Brad Chamberlain

> I think it's reasonable to consider the work tracked in this issue as completed when the initial PR was merged, even though there is still work remaining with the...

Belated thought in case it wasn't clear: I think all that would be required to close this "for real" would be to merge a PR capturing the current state of...

Catching up on this issue and the related Discourse thread, I was finding that I shared @damianmoz's confusion and didn't understand the explanations others were giving despite being familiar with...

> So what will be an acceptable resolution to this issue that will allow us to close it? A different text of the warning? If so, what? A switch to...

This (along with `param` domains and arrays) is a longstanding, but unfulfilled desire. The primary challenge is that most of these types are defined today as records/classes in module code,...

> It seems like this method isn't being picked up. @Iainmon : I think that this is because the `x=2` is being interpreted as an initialization (e.g., `init=` routine) rather...

This is related to the discussion in https://github.com/chapel-lang/chapel/issues/20873. Reading this issue, I was convinced by the argument that we could transform the code into just the evaluation of the LHS...

Just stumbled across this, which also may be related: https://github.com/chapel-lang/chapel/issues/20496

That's encouraging, thanks for the report, Paul! @ronawho, can you let us know what you remember here? Like did you do a full CHPL_COMM=gasnet paratest and had a few failures...

Just to let people my current thinking here: If we find that there are inherent problems with using multi-rail with a single locale per node, and we don't believe those...