Just Testing

Converting typed term representations: from HOAS to de Bruijn.

Fri, 03 Jul 2009 14:24:25 +1000

Converting typed term representations: from HOAS to de Bruijn.: Given a GADT representation of a typed higher-order term language using higher-order abstract syntax (HOAS), it is more difficult to convert to an alternative GADT representation using de Bruijn indices for bound variables than I at first expected. Here is a solution using explicit type representations (with Data.Typeable).

Instant Generics now has a website!

Sat, 30 May 2009 15:49:42 +1000

Instant Generics now has a website!: It also has a tar ball with some example code.

Instant Generics: Fast and Easy.

Mon, 11 May 2009 21:31:58 +1000

Instant Generics: Fast and Easy.: This paper introduces a novel approach to datatype-generic programming based on type classes and type families. It is fairly easy to use, is very expressive, and incurs minimal overhead over handwritten datatype-specific code.

Program for "Workshop on Exploiting Parallelism using GPUs and other Hardware-Assisted Methods (EPHAM 2009)".

Tue, 17 Mar 2009 21:49:01 +1100

Program for "Workshop on Exploiting Parallelism using GPUs and other Hardware-Assisted Methods (EPHAM 2009)".

Final version of "GPU Kernels as Data-Parallel Array Computations in Haskell".

Mon, 16 Mar 2009 21:32:52 +1100

Final version of "GPU Kernels as Data-Parallel Array Computations in Haskell".: This paper will be presented at the Workshop on Exploiting Parallelism using GPUs and other Hardware-Assisted Methods (EPHAM 2009), co-located with CGO’09 in Seattle, WA.

These graphs summarise the performance of Data Parallel Haskell...

Tue, 10 Mar 2009 15:39:44 +1100

These graphs summarise the performance of Data Parallel Haskell for three simple benchmarks on two different architectures, both of which have 8 cores. At the top, we have 8 cores with one hardware thread each, in the form of two Quad-Core 3GHz Xeon processors. At the bottom, we have 8 cores with 8 hardware threads each (for a total of 64 hardware threads), in the form of a single 1.4GHz UltraSPARC T2.

The benchmarks are the following: (1) sumsq computes the parallel sum of the squares from 1 to 10 million; (2) dotp computes the dot product of two dense vectors of 100 million double-precision floating-point numbers; and (3) smvm multiplies a sparse matrix with 10 million non-zero double-precision floating-point elements with a dense vector.

The graphs show the speedup of Data Parallel Haskell with respect to a sequential C implementation of each benchmark – whenever, a curve climbs above 1 on the y-axis, the parallel Haskell program beats the sequential C program in absolute runtime.

The scalability of these three programs in Data Parallel Haskell is very good, although dotp is limited by memory bandwidth on the Quad-Core Xeon processors, as discussed in my previous post. The grey curve is a parallel C implementation of dotp that is bandwidth limited in the same manner.

Installing GtK2Hs on a Mac with the native GTK+ OS X Framework.

Wed, 04 Mar 2009 14:06:49 +1100

Installing GtK2Hs on a Mac with the native GTK+ OS X Framework.

This is the performance of a dot product of two vectors of 10...

Tue, 03 Mar 2009 14:17:00 +1100

This is the performance of a dot product of two vectors of 10 million doubles each using Data Parallel Haskell. Both machines have 8 cores. Each core of the T2 has 8 hardware thread contexts. The performance is memory bound. The DPH code is

dotp' :: [:Double:] -> [:Double:] -> Double
dotp' v w = D.sumP (zipWithP (*) v w)

Interesting is that despite the much lower single-thread performance, the T2 makes it all up in excellent multi-threading performance. Interesting to Haskell folks is that the corresponding multi-threaded C code using pthreads is much harder to write and barely any faster when using all parallelism (the Sun C-Compiler still manages to reduce the runtime by about 30% with 64 threads, but on the Xeons, there is no significant difference).

NB: This uses the latest development version of GHC and DPH. Thanks to Ben Lippmeier for his nice work on the SPARC backend of GHC.

Call for Presentations: Commercial Users of Functional Programming Workshop (CUFP) 2009!

Tue, 03 Mar 2009 11:00:10 +1100

Call for Presentations: Commercial Users of Functional Programming Workshop (CUFP) 2009!: Show the world how you have used functional programming!

Complete and Decidable Type Inference for GADTs.

Tue, 03 Mar 2009 10:43:48 +1100

Complete and Decidable Type Inference for GADTs.

Industrial Haskell Group.

Tue, 03 Mar 2009 10:04:34 +1100

Industrial Haskell Group.

Version 0.16.0 of the C->Haskell interface generator.

Sun, 01 Mar 2009 12:19:03 +1100

Version 0.16.0 of the C->Haskell interface generator.

Mark your Haskell tweets with #haskell

Sat, 21 Feb 2009 23:03:38 +1100

This makes for easy searching and let’s us use Twitterfall (which is partly implemented in Erlang, btw).

Comparison of type families with functional dependencies.

Mon, 16 Feb 2009 14:10:43 +1100

Comparison of type families with functional dependencies.: This is a haskell-cafe post of a brief summary of the commonalities and differences of the two language features.

The slides of the invited talks about multicore programming from DAMP'09 are online.

Thu, 29 Jan 2009 13:56:54 +1100

The slides of the invited talks about multicore programming from DAMP'09 are online.: John Reppy talked about The Manticore Project and Ulf Wiger talked about Erlang Programming for Multi-core.

Some nice code examples showing how to use type families.

Fri, 16 Jan 2009 23:50:34 +1100

Some nice code examples showing how to use type families.: They are inspired by code from the MTL and from parser combinators.

Enforcing a relation between independent type families.

Fri, 16 Jan 2009 23:31:08 +1100

Enforcing a relation between independent type families.: The question of how to ensure that one type family is the inverse of another comes up once in a while. Here some concrete code.

Workshop on Exploiting Parallelism using GPUs and other Hardware-Assisted Methods (EPHAM 2009).

Mon, 12 Jan 2009 15:04:34 +1100

Workshop on Exploiting Parallelism using GPUs and other Hardware-Assisted Methods (EPHAM 2009).: Check out the CFP! EPHAM09 is co-located with 2009 International Symposium on Code Generation and Optimization (CGO) in Seattle, Washington.

GPU Kernels as Data-Parallel Array Computations in Haskell.

Tue, 06 Jan 2009 22:58:46 +1100

GPU Kernels as Data-Parallel Array Computations in Haskell.: This paper provides a brief overview of our approach to GPU programming with Haskell by embedding GPU kernels with a domain-specific language of data-parallel array computations in Haskell code. The paper also presents some preliminary benchmarks.

ICFP calls for functional programming Experience Reports.

Tue, 16 Dec 2008 12:22:00 +1100

ICFP calls for functional programming Experience Reports.: If you have used functional programming in a real world application, we want to hear about it!