Anshuman Mohan

I have now pushed two further components: - `map`: Covers 1d memory, items passed by reference, while loops, width inference, guarded assignments, and the higher-order helper `incr`. - `main`: Covers...

I am close to done when it comes to new code, so I welcome your feedback on that! Please see #2002 for the code that I've been writing. In some...

Ah actually there is a middle ground that saves on spurious primitive cells! They can use the slightly lower-level helper `binary_use`, which takes a cell and two inputs, and drives...

Yes, either `binary_use` or the more specific `_use` can be tossed in there right now! Putting `binary_use` in there would produce the same Calyx code as the clients do currently....

Yup, this is now that issue! I’ll just change the name…

In #2034 and #2036, I have made changes of the sort proposed above in the NTT and Systolic frontends respectively. I'd be happy to hear of other big chunks of...

Okay I've taken a look at the other frontends and I'm pretty sure there's not a ton to be done in those cases. MrXL could be reworked a ton, but...

Ah, sorry! I'll edit the text above to explain, but yes, this is a different scheduling policy that I think _is_ easily within our grasp using just our simple approach....

Just documenting a wrinkle that has come up now that @csziklai has started working on this! Let's just focus on round-robin scheduling with 3 flows. #### Review: `n=2` As you...

Here's a sketch proposal. Warning that it's a quick and dirty sketch mostly focussed on `n=3`, but I'd like to surface it so we can pick holes in it later...