On Wed, Mar 24, 2021 at 09:44:21AM -0400, Josef Bacik wrote:
> Looking at perf data for a fio workload I noticed that we were spending
> a pretty large chunk of time (around 5%) doing percpu_counter_sum() in
> need_preemptive_reclaim. This is silly, as we only want to know if we
> have more ordere
Looking at perf data for a fio workload I noticed that we were spending
a pretty large chunk of time (around 5%) doing percpu_counter_sum() in
need_preemptive_reclaim. This is silly, as we only want to know if we
have more ordered than delalloc to see if we should be counting the
delayed items in