calyx
calyx copied to clipboard
calyxir
Intermediate Language (IL) for Hardware Accelerator Generators
It seems that portions of Calyx's standard library written in SystemVerilog is not able to be simulated by Icarus. ```sh [nix-shell:~/Repos/calyx] 11:12:38 $ iverilog -v Icarus Verilog version 12.0 (stable)...
The new `pip` restrictions force you to install everything in a virtualenv. This makes it very annoying to run fud/fud2 commands that rely on this without resorting to disabling this...
In all the excitement of #2191, a piece of the puzzle has gone missing from `sdn.py`: we are no longer calling the stats component as we used to. This omission...
See #2191. The goal is as follows: Make our oracles as clean, simple, and believable as possible. If some of our queues are meant to behave identically, they should use...
See #2191. We are already handling documentation for several queues that exist in the system. Once our implementation for PIEOs have landed, we should correspondingly add that new documentation in.
## Problem Statement This is partially redundant to a comment on #2113 but that's okay. Read that for more information on current implemented planners and more words about this problem...
## Issue Currently generated Ninja can be redundant. The two most egregious examples - [ ] Duplicate rules with different names are not collected into a single rule and reused....
(Some of this post is duplicated from my lab notebook [entry](https://github.com/orgs/calyxir/discussions/2202#discussioncomment-10014608)). # Recap of Problem Suppose we have the following control structure (assume all groups take 1 cycle): ``` static...
It should probably be verilator for the sake of matching the original `fud` and for overall stability
Probably extremely minor but the error message ``` Error: could not find path ``` caused me some unnecessary confusion because I wasn't sure if it meant that it couldn't find...