Deepening LLDB Integration

A previous post covered the subject of the initial LLDB and Dylan integration.

While much of this post discusses what we did to improve Dylan integration, the overall techniques are broadly useful for many languages. In fact, if anyone is interested, I am available for consulting on this topic. Just email me.

Catching Up

In the previous post, we talked about changing the base representation from void* to uintptr_t to work around a bug in LLDB. This was, in part, due to our goal to try to have the integration work with the currently shipping version of LLDB.

Unfortunately, this work won't be completed and merged. In separate developments, it has emerged that it can be unsafe to treat a pointer as an integer when working with a garbage collector. The details of this are quite interesting, so out of an abundance of caution, we will not be going down this route.

Expanding Objects

We are able to use the "pointer depth" parameter to commands like expression and frame variable to show the "expanded" form of objects:

(lldb) frame variable -P 1 T15
(dylan_value) T15 = 0x017e8000 {<buffer>} {
  [buffer-next] = 0x00000001 {<integer>: 0}
  [buffer-end] = 0x00000001 {<integer>: 0}
  [buffer-position] = 0x00000001 {<integer>: 0 ...

There are comments.

Dylan and Gradual Typing

As we look to the future and what we would like for Dylan to become and investigate how we would like for Dylan to evolve, it is helpful to look at some of the current work and how Dylan compares, where Dylan falls down and whether or not we can improve it.

One of those areas is in the guarantees offered by the type system. While Dylan is seen as a dynamic language, it has a number of features that help provide optional static type checking. As we'll see, there is a lot of room for improvement in this area.

In this post, using a missing compile-time warning as the driver, we'll walk through some details of the Dylan type system and then see how it differs from a gradually typed system. We'll see that type annotations are interpreted very differently under a gradual typing regime versus the Open Dylan compiler.

I've previously written a Type System Overview which may be useful, but hopefully this post can stand on its own.

One particularly interesting body of work is that on Gradual Typing. From What is Gradual Typing:

Gradual typing is a type system I [Jeremy Siek ...

There are comments.

The ALGOL Roots of Dylan

Dylan is often seen as a descendent of the Lisp family of languages. It was designed (and implemented) by people from the Common Lisp world and borrowed lots from Scheme, EuLisp and other Lisp dialects. On the other hand, it was given a syntax from the ALGOL tradition rather than using s-expressions.

It is interesting to compare some of what is present within the Dylan language design with ALGOL though and see if perhaps some of the ALGOL influence wasn't just in the syntax. It is difficult today to know the extent to which ALGOL 68 was a direct influence versus having been filtered through Common Lisp and other languages. After all, Dylan was designed some 25 years or so after ALGOL 68. (At the time of this writing, Dylan itself is over 20 years old.)

It is clear though that many languages have picked up ideas from ALGOL over the years, including languages that were direct influencers of Dylan such as Scheme.

This post isn't intended to say "this concept came from ALGOL" but just to look at some of the interesting similarities.

Orthogonal Design

ALGOL 68 as defined in the Revised Report on the Algorithmic Language ...

There are comments.

Integrating with LLVM

Many compilers are now being outfitted with LLVM backends. There are a variety of ways to do this, and we'll take a look at some of them here.

Calling LLVM APIs Directly

One common and easy way to integrate with LLVM is to directly link against the LLVM libraries and invoke the LLVM API directly.

C++ API

LLVM's native APIs are C++. Examples of languages that talk to LLVM via the C++ APIs are Julia and CLASP.

This is also the approach taken by Clang, however, Clang lives within the same code repository as LLVM and shares many developers. It is part of the LLVM project rather than a separate compiler that is using LLVM as a backend.

While this seems like an attractive option, there are some issues with it:

Your code must either be written in C++ or be able to invoke C++ code via an FFI (foreign function interface).
You are tied to a particular version of the LLVM API at compile time.
It is harder to re-use someone's existing installation of LLVM as you are tied to a particular version.

C API

LLVM also provides a C wrapper around the C++ APIs. This ...

There are comments.

Saying Good-bye: HARP

This begins a new series of blog posts that will continue over the next few months as we say "Good-bye!" to parts of Open Dylan.

Freedom comes when you learn to let go
Creation comes when you learn to say no

  -- Madonna, The Power of Good-Bye

We're beginning a process by which we'll start slimming down the compiler and the libraries, letting go of some major chunks of code, with the goal of improving the hackability of the compiler and enabling us to make new leaps in functionality.

What is HARP?

HARP is the Harlequin Abstract RISC Processor and was designed and developed at Harlequin in the late 1980s. It was used in Harlequin's LispWorks and later translated to Dylan for use in Harlequin's DylanWorks (which is now Open Dylan).

Clive Tong, an engineer at Harlequin in 1989, briefly described it as:

The compiler targeted an instruction set known as HARP (Harlequin Abstract RISC Processor), and then HARP instructions were translated into machine instructions using a template matching scheme. HARP had an infinite set of registers, and the register colouring happened as part of this templating processing.

Some additional details about the early design of HARP ...

There are comments.

Type System Overview

The type system and how it is used is a commonly misunderstood aspect of the Dylan language. Although it lacks some forms of expressiveness in the current incarnation, it also has some features that aren't found in many languages, such as singleton types. It is also very important in helping the compiler to generate faster yet still safe code.

One interesting feature in Dylan is that it is optionally typed. While this is more common today and sometimes has fancy names applied like 'gradually typed', the overall point is the same: Your code can start out untyped and looking like code does in Ruby or Python. However, when you want or need additional performance or correctness guarantees, you can supply type annotations that the compiler can use. The compiler can also infer some types from the values used or other type annotations.

In this post, we'll explain some of the basic concepts of the Dylan type system and show how it is used by the compiler.

Type and Value Relationships

There are 2 important relationships between values and types in Dylan.

They are instance? and subtype?. Other relationships, such as known-disjoint? are used within the compiler to assist ...

There are comments.

Function Types and Dylan 2016

Moving towards Dylan 2016, the Dylan community would like to address some weaknesses in the language specification and what can be readily expressed in Dylan code. In this post, we'll look at function types as well as provide a brief introduction to some details of the type system implementation within the Open Dylan compiler.

Function Types

One of the big holes in the Dylan type system is the inability to specify function types. What this means is that you can only say that a value is of type <function> and can't indicate anything about the desired signature, types of arguments, return values, etc. This is unfortunate for a number of reasons:

Poor static type safety. The compiler can verify very little involving a function value. It can't warn when the wrong number of arguments or the wrong types of arguments are passed.
Less clear interfaces. The type signature of a function must be documented clearly rather than being expressed clearly within the code.
Optimization is more difficult. Since the compiler can't perform as many checks at compile time, more checks need to be performed at run-time, which limits the amount of optimization that can be performed ...

There are comments.

HTTP and Stream Processing

As we discussed in the previous post, we are thinking about a new design and implementation for the streams library in Open Dylan.

While the examples in this post are in Dylan and are using code from our HTTP server, these issues exist in HTTP frameworks in other languages. The code should be clear enough that little to no Dylan knowledge is required to understand the points being made here.

What does this have to do with HTTP? There are several pain points in our HTTP stack as it is currently written:

Requests are read in their entirety into memory, so a large request (such as a file upload) takes a significant amount of memory.
Responses often buffer their entire output in memory as well.
Because of the use of the existing streams library, we don't handle non-blocking sockets and require a thread per socket.
We don't have a good model for handling long-lasting connections such as might be used with Server Sent Events or WebSockets without tying up a thread for the duration of the socket being open.

We don't know yet what the new streams API will look like, but we can look at our ...

There are comments.

Beginning to Rethink Streams

Dylan's current streams library has served us moderately well over the years. However, it has some issues which can be addressed by a new design, expanding the range of problems for which it is suited.

How things are now

According to the current streams library's documentation, the design goals were:

A generic, easy-to-use interface for streaming over sequences and files. The same high-level interface for consuming or producing is available irrespective of the type of stream, or the types of the elements being streamed over.
Efficiency, especially for the common case of file I/O.
Access to an underlying buffer management protocol.

One of the things it was explicitly not designed to handle was, again, according to the design goals in the documentation:

A comprehensive range of I/O facilities for using memory-mapped files, network connections, and so on.

Unfortunately, the primary interface to our current network library is based on these very streams for which network connections were not a design goal. While this works in practice, it imposes some important limitations on our networking code. The biggest of these is that sockets can not be non-blocking as it is expected that reads and writes will complete ...

There are comments.

Integrating with LLDB

I spend a lot of time debugging Dylan code. Up until now, this has been a somewhat painful process when not using the IDE on Windows. (And I don't really use the IDE on Windows as it doesn't fit well into my workflow.) I finally reached the point where I decided that I wanted to improve our debugger integration.

Much of what is described below may be applicable to people working on debugging support for other languages.

Current State of Dylan Debugging

We have a debugger on Windows integrated with our IDE. This facility is not available on the other platforms that we support. There are many reasons for this:

The debug info that Open Dylan generates is only done on Windows.
The debugger code for interacting with the OS is only implemented for Windows.
The IDE itself is only available on Windows.

This means that debugging on Linux, FreeBSD and Mac OS X has traditionally been more challenging. We often resort to "printf debugging" and have some basic debug printing functions that can be invoked from the compiler so long as they don't crash. Debugging is really only effective with the C back-end as the HARP ...

There are comments.