
901	libggml-cpu.so - binary-ops.cpp:10-32 [...]	ggml_compute_forward_add_non_quantized	Innermost	0.22	0.00	0.00	0.15	0.15	0.00	0.00	1	164.71	0	6.25	1	1.33	16	1	0	3	0	0	0	100.00
2864	libggml-cpu.so - ops.cpp:4325-4326	ggml_compute_forward_rms_norm	Innermost	0.13	0.00	0.00	0.09	0.09	0.00	0.00	1	554.33	0	7.81	1	1.98	13.02	1	0	1	0	0	0	100.00
1149	libggml-cpu.so - binary-ops.cpp:18-32 [...]	ggml_compute_forward_mul	Innermost	0.12	0.00	0.00	0.08	0.08	0.00	0.00	1	356.35	0	6.25	1	1.33	16	1	0	3	0	0	0	100.00
3924	exec - sampling.cpp:125-126	_ZN14common_sampler10set_logitsEP13llama_contexti.A	Single	0.10	0.00	0.00	0.07	0.07	0.00	0.00	1	0.00	33.33	8.33	1.5	1	12	1	0	2	0	0	0	100.00

×