- Haskell-Cafe - Haskell.org

Re: [Haskell-cafe] ANN: unordered-containers - a new, faster hashing-based containers library
by Louis Wasserman 14 Mar '11

14 Mar '11

A couple thoughts: size takes O(n). That's just depressing. Really. Do you think union, intersection, etc. could be supported? Louis Wasserman wasserman.louis(a)gmail.com http://profiles.google.com/wasserman.louis

21 48

Some quick experiments with GHC 7.0.2 in Intel's Manycore Testing Lab (32 cores)
by José Pedro Magalhães 14 Mar '11

14 Mar '11

Hi all, I've played a bit with Intel's Manycore Testing Lab ( http://software.intel.com/en-us/articles/intel-many-core-testing-lab/) Part of the agreement to use it requires that you report back your experiences, which I did in an Intel forum post ( http://software.intel.com/en-us/forums/showthread.php?t=81396) I thought this could be interesting to the Haskell community in general as well, so I'm reposting here, and pasting the text below for convenience. I've replaced the images with links. Cheers, Pedro As per the agreement with Intel, I am reporting my experiences with the > Intel Manycore Testing Lab (Linux). This was my first time in the lab, and I > wanted to test GHC's [1] SMP parallelism [2] features. > > The first challenge was to actually get GHC to work on the lab. There was a > working version of ghc under /opt/ghc6.13/bin/ghc, but I really needed GHC > 7. So first I built GHC 7.0.2-rc2, which worked without much trouble. > > Next step was to get all the necessary libraries in place. Since the lab > has no direct internet access, cabal-install [3] wouldn't be of much use. > Instead, I downloaded a snapshot of hackage [4] with the latest version of > every package and manually installed the packages I needed. A bit boring, > but doable. > > Finally I was ready to compile my programs and test. First thing I tried > was an existing algorithm I had which, at some point, takes a list of about > 500 trees and, for each tree, computes a measure which is expressed as a > floating point number. This is basically a map over a list transforming each > tree into a float. Each operation is independent of the others, and all > require the same input, so it seems ideal for parallelisation. A quick > benchmark revealed the following running times: > > http://dreixel.net/images/perm/ParList.png > > (Note the non-linear number of cores at the end of the x-axis.) Apparently > there are performance gains with up to 6 cores; adding more cores after this > makes the total running time worse. > > While this might sound bad, do note that all that was necessary to > parallelise this algorithm was a one line change: basically, at the point > where the list of floats @l@ is generated, it is replaced with @l `using` > parList rdeepseq@. This change, together with recompilation using > -threaded, is all that is necessary to parallelise this program. > > Later I performed a more accurate benchmark, this time using the equality > function (take two elements and compare them for equality). The first step > was to parallelise the equality function, which, again, is a very simple > task: > > -- Tree datatype > data Tree a = Leaf | Bin a (Tree a) (Tree a) > > -- Parallel equality > eqTreePar :: Tree Int -> Tree Int -> Bool > eqTreePar Leaf Leaf = True > eqTreePar (Bin x1 l1 r1) (Bin x2 l2 r2) = x1 == x2 && par l (pseq r (l && > r)) > where l = eqTreePar l1 l2 > r = eqTreePar r1 r2 > eqTreePar _ _ = False > > `par` and `pseq` are the two primitives for parallelisation in GHC [5]. The > performance graph follows: > > http://dreixel.net/images/perm/ParEq.png > > (This time I ran the benchmark several times; the error bars on the graph > are the standard deviations.) Again we get performance improvements with up > to 6 cores, and after that performance decreases. What I find really nice is > the improvement with two cores, which is almost a 50% decrease in running > time. The ratios for 2 to 4 cores wrt. the running time with 1 core are > 0.52, 0.39, and 0.35, respectively. This is really good for such a simple > change in the source code, and most people only have up to 4 cores anyway. > In any case, the results of this (very preliminary) experiment seem to > indicate that GHC's SMP parallelism is not particularly optimized for a high > number of cores (yet). > > I'm planning to explore this line of research further, and I'm hoping to be > able to conduct more experiments in the near future. Feel free to contact me > if you want more information on what I've done. > > > Cheers, > Pedro > > [1] http://www.haskell.org/ghc/ > [2] http://www.haskell.org/ghc/docs/latest/html/users_guide/using-smp.html > [3] http://hackage.haskell.org/package/cabal-install > [4] http://hackage.haskell.org > [5] > http://hackage.haskell.org/packages/archive/parallel/latest/doc/html/Contro… >

2 1

Re: [Haskell-cafe] Data.ByteString.Lazy.ByteString vs Data.ByteString.Lazy.Internal.ByteString
by Antoine Latter 14 Mar '11

14 Mar '11

The ghc-binary package is used internal to GHC, and isn't gauranteed to be present from one version to the next, nor do I expect the GHC team to promise it will have a stable interface. You'd really be better of instaling the package "binary", or something similar. On Mar 14, 2011 5:34 AM, "Christian Maeder" <Christian.Maeder(a)dfki.de> wrote:

1 0

ICFP 2011 Deadline Extension
by Wouter Swierstra 14 Mar '11

14 Mar '11

===================================================================== ICFP 2011: International Conference on Functional Programming http://www.icfpconference.org/icfp2011 ===================================================================== On behalf of the Program Committee of ICFP 2011, I would like to announce a deadline extension for authors affected by the recent earthquake in Japan. The Program Committee will accept titles and abstracts until Monday 4 April at 23:59, and full submissions until Thursday 7 April at 23:59 from those authors affected by the earthquake. Our thoughts go out to the victims of this tragedy. Wouter Swierstra ICFP Publicity Chair

1 0

Re: [Haskell-cafe] [Haskell] Linker flags for foreign export.
by Max Bolingbroke 14 Mar '11

14 Mar '11

Hi Jason, On 8 March 2011 05:28, Jason Dusek <jason.dusek(a)gmail.com> wrote: > gcc -g -Wall -O2 -fPIC -Wall -o import \ > -I/usr/lib/ghc-6.12.1/include/ \ > import.c exports.so In my experience, the easiest way to do this is to use gcc to build object files from C source files, and then specify those object files on the ghc command line in order to get GHC to do the linking step. This will deal with linking in the correct RTS and, if you specify appropriate -package flags, dependent packages as well. If you want a C main function then see user guide section 8.2.1.1 at http://www.haskell.org/ghc/docs/latest/html/users_guide/ffi-ghc.html. Cheers, Max

2 7

Yesod 0.7.1 using GHC 7.0.2 on Arch Linux
by Ertugrul Soeylemez 14 Mar '11

14 Mar '11

Hello there, today I have upgraded to GHC 7.0.2 by doing a system update on my Arch Linux box (x86, 32 bits), and I'm having difficulties to install Yesod. Its dependencies seem to have been installed without problems, but the Yesod package itself fails. The complete build log follows: % cabal install yesod Resolving dependencies... Configuring yesod-0.7.1... Preprocessing library yesod-0.7.1... Preprocessing executables for yesod-0.7.1... Building yesod-0.7.1... [1 of 1] Compiling Yesod ( Yesod.hs, dist/build/Yesod.o ) Registering yesod-0.7.1... [1 of 2] Compiling CodeGen ( CodeGen.hs, dist/build/yesod/yesod-tmp/CodeGen.o ) [2 of 2] Compiling Main ( scaffold.hs, dist/build/yesod/yesod-tmp/Main.o ) Loading package ghc-prim ... linking ... done. Loading package integer-gmp ... linking ... done. Loading package base ... linking ... done. Loading package bytestring-0.9.1.10 ... linking ... done. Loading package base64-bytestring-0.1.0.2 ... linking ... done. Loading package array-0.3.0.2 ... linking ... done. Loading package containers-0.4.0.0 ... linking ... done. Loading package cereal-0.3.0.0 ... linking ... done. Loading package filepath-1.2.0.0 ... linking ... done. Loading package old-locale-1.0.0.2 ... linking ... done. Loading package old-time-1.0.0.6 ... linking ... done. Loading package unix-2.4.2.0 ... linking ... done. Loading package directory-1.1.0.0 ... linking ... done. Loading package binary-0.5.0.2 ... linking ... done. Loading package data-default-0.2.0.1 ... linking ... done. Loading package tagged-0.2 ... linking ... done. Loading package crypto-api-0.5.2 ... linking ... done. Loading package pureMD5-2.1.0.3 ... linking ... done. Loading package pretty-1.0.1.2 ... linking ... done. Loading package template-haskell ... linking ... done. Loading package transformers-0.2.2.0 ... linking ... done. Loading package deepseq-1.1.0.2 ... linking ... done. Loading package text-0.11.0.5 ... linking ... done. Loading package blaze-builder-0.2.1.4 ... linking ... done. Loading package blaze-html-0.4.1.0 ... linking ... done. Loading package file-embed-0.0.3.1 ... linking ... done. Loading package time-1.2.0.3 ... linking ... done. Loading package unix-compat-0.2.1.1 ... linking ... done. Loading package enumerator-0.4.7 ... linking ... done. Loading package mtl-2.0.1.0 ... linking ... done. Loading package parsec-3.1.1 ... linking ... done. Loading package network-2.2.1.10 ... linking ... done. Loading package wai-0.3.2 ... linking ... done. Loading package utf8-string-0.3.6 ... linking ... done. Loading package web-routes-0.23.4 ... linking ... done. Loading package wai-app-static-0.0.1.1 ... linking ... done. Loading package random-1.0.0.3 ... linking ... done. Loading package clientsession-0.4.1 ... linking ... done. Loading package cookie-0.0.0 ... linking ... done. Loading package failure-0.1.0.1 ... linking ... done. Loading package extensible-exceptions-0.1.1.2 ... linking ... done. Loading package QuickCheck-2.4.0.1 ... linking ... done. Loading package blaze-builder-enumerator-0.2.0.1 ... linking ... done. Loading package json-types-0.1 ... linking ... done. Loading package json-enumerator-0.0.1 ... linking ... done. Loading package hamlet-0.7.2 ... linking ... done. Loading package monad-peel-0.1 ... linking ... done. Loading package zlib-0.5.3.1 ... linking ... done. Loading package zlib-bindings-0.0.0 ... linking ... done. Loading package wai-extra-0.3.3 ... linking ... done. Loading package web-routes-quasi-0.6.3.1 ... linking ... done. Loading package yesod-core-0.7.0.1 ... linking ... done. Loading package yesod-static-0.0.0.1 ... linking ... done. Loading package SHA-1.4.1.3 ... linking ... done. Loading package RSA-1.0.6.1 ... linking ... done. Loading package data-object-0.3.1.6 ... linking ... done. Loading package attoparsec-0.8.5.0 ... linking ... done. Loading package bytestring-nums-0.3.2 ... linking ... done. Loading package bytestring-trie-0.2.3 ... linking ... done. Loading package JSONb-1.0.3 ... linking ... done. Loading package attempt-0.3.0 ... linking ... done. Loading package convertible-text-0.3.0.9 ... linking ... done. Loading package data-object-json-0.3.1.5 ... linking ... done. Loading package attoparsec-enumerator-0.2.0.3 ... linking ... done. Loading package network-bytestring-0.1.3.4 ... linking ... done. Loading package asn1-data-0.4.6 ... linking ... done. Loading package certificate-0.7.0 ... linking ... done. Loading package primitive-0.3.1 ... linking ... done. Loading package vector-0.7.0.1 ... linking ... done. Loading package cryptocipher-0.2.6 ... linking ... done. Loading package cryptohash-0.6.3 ... linking ... done. Loading package tls-0.3.3 ... linking ... done. Loading package http-enumerator-0.3.1 ... linking ... done. Loading package tagsoup-0.12 ... linking ... done. Loading package xml-1.3.7 ... linking ... done. Loading package authenticate-0.8.2.2 ... linking ... done. Loading package control-monad-attempt-0.3.0 ... linking ... done. Loading package dataenc-0.13.0.5 ... linking ... done. Loading package process-1.0.1.5 ... linking ... done. Loading package mime-mail-0.1.0.1 ... linking ... done. Loading package stm-2.2.0.1 ... linking ... done. Loading package pool-0.0.1 ... linking ... done. Loading package persistent-0.4.1 ... linking ... done. Loading package ranges-0.2.3 ... linking ... done. Loading package email-validate-0.2.6 ... linking ... done. Loading package xss-sanitize-0.2.6 ... linking ... done. Loading package yesod-persistent-0.0.0.1 ... linking ... done. Loading package yesod-form-0.0.0.2 ... linking ... done. Loading package yesod-json-0.0.0.1 ... linking ... done. Loading package yesod-auth-0.3.1 ... linking ... done. Loading package cmdargs-0.6.8 ... linking ... done. Loading package MonadCatchIO-mtl-0.3.0.2 ... linking ... done. Loading package Cabal-1.10.1.0 ... linking ... done. Loading package ghc-binary-0.5.0.2 ... linking ... done. Loading package bin-package-db-0.0.0.0 ... linking ... done. Loading package hpc-0.5.0.6 ... linking ... done. Loading package ghc-7.0.2 ... ghc: /usr/lib/ghc-7.0.2/ghc-7.0.2/libHSghc-7.0.2.a: no string tables, or too many ghc: panic! (the 'impossible' happened) (GHC version 7.0.2 for i386-unknown-linux): loadArchive "/usr/lib/ghc-7.0.2/ghc-7.0.2/libHSghc-7.0.2.a": failed Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug cabal: Error: some packages failed to install: yesod-0.7.1 failed during the building phase. The exception was: ExitFailure 1 Is this really a bug? If yes, is it a known one? I couldn't find anything on the internet so far. Many thanks in advance. Greets, Ertugrul -- nightmare = unsafePerformIO (getWrongWife >>= sex) http://ertes.de/

4 6

ANN: hsmagick 0.5
by Vincent Gerard 13 Mar '11

13 Mar '11

Hello Haskellers, I am currently taking over the maintainership of hsmagick, a FFI bindings library to ImageMagick. Therefore, I am pleased to announce the 0.5 release of hsmagick. http://hackage.haskell.org/package/hsmagick Major changes: ------ * imageToBlob and blobToImage functions using ByteStrings * Use of ForeignPtr instead of Ptr as memory references representation. Thanks to Steffen Siering patches, the internal representation has been changed from Ptr to ForeignPtr. The finalization of ForeignPtr should resolve space leaks issues. * destroyImage function is exposed, to release the memory immediately. * hsmagick builds with GHC 7 The source repository is now hosted on github https://github.com/vincentg/hsmagick Please, do not hesitate to contact me by email or with github for any questions, requests, bugs, patches, anything about hsmagick. Cheers, Vincent Gerard (Please make sure not to reply to haskell(a)haskell.org.)

1 0

binary twidling modules
by rusi 13 Mar '11

13 Mar '11

I am not sure what would be the generic names I am asking for... examples are the nearest I can get. In perl there is pack/unpack http://perldoc.perl.org/perlpacktut.html Likewise in python there is struct module http://docs.python.org/library/struct.html What is/are the Haskell equivalents?

2 1

Haddock fails on parsing Hamlet
by Mathew de Detrich 13 Mar '11

13 Mar '11

Since the site for signing up to haddock mailing list seems to be down ( http://projects.haskell.org/cgi-bin/mailman/listinfo/haddock) I thought I would post this here Haddock fails to parse (and therefore closes prematurely) when generating documentation for the Hamlet package found here ( http://hackage.haskell.org/package/hamlet) This furthermore prevents the hamlet from being built on archlinux using cabal2arch/AUR/bauerbill The error when creating the documentation is Warning: Text.Hamlet.Quasi: We do not support associated types in instances yet. These instances are affected: Text.Hamlet.Quasi.HamletValue (Text.Hamlet.Quasi.Hamlet url), Text.Hamlet.Quasi.HamletValue Text.Blaze.Internal.Html haddock coverage for dist/build/tmp17309/Text/Hamlet/Quasi.hs: 9/14 64% dist/build/tmp17309/Text/Cassius.hs:340:23: parse error on input `AbsoluteUnit' Aborting... On the previous haskell-platform this did not happen Thanks

3 3

ANN: bash-0.0.0
by Jason Dusek 13 Mar '11

13 Mar '11

Many-a-time, have I used higher level languages to generate Bash scripts. Here, for the first time, I have taken the time to write a structured and safe Bash pretty printer, leveraging some work on shell escaping I did some months ago. http://hackage.haskell.org/packages/archive/bash/0.0.0/doc/html/Language-Ba… A wide range of Bash constructs are supported. It's easiest to point out what's not supported at present: . Arithmetic substitutions (those within "$(( ))"). . Bash's test special form (using "[[ ]]"). Ordinary uses of the test command are of course fully supported. . Specialized expression de-referencing forms that allow substitution and substring selection; the only thing supported in this domain is defaulting and length: ${#var} # Length of var in bytes. ${var:-} # If var is not set, yield the empty string. Many other forms that are tedious to get right -- special variable names, array expansions, even eval -- are fully supported. I'll be using this tool to support a config generation system. Another interesting application is generating Bash completion scripts. If you decide to use it and have trouble please let me know; this is the first Language.Something that I have written and it may not present the friendliest interface. Do let me know, also, if the omitted syntactic structures would be helpful to you and I will see what I can do to include them. -- Jason Dusek Linux User #510144 | http://counter.li.org/

4 4