Anshuman Mohan

Thanks Cassandra! So yes, to sum: the problem I chalked out was reasonable, but my proposed code was actually still a little off. The code above takes a slightly different...

I gave it some more thought. I think an array needs to be banked per the _[LCM](https://en.wikipedia.org/wiki/Least_common_multiple) of the various parallelism factors that ever parallelize the array_. For example, ```...

After the meeting we just had, it appears I have arrived at the need for _arbitration logic_. Rachit thinks that this is going to be hard, harder possibly than going...

Done with 1 and 2 above. PR https://github.com/cucapra/calyx/pull/1473 is once again ready for review.

I'm taking this off the tutorial milestone; the new in-milestone target is #1474. I will incorporate some of the discussion above into my educational slide/handout so that they can see...

@rachitnigam should we just close this as a wontfix? Or maybe convert it into a discussion or somesuch?

Done deal!

We are limiting this to the alternating style because any non-alternating case can just be crafted using the alternating style (above) plus calls to appropriate control operators. Say you need:...

Perhaps silly, but do we need to do anything to prevent: ``` @qt seq { d1; empty_static_group_that_takes_time_0; d2 } ``` ? Are static groups with latency 0 even allowed?

Whoops mis-click...