September 2006 - Glasgow-haskell-users

6.6 candidate
by J. Garrett Morris 06 Sep '06

06 Sep '06

Hello, I successfully downloaded and installed ghc-6.5.20060901 on Windows XP (SP2 etc.). However, when attempting to build fps-0.8, I received a large number of errors stemming from gcc being unable to find Stg.h or HsBase.h. As far as I could tell using -v, gcc is still being passed the old path to the include files (ghc-dir\include, perhaps) whereas said files were actually located in ghc-dir\lib\i386-unknown-mingw32\include. Moving the files from the new location to the old solved the compilation problem. Was this a problem in the distributed build, or did I do something wrong when I was installing it? Thanks. /g -- It is myself I have never met, whose face is pasted on the underside of my mind.

3 2

Floating point problems
by Jamie Brandon 05 Sep '06

05 Sep '06

I recently defied my supervisor and used Haskell to write my coursework instead of C. All went well until I needed floating point and started having odd results. As far as I can tell it isn't substantially affecting my results but it is rather embarrassing after slagging off C so much. Here are some examples: *Main> 0.2 + 0.1 0.30000000000000004 *Main> 0.200000000000000 + 0.100000000000000000 0.30000000000000004 *Main> 0.3 + 0.1 0.4 *Main> 0.2 + 0.1 0.30000000000000004 *Main> it + 0.1 0.4 I assume this is a result of the discrepancy between binary and decimal representations of the numbers. Is there any way around? For a start, it would be nice to have a simple way to get 0.1 + 0.2 == 0.3 = True This is with GHC 6.4.1 and GCC 4.0.3 Thanks, Jamie

9 16

GHC non-termination
by Robert Dockins 05 Sep '06

05 Sep '06

Hello all, I've discovered that GHC doesn't deal very well with the following program. It appears to diverge when running the following program with 'runghc'. The main compiler can also be persuaded to diverge in a similar fashion. Hugs exhibits correct behavior, ie, it prints "hello". This is with GHC 6.4.2. I've tested on linux x86 and PPC OS X. newtype Mu a = Roll { unroll :: Mu a -> a } omega :: a omega = (\x -> (unroll x) x) (Roll (\x -> (unroll x) x)) main = putStrLn "hello" Rob Dockins Speak softly and drive a Sherman tank. Laugh hard; it's a long way to the bank. -- TMBG

2 2

Program slower with local definitions
by Michael Marte 05 Sep '06

05 Sep '06

Hello *, I noticed a five percent speed up by making local definitions global. This is the first version with local definitions: f = <some expression based on mySet> where mySet = foldl (flip Set.insert) Set.empty myList myList = <some constant list> Then I moved mySet and myList to the top level which gave me the speed up. Shouldn't both versions show the same behaviour with regard to run time? Doesn't the compiler lift mySet and myList to top level automatically? Are myList and mySet evaluated in every call to f when defined locally? I am using ghc 6.4.2 on Windows XP. Michael

3 2

Re: cvs commit: hugs98 Makefile RPM.mk hugs98/libraries/tools convert_libraries
by Malcolm Wallace 05 Sep '06

05 Sep '06

> > > HaXml (no longer builds) > > > > In what way does HaXml fail to build for Hugs? Is it easily > > fixable? > > ... and there's the famous Data.FiniteMap. So does anyone have any objections if I go ahead and commit the replacement (compatibility) implementation of Data.FiniteMap to the main repository for packages/base? Regards, Malcolm

4 7

instance overaps with lib
by Serge D. Mechveliani 05 Sep '06

05 Sep '06

More about instance overlaps with the GHC library: I need to print in a special way the data of [Equation], (Term, Term), [(Term, Term)], (Equation, Equation). The first can be by defining showList in instance Show Equation. But Show has not a method of showPair. So, I need to write the function showsTermPair and to use it together with another home-made function showsListGeneric. And in ghc-6.4.1, it was sufficient to define the instances and to use only `shows'. About prompt feed-backs from users: this issue of possible fisfeature with overlaps does not look tragic. But with the Map library for Haskell-98, I take it as a considerable loss. It was said that I had a chance to make it standard, if asked about it in time. ----------------- Serge Mechveliani mechvel(a)botik.ru

2 2

overlapping instances in 6.6 candidate
by Serge D. Mechveliani 04 Sep '06

04 Sep '06

Dear GHC developers, Can you, please, explain what has happened in 6.5.20060831 with treating of overlapping instances? It seems to understand them differently than ghc-6.4.1. For example, I compile the module ------------------------ class C a where c :: a -> Bool instance C Int where c n = n > 0 instance C a => C [a] where c xs = and $ map c xs instance C [Int] where c xs = c $ head xs main = putStr (shows (c [1, (-2 :: Int)]) "\n") instance Show [Bool] where showsPrec _ xs = shows (head xs) ------------------------ with the options $dmCpOpt = -fglasgow-exts -fallow-undecidable-instances -fallow-overlapping-instances -fno-warn-overlapping-patterns -fwarn-unused-binds -fwarn-unused-matches -fwarn-unused-imports : > ghc $dmCpOpt --make Main It reports -------------------------------------------------------- [1 of 1] Compiling Main ( Main.hs, Main.o ) Main.hs:11:0: Overlapping instances for Show [Bool] arising from use of `GHC.Show.$dmshowList' at Main.hs:11:0 Matching instances: instance (Show a) => Show [a] -- Defined in GHC.Show instance [overlap ok] Show [Bool] -- Defined at Main.hs:11:0 In the expression: GHC.Show.$dmshowList In the definition of `showList': showList = GHC.Show.$dmshowList In the definition for method `showList' -------------------------------------------- And it takes as all right the overlaps for the class C. [Bool] is a substitutional instance of [a], and according to the GHC specification, the Show instance for [Bool] must override the more generic one for Show a => [a]. Thank you in advance for explanation. ----------------- Serge Mechveliani mechvel(a)botik.ru

2 1

Re[2]: [GHC] #876: stack overflow on 'length . filter odd $ [0 .. 999999999]'
by Bulat Ziganshin 04 Sep '06

04 Sep '06

Hello GHC, Tuesday, August 29, 2006, 10:54:07 PM, you wrote: > It makes sense to me that the above behaviour is seen: length is now a > good > consumer, but it generates 1+(1+(1+(1+... as it is consuming, and this > causes a stack overflow. I don't think we can fix this while staying with > fold/build fusion, so it looks to me like the patch should be reverted and > the whole problem looked at post-6.6. in general, for any function we can either implement 1) good consumer based on using foldr 2) tail-recursion are you agree? in particular, filterM, replicateM and other *M list operations (their problems was mentioned earlier) can be implemented in first or second way, but not both may be this can be fixed with some smart rules, which selects first or second implementation depending on context? -- Best regards, Bulat mailto:Bulat.Ziganshin@gmail.com

5 7

binary distribution of ghc-6.5.20060831 for i686-apple-darwin
by Hiroaki Nakamura 04 Sep '06

04 Sep '06

Hello! I am a newbie to Haskell, and this is my first time to post here. Well, I successfully built a binary distribution of ghc-6.5.20060831 for i686-apple-darwin. So I would like to contribute this, but is there a place to upload this binary distribution? What I did is: First I built ghc-6.5.20060828 using binary distribition http://pugs.blogs.com/dist/ghc-6.5.20060608.tar.bz2 contributed by AudreyTang. Thanks to useful information on http://cvs.haskell.org/trac/ghc/wiki/X86OSXGhc I followed the instruction about unpull and pull the "Fix Array imports" patch. Then I built ghc-6.5.20060831 using ghc-6.5.20060828 and it seems ok. But I am not sure because I am just a newbie to Haskell. I think it is useful to have a binary distribution on the GHC home page so that other newbie Intel Mac users can test this release candidate easily without compiling themselves. Thanks in advance. -- )Hiroaki Nakamura) hnakamur(a)gmail.com

2 1

making 6.6-candidate
by Serge D. Mechveliani 04 Sep '06

04 Sep '06

Dear GHC team, I am trying to make ghc-6.5.20060831 from source by ghc-6.4.1 under Linux, i386-unknown. It has got to /home/mechvel/ghc/6.4.1/inst/bin/ghc -H16m -O -istage1/utils -istage1/basicTypes -istage1/types -istage1/hsSyn -istage1/prelude -istage1/rename -istage1/typecheck -istage1/deSugar -istage1/coreSyn -istage1/specialise -istage1/simplCore -istage1/stranal -istage1/stgSyn -istage1/simplStg -istage1/codeGen -istage1/main -istage1/profiling -istage1/parser -istage1/cprAnalysis -istage1/ndpFlatten -istage1/iface -istage1/cmm -istage1/nativeGen -Istage1 -cpp -fglasgow-exts -fno-generics -Rghc-timing -I. -IcodeGen -InativeGen -Iparser -package unix -ignore-package lang -recomp -Rghc-timing -H16M '-#include "cutils.h"' -i../compat -ignore-package Cabal -funbox-strict-fields -c parser/Lexer.hs -o stage1/parser/Lexer.o -ohi stage1/parser/Lexer.hi and looped at this point during about 50 minutes, without any report. Finally, it was done, and now, I hope, it will make the rest. But I wonder whether this loop is suspicious. The command `top' has been showing CPU %MEM 20633 mechvel ... 99.9 2.1 ... ghc-asm for 50 minutes (`as' is assembly, and who is ghc-asm, I wonder). The machine is i-686, 512 Mb RAM, vendor_id : GenuineIntel cpu family : 6 model : 8 model name : Pentium III (Coppermine) stepping : 3 cpu MHz : 601.371 cache size : 256 KB autoreconf has reported it could not find this and that, but it is, probably, all right (?) config.log is --------------------------------------- ... It was created by The Glorious Glasgow Haskell Compilation System configure 6.5.20060831, which was generated by GNU Autoconf 2.59. Invocation command line was $ ./configure --prefix=/home/mechvel/ghc/6.6.pre/inst uname -m = i686 uname -r = 2.4.27-2-686-smp uname -s = Linux uname -v = #1 SMP Mon May 16 16:55:31 JST 2005 ... ------------------------------------ Regards, ----------------- Serge Mechveliani mechvel(a)botik.ru

2 1