Chisel: Constructing Hardware in a Scala Embedded Language
chisel.eecs.berkeley.edu
chisel.eecs.berkeley.edu
The industry lack's of wide profiles (hardware engineer with good software understanding) is really hurting any new innovative HDL
For what it's worth, I don't thin Chisel is really standardizable as it stands, but something like it could be.
Update: read the paper and they cover a few examples in it. Don't have enough domain knowledge to judge whether the claims are accurate, nor whether they generalise to something like clash-lang.
I wasn't sure how seriously to take Chisel until I saw the RISC-V BOOM definition https://github.com/ucb-bar/riscv-boom. It's clean, succinct, and I think Chisel is quite viable.
It's also a testament to the symbolic nature of Scala. It holds together pretty well in JVM, ScalaJS, Chisel, and scala-native. Scala as a language is remarkably simple at the core, which Odersky does a good job of showing in his book, even though there are a lot of library features on top that make it complex and a more difficult language in general use. In this way of having a cute core, it is very unlike C++.
I tried to make a verilog DSL in julia, which was a lot of fun, but I got stuck on how to implement sequential logic (instead of just combinatorial logic). Even better (and should be reasonably doable with clever application of julia's macro and type system) would be to take a julia function and transparently generate verilog from it.
Sure, there are many, many more temporary variables created between nodes, but what I care about when debugging is "what are the values of my Chisel nodes"? It's like debugging by looking at assembly, but with your C/C++ variable names carried through to the disassembly.