You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently nanobench uses a few hardware performance counters - number of instructions retired, cycles, etc..
I would like to have data on cache performance. Specifically, there are counters for cache references and cache misses.
In Linux, there are portable counters PERF_COUNT_HW_CACHE_REFERENCES and PERF_COUNT_HW_CACHE_MISSES,
and also there are more specialized events PERF_COUNT_HW_CACHE_L1D etc.
Is there a reason to not use these? If not, I may be able to work on this.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
Currently nanobench uses a few hardware performance counters - number of instructions retired, cycles, etc..
I would like to have data on cache performance. Specifically, there are counters for cache references and cache misses.
In Linux, there are portable counters
PERF_COUNT_HW_CACHE_REFERENCES
andPERF_COUNT_HW_CACHE_MISSES
,and also there are more specialized events
PERF_COUNT_HW_CACHE_L1D
etc.Is there a reason to not use these? If not, I may be able to work on this.
Beta Was this translation helpful? Give feedback.
All reactions