CRITICAL Incorrect application of interlocked increment

[michael-michalski]

The static integer might be incremented additionally between the atomic operation and its application in a parallel scenario. It is the result of the atomic operation, which contains the correct value.

https://learn.microsoft.com/en-us/dotnet/api/system.threading.interlocked.increment?view=net-8.0

Returns
UInt32

The value of the variable immediately after the increment operation finished.

[aevitas]

Good catch. Did you run into any actual instances of this occurring? The odds seem incredibly small, though I agree it is theoretically possible for two threads to read the same increment in the current implementation.

[KrFriis]

Oops, I was going fast through a review, and a coworker had sent this to me.
I accidentally approved this, I cannot figure out how to revoke whatever I did.

My bad.

[michael-michalski]

Yes, it occurred in a real scenario :)

[aevitas]

Yes, it is based on an event from the trenches :)

Hope it wasn't too bad, thanks for the PR - I'll get a release out asap

[michael-michalski]

I found a fairly quick remedy, thank you :)