Is your feature request related to a problem? Please describe.
I am trying to optimize an application and would like better visibility into how to get good reuse.
Describe alternatives you've considered
I've considered using the profiler and microbenchmarks to get XCD-level information and hacking together my own data analysis infrastructure.
Is your feature request related to a problem? Please describe. I am trying to optimize an application and would like better visibility into how to get good reuse.
Describe the solution you'd like I would like the option for the memory map to show memory metrics hierarchically (for example, per-XCD) as the memory is disaggregated across the hierarchy (as mentioned here : https://www.amd.com/content/dam/amd/en/documents/instinct-tech-docs/white-papers/amd-cdna-3-white-paper.pdf)
Describe alternatives you've considered I've considered using the profiler and microbenchmarks to get XCD-level information and hacking together my own data analysis infrastructure.
Thanks!