Would need to be rewritten to harden against thread aborts. Let me know if interested.

Related Reddit post by @MarkCiliaVincenti which provides a nice background summary.

While in the longer term, this is, of course, very interesting, I can't see it being much of a priority given that there is no performance gain unless running on .NET 9. Just my thoughts.

Related Reddit post by @MarkCiliaVincenti which provides a nice background summary.

While in the longer term, this is, of course, very interesting, I can't see it being much of a priority given that there is no performance gain unless running on .NET 9. Just my thoughts.

That is correct. The performance gain is only on .NET 9 or greater, but there is no performance loss for .NET 8 or earlier.

More and more people will start adopting .NET 9, and later .NET 10, at which point the assumption is that .NET 8 would still be targeted by Lucene.NET. Therefore it stands to reason that one should ask whether or not it should be done. If the answer is yes, delaying it provides no value, except sensible ones such as waiting for .NET 9.0 to be officially released.

@MarkCiliaVincenti Thank you for this PR, but there are a few reasons why we are going to close this at this time. First, as you might have noticed, it has been quite some time since our last release. At this time, to prevent having to continually shift our targets, we are not planning on targeting .NET 9 for this next release. (Additionally, if we were going to consider that, we would want that change done separate from locking changes.) We also have just gone through fixing a number of concurrency bugs for this release, and are hesitant to make any changes at this time. We will keep the issue open and reconsider this for a future release. Thanks again!