Motivated by Tiger, a tool for generating interpreters, being mentioned on Twitter, I had a brief look at vmgen, Tiger, eJSTK, Truffle DSL, and DynSem. What follows are my rather rough notes and pointers. So, this is by no means a careful literature study, and I welcome further pointers.
It has been a while since we put together a release for SOMns. And it has been even longer, since I last wrote about it on this blog.
Writing specializations is generally pretty straight forward, but there is at least one common pitfall. When designing specializations, we need to remind ourselves that type-based specializations are technically guards.
When we have to debug applications that use concurrency, perhaps written in Java, all we get from the debugger is a list of threads, perhaps some information about held locks, and the ability to step through each thread separately.
Note: This post is meant for people familiar with Truffle. For introductory material, please see for instance this list.
This post, the fourth in the series, is about my current work on concurrency and tooling. As mentioned before, I believe that there is not a single concurrency model that is suitable for all problems we might want to solve. Actually, I think, this can be stated even stronger: Not a single concurrency model is appropriate for a majority of the problems we want to solve.
The third post of this series is about how I started using Truffle and Graal, pretty much 4 years ago. It might be in parts ranty, but I started using it when it was in a very early stage. So, things are a lot better today.
Since SOMns is a pure research project, we aren’t usually doing releases for SOMns yet. However, we added many different concurrency abstractions since December and have plans for bigger changes. So, it seems like a good time to wrap up another step, and get it into a somewhat stable shape.
One possible way for modeling concurrent systems is Tony Hoare’s classic approach of having isolated processes communicate via channels, which is called Communicating Sequential Processes (CSP). Today, we see the approach used for instance in Go and Clojure.
Research on programming languages is often more fun when we can use our own languages. However, for research on performance optimizations that can be a trap. In the end, we need to argue that what we did is comparable to state-of-the-art language implementations. Ideally, we are able to show that our own little language is not just a research toy, but that it is, at least performance-wise, competitive with for instance Java or JavaScript VMs.
Next weekend starts one of the major conferences of the programming languages research community. The conference hosts many events including our Meta’16 workshop on Metaprogramming, SPLASH-I with research and industry talks, the Dynamic Languages Symposium, and the OOPSLA research track.
With the Truffle language implementation framework, we got a powerful foundation for implementing languages as simple interpreters. In combination with the Graal compiler, Truffle interpreters execute their programs as very efficient native code.
Now that we got just-in-time compilation essentially “for free”, can we get IDE integration for our Truffle languages as well?
Beside the great performance after just-in-time compilation, the Truffle Language implementation framework provides a few other highly interesting features to language implementers. One of them is the instrumentation framework, which includes a REPL, profiler, and debugger.
Last December, we got a research project proposal accepted for a collaboration between the Software Languages Lab in Brussels and the Institute for System Software here in Linz. Together, we will be working on tooling for complex concurrent systems. And with that I mean systems that use multiple concurrency models in combination to solve different problems, each with the appropriate abstraction. I have been working on these issues already for a while. Some pointers are available here in an earlier post: Why Is Concurrent Programming Hard? And What Can We Do about It?
Continuing a little bit with writing notes on Truffle and Graal, this one is based on my observations in SOMns and changes to its message dispatch mechanism. Specifically, I refactored the main message dispatch chain in SOMns. As in Self and Newspeak, all interactions with objects are message sends. Thus, field access and method invocation is essentially the same. This means that message sending is a key to good performance.
Over the course of the next four weeks, I plan to publish a new post every Tuesday to give a detailed introduction on how to use the Graal compiler and the Truffle framework to build fast languages. And this is the very first post to setup this series. The next posts are going to provide a bit of background on Golo, the language we are experimenting with, then build up the basic interpreter for executing a simple Fibonacci and later a Mandelbrot computation. To round off the series, we will also discuss how to use one of the tools that come with Graal to optimize the performance of an interpreter. But for today, let’s start with the basics.
The past few month, I have been busy implementing a fast actor language for the JVM. The language is essentially Newspeak with a smaller class library and without proving access to the underlying platform, which can lead to violations of the language’s guarantees.
Back in 2013 when looking for a way to show that my ideas on how to support concurrency in VMs are practical, I started to look into meta-compilation techniques. Truffle and RPython are the two most promising systems to build fast language implementations without having to implement a compiler on my own. While these two approaches have many similarities, from a conceptual perspective, they take two different approaches that can be seen as the opposite ends of a spectrum. So, I thought, it might be worthwhile to investigate them a little closer.
Runtime metaprogramming and reflection are slow. That’s a common wisdom. Unfortunately. Using refection for instance with Java’s reflection API, its dynamic proxies, Ruby’s #send
or #method_missing
, PHP’s magic methods such as __call
, Python’s __getattr__
, C#’s DynamicObject
s, or really any metaprogramming abstraction in modern languages unfortunately comes at a price. The fewest language implementations optimize these operations. For instance, on Java’s HotSpot VM, reflective method invocation and dynamic proxies have an overhead of 6-7x compared to direct operations.
Today, I gave a talk on implementing languages based on the ideas behind RPython and Truffle at FOSDEM on the main track. Please find abstract and slides below.
Today, I got a few more benchmarks running to get a better idea of where RTruffleSOM and TruffleSOM stand in terms of their absolute performance.
The first results of my experiments with self-optimizing interpreters was finally published in IEEE Software. It is a brief and very high-level comparison of the Truffle approach with a classic bytecode-based interpreter on top of RPython. If you aren’t familiar with either of these approaches, the article is hopefully a good starting point. The experiments described in it use SOM, a simple Smalltalk.
Today at FOSDEM, I gave a brief talk on implementing SOM, a little Smalltalk, with RPython and Truffle. RPython, probably best known for the PyPy implementation, uses meta-tracing JIT compilation to make simple interpreters fast. Truffle, a research project of Oracle Lab, is an approach for building self-optimizing interpreters and in combination with Graal, it gives a JIT compiler for AST-like interpreters. In the talk, I briefly sketch both of them, without going into many details.