77 points aray07 4 hours ago 15 comments

measurablefunc 3 hours ago | parent

Combining this w/ a JIT gets you truffle/graal.

grashalm 3 hours ago | parent

If you are curious how this works in Truffle. Here are some docs:

https://www.graalvm.org/latest/graalvm-as-a-platform/languag...

arvyy 1 hour ago | parent

truffle is great, it's the most exciting tech in pldev domain to me. I'll just mention truffle has few other perks than just performance boost -- you get a debugger with a fairly small effort, and you get interop with other languages implemented using truffle (which does include fairly mainstream ones like python, js, and others).

owlstuffing 2 hours ago | parent

This type of optimization is also known as compile-time constant expression folding.

The problem with expanding it to include function calls is that it compromises binary compatibility. A code change on a function that results in a different range or has different side effects, will not be realized unless the caller is recompiled.

cogman10 1 hour ago | parent

Which is why the second most important optimization of compilers is inlining. If the callee is small enough then hoisting it up into the caller can expose additional optimizations, including noticing that parameters in the caller to the callee are in fact constant.

apignotti 2 hours ago | parent

We implemented a similar concept in our C++ -> JavaScript/WebAssembly compiler (Cheerp) several years ago. It was a nice challenge.

https://labs.leaningtech.com/blog/reducing-webassembly-size-...

infogulch 2 hours ago | parent

Doesn't Zig have a pretty nice implementation of compile-time evaluation built in?

jimmyhmiller 1 hour ago | parent

Comptime in zig is definitely quite cool, but not quite the same.

Consider this example. https://jimmyhmiller.com/split-shot/#/vm

Make the code a comptime argument in zig and you get "store to comptime variable depends on runtime condition".

hansvm 1 hour ago | parent

Zig isn't perfect, but I've definitely done this futamura stuff before. It's one of my favourite ways to implement a Forth in Zig (which, incidentally, is the default way that I introduce a backprop library or other DSL as I'm prototyping an idea). Interpreters are dead-simple to write, and I just write an interpreter with a comptime code argument. It definitely works, and the codegen is reasoanble out of the box.

artrax 1 hour ago | parent

We build this as a c++ library, do to query-compilation for our stream processing engine.

https://github.com/nebulastream/nautilus

It allows us to implement operators or functions without having to write them in terms of LLVM IR DSL (or similar). Rather, you write your functions over a wrapper type (and returning one), for which there is an overload of anything which you can overload in c++.

You can instantiate a wrapper from some c++ value, in which case the wrapper will just forward everything to this value, which was nice for debugging in the preagentic era. But, you can also instantiate such a wrapper as a tracer, and the result of the functions evaluation will be a trace.

For controlflow, we record whenever a wrapper type is evaluated as a bool, first return true, and when we reach the end of one branch, throw an exception to get back to the beginning, and start all over again, but evaluate it to false.

The resulting trace we can turn into LLVM IR and compile, or run for example in wasm

johntortugo 1 hour ago | parent

This seminar explains how GraalVM Truffle uses partial evaluation: https://youtu.be/pksRrON5XfU

airtnp 57 minutes ago | parent

I keep thinking the partial evaluator and Futamura projection application on LLM context.

Skill => Generic Program LLM skill optimizer => Partial evaluator Context => Env/Input Agent => Interpreter

First projection: mix(interpreter, program) => specialized skill AOT compiling => offline skill rewriting JIT compiling => runtime specializing, defer loading

Second projection: mix(mix, interpreter) => Agent/Model specific skill optimizer

Third projection: mix(mix, mix) => self-improvement skill optimizer

Whatever maybe I'm just thinking LLM as magic partial evaluator for any tasks

iFreilicht 27 minutes ago | parent

Very interesting concept! It reminds me of Hazel (https://hazel.org/) which takes the idea further and into the editor; any program written in it can be partially evaluated, even if variables or even part of the syntax are missing. Missing parts are called "holes", and those are preserved during evaluation so that something like

abs( + 2 * 3 - 10)

becomes

abs( - 4)

which is great fun. My first attempts using this showed me that if you're not used to this, it actually gets pretty confusing quite quickly, but I guess it can be learned.

fmbb 16 minutes ago | parent

Is that a unary minus denoting a negative number or a binary minus yielding a function with one hole?

TuringTest 13 minutes ago | parent

> For me, the most beautiful pieces of software are those that reveal something about the nature of programming languages.

There are three moments that make the functional programmer evolve their understanding of how a programming language really works as a symbolic representation of human thought, beyond the metaphor taught in school of a computer program as a grinder that crunches input data and produces results.

The first one is when you really understand recursion and how a function can be defined in terms of itself (with varying parameters for each call).

The second is when you discover Prolog variable unification or Haskell lazy evaluation and how you can represent infinite data structures without having the program hang in an infinite loop.

Third one is when you learn that the behavior of any imperative language (down to the working of the microprocessor data paths) can be fully represented in a pure functional rewrite language (without destructive assignment / cell state changes).

Those insights make one realize that all programming paradigms have the nature of symbols processing in common and there's no "true description" of a program's runtime execution - the imperative, functional and logic views of the running program are all equivalent representations.