Should Handle Error.vi be non-reentrant?

[r-good]

I just figured out that Handle Error.vi is a non-reentrant vi. This is surprising to me because everything else in Actor Core.vi, and including that vi, are all reentrant. Is this intentional? If so, why? it seems to me that all of these vis should be reentrant to allow for true asynchronous execution. I could totally be wrong in my understanding here, but this struck me as odd

[niACS]

From Stephen Loftus-Mercer:

Handle Error.vi could be reentrant. Just an oversight.

To clarify: not entirely an oversight. I tend to only add reentrancy when necessary. It makes debugging notably harder and will bloat the size of applications notably (one extra reentrant clone per actor potentially). The gain is the thread friction decrease. The oversight was I never really analyzed whether it was worth doing. It never came up... Casey [Lamers - ed] with his bijillion actors wasn't seeing it as a bottleneck... and I liked the ease of setting a breakpoint to catch an error. I wasn't really looking for reasons to change it.

Changing it at this point would be a breaking change as all the overrides would need to be made reentrant when upgrading.

[niACS]

So, 1) not necessary and 2) a breaking change. I think this stays as is.

[CaseyZS]

Discussed by the SteerCo and there was general agreement that this should stay as it is because of the reasons Allen enumerated.