Add counters to smr.

[jeff]

Summary

These should eventually be per-smr I suspect but for now this is helpful.

[rlibby]

Logic looks fine. I trust you've checked any performance effects.

In advance you count on entry, and then also on the slow path, whereas in poll you count either fast or slow. Did you mean to approach those differently?

sys/kern/subr_smr.c

168 ↗

(On Diff #67646)

Should they be initialized with EARLY_COUNTER? Since it seems that UMA SMR is available after SI_SUB_VM.

[markj]

sys/kern/subr_smr.c

39 ↗

(On Diff #67646)

Sort includes while here?

316 ↗

(On Diff #67646)

This and the call below can be counter_u64_add_protected() since you're in a critical section.

[jeff]

In D23463#514971, @rlibby wrote:

Logic looks fine. I trust you've checked any performance effects.

In advance you count on entry, and then also on the slow path, whereas in poll you count either fast or slow. Did you mean to approach those differently?

It might make more sense to count all polls and then number of polls that scan. I will look at that.