Rachit Nigam

icon indicating a website link rachit.pl icon indicating an email [email protected]

icon company Massachusetts Institute of Technology icon location Incoming EECS professor at MIT. Creator of @calyxir. Up to no good.

Results 525 comments of Rachit Nigam

[Tracker] Lowering high-level Calyx programs

Awesome! Looking forward to the PR!

Calyx Sharing Results-- To-do Tasks

@calebmkim can you update this tracker to reflect the current state of things. If we've moved the tracking to the evaluation repository, let's close this issue with a link to...

Calyx Sharing Results-- To-do Tasks

Agreed with @sampsyo

Source position integration through CIRCT SCF

Looks like this has been fixed. @EclecticGriffin feel free to re-open if that is not the case

[Tracking] Frontend data structure audit

I assume we're not going to work on this so closing the issue

[CIRCT] Lowering high level MLIR programs to Calyx.

@cgyurgyik are there action items for the this tracker that we should be working on? If not, let's wrap up the issue

`remove-comb-groups` should be optional

While working on fixing this, I came across this program which cannot be compiled correctly: ``` seq { if lt.out with C1 { A; } if lt.out with C2 {...

`remove-comb-groups` should be optional

Okay, so I've realized that loops where we chain the condition check with the next group will also potentially suffer from the above problem. Consider: ``` while lt.out with C1...

Typechecker Issue

Oh odd! Maybe it is not assigning a dynamic index type to the accesses memories?

Typechecker Issue

Hm, the problem is that `perm[i]` needs to imply that all banks in a particular dimension need to be consumed. Currently, it is only saying that one bank is being...