rocm-systems/tests/workloads/vcopy/MI100/SQ_INST_LEVEL_VMEM.csv

1	Dispatch_ID	Kernel_Name	GPU_ID	queue-id	queue-index	pid	tid	Grid_Size	Workgroup_Size	LDS_Per_Workgroup	Scratch_Per_Workitem	Arch_VGPR	Accum_VGPR	SGPR	wave_size	sig	obj	SQ_INSTS_VMEM	SQ_INST_LEVEL_VMEM	SQ_ACCUM_PREV_HIRES	DispatchNs	Start_Timestamp	End_Timestamp	CompleteNs
2	0	vecCopy(double*, double*, double*, int, int)	2	0	0	1330557	1330557	1048576	256	0	0	8	8	16	64	0x0	0x7f6ef60aae80	32768	646195	82717292	5862035909155076	5862046126501116	5862046126524956	5862035917308328
3	1	vecCopy(double*, double*, double*, int, int)	2	0	2	1330557	1330557	1048576	256	0	0	8	8	16	64	0x0	0x7f6ef60aae80	32768	679196	86941576	5862035917329729	5862046126616796	5862046126635516	5862035917590542
4	2	vecCopy(double*, double*, double*, int, int)	2	0	4	1330557	1330557	1048576	256	0	0	8	8	16	64	0x0	0x7f6ef60aae80	32768	648200	82969328	5862035917619637	5862046126657436	5862046126676476	5862035917766324