February 2010 - Haskell-Cafe

Scarry parallel garbage collection behavior (6.12.1, Mac OSX)
by Pavel Perikov 22 Feb '10

22 Feb '10

Hi, list. This is a long post, sorry. The bottom line: specifying +RTS -Nn where n > number of cores slows the program compiled with 6.12.1 by orders of magnitude under Mac OSX. Turning off parallel garbage collections with -qg resolves the problem. Independent verification is appreciated. I have a very simple program: import Control.Parallel main=do let s = s1 `par` ( s2 `pseq` s1 + s2) s1 = sum [1..10000000] s2 = sum [1..10000005] print s compiled with ghc -O -threaded running it with without any options time ./clocktime produces the following times: real 0m3.779s user 0m3.644s sys 0m0.053s Giving -N option to RTS (or -N2, I have two core macbook) shows decent speedup: 100000060000015 real 0m2.678s user 0m4.762s sys 0m0.043s though increase in processor time by 30% seems suspicious. Things become weird if I increase the number of threads. I couldn't expect any further speedup for this task but what I get is absolutely unexpected: time ./clocktime +RTS -N3 100000060000015 real 0m57.071s user 1m40.910s sys 0m0.921s Almost a minute!! Adding more threads does not help either :) time ./clocktime +RTS -N4 100000060000015 real 1m48.281s user 3m10.832s sys 0m1.590s A bit of playing with Instruments/DTrace shows that runaway garbage collection puts the program on its knees. I the case of -N1 the top time consumers look like this: Self % Total % Self Run % Running % # Self # Samples ms Self Run ms Running Parent % Library Source Path Symbol Name 12.8 12.8 408 clocktime __gmpz_add 10.7 10.7 341 clocktime integer_cmm_plusIntegerzh 10 10 321 clocktime allocateLocal etc. This looks absolutely reasonable. Going to -N2 changes the picture to: Self % Total % Self Run % Running % # Self # Samples ms Self Run ms Running Parent % Library Source Path Symbol Name 15.6 15.6 989 clocktime gcWorkerThread 11.6 11.6 736 clocktime waitForGcThreads 9.5 9.5 603 clocktime scavenge_until_all_done 7.4 7.4 471 clocktime setContextSwitches 6.2 6.2 396 clocktime __gmpz_add -N3 Is awful: Self % Total % Self Run % Running % # Self # Samples ms Self Run ms Running Parent % Library Source Path Symbol Name 51.7 51.7 29514 clocktime gcWorkerThread 25.5 25.5 14591 clocktime scavenge_until_all_done 7.2 7.2 4114 clocktime waitForGcThreads 5.5 5.5 3180 clocktime GarbageCollect 5.2 5.2 2998 clocktime setContextSwitches 0.4 0.4 240 clocktime integer_cmm_plusIntegerzh Pavel

2 2

A small oversight
by Andrew Coppin 22 Feb '10

22 Feb '10

I just discovered the highly useful function Data.Function.on. I vaguely recall a few people muttering a couple of years back that this would be a useful thing to have, but I had no idea it was in the standard libraries now. Anyway, while using it, I discovered a small omission from the Haskell libraries: We have min, max, minimum and maximum. We also have minimumBy and maximumBy. But there's no sign of minBy or maxBy. You can do, for example, (min `on` postcode) customer1 customer2 but that gives you the lowest postcode, not the customer to which this postcode belongs. By contrast, minimumBy (compare `on` postcode) [customer1, customer2] gives you the corresponding customer. But it seems silly to have to actually construct a list just for this. So... is there any danger of getting minBy and maxBy added? (I don't actually know off the top of my head where min and max are defined - the Prelude?) Also, constructions like sortBy (compare `on` foo) must surely be very common. Perhaps it would be an idea to define a family of functions like sortOn :: (Ord y) => (x -> y) -> [x] -> [x] sortOn foo = sortBy (compare `on` foo) Just an idea. Finally, take a look at this: newtype SwapOrd x = SwapOrd (unswap_ord :: x) deriving Eq instance Ord x => Ord (SwapOrd x) where compare x y = swap_ord $ compare x y swap_ord :: Ordering -> Ordering swap_ord o = case o of EQ -> EQ GT -> LT LT -> GT Just in case you wanted to sort things in reverse order. I think having swap_ord would be nice if nothing else...

7 7

Pointfree composition for higher arity
by Sean Leather 22 Feb '10

22 Feb '10

I find myself often writing this pattern: someFun x y z = ... fun y z = runFun $ someFun someDefault y z > or, alternatively: fun y = runFun . someFun someDefault y > The second option approaches the ideal pointfreeness (or pointlessness if you prefer), but I'd like to go farther: (...) :: (c -> d) -> (a -> b -> c) -> a -> b -> d > (...) f g x y = f (g x y) > infixr 9 ... > > fun = runFun ... someFun someDefault > There, that's better. More points for fewer points (which means I should really change the name from fun to pun). Does anybody else care about this? What are some alternative solutions? I'd love to have something like this available in the Prelude or a library. (I have no strong feelings about the particular operator.) Regards, Sean

9 14

Re: [Haskell] Functional Programming User Group Ghent
by minh thu 22 Feb '10

22 Feb '10

[moving to café from haskell] 2010/2/21 Jeroen Janssen <jejansse(a)gmail.com>: > Dear all, > We are in the progress of starting up a Functional Programming User Group in > Ghent. The basic idea of the group is to have occasional informal meetings > where people can give talks, where we have invited speakers, where we go to > a bar to chat about all things Functional Programming, to organize Haskell > Hacking events etc. > To see if enough people would be interested in such a group, and to make the > organizing process of the first meeting easier, we would like to ask anyone > interested to tick off the dates he/she is available on the Doodle-page > provided below. Note that the indicated time is only a rough guideline; if > you would prefer another time of day, you can indicate this in the Comments > section of the Doodle. > http://www.doodle.com/xq2i5hsueidwkh2f > If enough people are interested, and a date is decided upon, we will e-mail > the Haskell list once more to confirm the date and give anyone who was > interested, but could not fill in the doodle, a chance to be present. Also, > this ensures that you do not have to leave your e-mail address on Doodle. > If you have any questions, do not hecessitate to contact one of us at either > Bart.Coppens(a)UGent.be > jejansse(a)gmail.com > Kind Regards, > Bart Coppens > Jeroen Janssen Hi Jeroen, Great, a fp user group in Belgium! Alas my dutch is too rusty too even understand the Doodle poll description. Is the group planned only for dutch-speaking programmers ? Cheers, Thu

3 3

Haskell Weekly News: Issue 150 - February 22, 2010
by jfredett＠gmail.com 22 Feb '10

22 Feb '10

--------------------------------------------------------------------------- Haskell Weekly News http://sequence.complete.org/hwn/20100222 Issue 150 - February 22, 2010 --------------------------------------------------------------------------- Welcome to issue 150 of HWN, a newsletter covering developments in the [1]Haskell community. From the very pits of Harddrive failure I return... The Haskell Weekly News Reborn! Last week, on friday, my HWN machine suffered what can only be appropriately named a unbelievably poorly timed hard drive failure, hence the lack of an HWN. I trumphently return this week with a megaedition. Tons and tons of awesome stuff happened in the last two weeks. So I won't make you wait any longer, haskellers, your Haskell Weekly News! Announcements vty-ui 0.3. Jonathan Daugherty [2]announced a new version of vty-ui, featuring a richer rendering engine, generalized text transformations, and a more functional style. darcs 2.4 release candidate 2. Reinier Lamers [3]announced a new release candidate for darcs 2.4. iteratee-parsec 0.0.1. Maciej Piechotka [4]announced the release of iteratee-parsec, a library whih provides support for parsec parsers in an Iteratee monad. concurrent-extra-0.1. Roel van Dijk [5]announced the release of concurrent-extra, which provides a few new synchronization primitives. wyvern, a Dragon Go Server 'client'. wagnerdm [6]announced wyvern, a Dragon Go Server 'client', which plays moves for you on the [7]Dragon Go Server. atom-1.0.0. Tom Hawkins [8]announced a new major release of the atom library. Call for Copy: Monad.Reader Issue 16. Brent Yorgey [9]announced the call for copy for Monad.Reader issue 16. hmatrix 0.8.3. Alberto Ruiz [10]announced a new version of hmatrix. First Atlanta Function Programming Users Group Meeting. David Vollbracht [11]announced the first Atlanta FP Group meetup. TFP 2010 - Call for Papers and Participation. Simon Marlow [12]announced a call for papers and particpation in TFP 2010, the 11th Symposium on Trends in Functional Programming. CMCS 2010: Call for Short Submissions & Call for Participation. Alexandra Silva [13]announced a call for short submissions and participation in the 10th International Workshop on Coalgebraic Methods in Computer Science. WGP 2010 Call for Papers. Bruno Oliveira [14]announced a call for papers for Workshop on Generic Programming. ICE 2010: First call for papers. Alexandra Silva [15]announced the first call for papers for ICE2010, the 3rd Interation and Concurrency Experience workshop. Call for Papers: Haskell Symposium 2010. Jeremy.Gibbons [16]announced a call for papers for the 2010 Haskell Symposium. Dungeons of Wor - a largish FRP example and a fun game, all in one! Patai Gergely [17]announced Dungeons of Wor, a Game written in Haskell using the Elerea library and Functional Reactive Techniques. Discussion The Related monad and constant values in type classes. Jonas Almstroem Duregard [18]talked about a common pattern which occurs when associating data with a type (as opposed to a value). Heterogeneous Data Structures - Nested Pairs and functional references. Guenther Schmidt [19]asked about different ways to implement heterogeonous datastructures. Blog noise [20]Haskell news from the [21]blogosphere. Blog posts from people new to the Haskell community are marked with >>>, be sure to welcome them! * Michael Snoyman: [22]Four HTTP Request Body Interfaces (repost). * Brent Yorgey: [23]Math.Combinatorics.Multiset. * Darcs: [24]darcs weekly news #55. * Galois, Inc: [25]Tech Talk: Modern Benchmarking in Haskell. * Ketil Malde: [26]Tools for pyrosequencing analysis. * Arch Haskell News: [27]Arch Haskell News: February 2010. * Don Stewart (dons): [28]Migrating from uvector to vector. * GHC / OpenSPARC Project: [29]Memory Barriers and GHC 6.12.1. * Darcs: [30]darcs weekly news #54. * Gergely Patai: [31]Behind the dungeons. * David Amos: [32]How to find a strong generating set. Quotes of the Week * fasta: And by cool, I mean 'that is not supposed to happen'. * uncyclopedia: Calculations which are undefined are denoted by the _|_ symbol, pronounced Bottom, which the documentation explains as the compiler giving you the finger. * Draconx|Laptop: mathematics is the art of finding different ways to write "therefore". * medfly: Cale, #haskell wants to know all your personal preferences so we can all copy you * Cale: Gentoo is a massive waste of electricity. * hobophobe: So, I can only conclude that Haskell is a memetic virus, and monads are the eggs it lays out in innocent programming forums to entice others to become infected. * benmachine: unsafeCoerce is just a generalisation of id * jmcarthur: what good is a state monad if your state is basically 'i have no idea what the hell my state is'? About the Haskell Weekly News New editions are posted to [33]the Haskell mailing list as well as to [34]the Haskell Sequence and [35]Planet Haskell. [36]RSS is also available, and headlines appear on [37]haskell.org. To help create new editions of this newsletter, please see the information on [38]how to contribute. Send stories to jfredett . at . gmail . dot . com. The darcs repository is available at darcs get [39]http://patch-tag.com/r/jfredett/HWN2/pullrepo HWN2 . References 1. http://haskell.org/ 2. http://article.gmane.org/gmane.comp.lang.haskell.cafe/70589 3. http://article.gmane.org/gmane.comp.lang.haskell.cafe/70563 4. http://article.gmane.org/gmane.comp.lang.haskell.cafe/70533 5. http://article.gmane.org/gmane.comp.lang.haskell.cafe/70485 6. http://article.gmane.org/gmane.comp.lang.haskell.cafe/70344 7. http://www.dragongoserver.net/ 8. http://article.gmane.org/gmane.comp.lang.haskell.cafe/70328 9. http://article.gmane.org/gmane.comp.lang.haskell.cafe/70314 10. http://article.gmane.org/gmane.comp.lang.haskell.cafe/70290 11. http://article.gmane.org/gmane.comp.lang.haskell.general/17802 12. http://article.gmane.org/gmane.comp.lang.haskell.general/17799 13. http://article.gmane.org/gmane.comp.lang.haskell.general/17791 14. http://article.gmane.org/gmane.comp.lang.haskell.general/17783 15. http://article.gmane.org/gmane.comp.lang.haskell.general/17779 16. http://article.gmane.org/gmane.comp.lang.haskell.general/17778 17. http://article.gmane.org/gmane.comp.lang.haskell.cafe/70188 18. http://thread.gmane.org/gmane.comp.lang.haskell.cafe/70492 19. http://thread.gmane.org/gmane.comp.lang.haskell.cafe/70450 20. http://planet.haskell.org/ 21. http://haskell.org/haskellwiki/Blog_articles 22. http://snoyberg.wordpress.com/2010/02/21/four-http-request-body-interfaces-… 23. http://byorgey.wordpress.com/2010/02/20/math-combinatorics-multiset/ 24. http://blog.darcs.net/2010/02/darcs-weekly-news-55.html 25. http://www.galois.com/blog/2010/02/19/tech-talk-modern-benchmarking-in-hask… 26. http://blog.malde.org/index.php/2010/02/19/tools-for-pyrosequencing-analysi… 27. http://archhaskell.wordpress.com/2010/02/19/arch-haskell-news-february-2010/ 28. http://donsbot.wordpress.com/2010/02/15/migrating-from-uvector-to-vector/ 29. http://ghcsparc.blogspot.com/2010/02/memory-barriers-and-ghc-6121.html 30. http://blog.darcs.net/2010/02/darcs-weekly-news-54.html 31. http://just-bottom.blogspot.com/2010/02/behind-dungeons.html 32. http://haskellformaths.blogspot.com/2010/02/how-to-find-strong-generating-s… 33. http://www.haskell.org/mailman/listinfo/haskell 34. http://sequence.complete.org/ 35. http://planet.haskell.org/ 36. http://sequence.complete.org/node/feed 37. http://haskell.org/ 38. http://haskell.org/haskellwiki/HWN 39. http://patch-tag.com/r/jfredett/HWN2/pullrepo%20HWN2

1 0

Threading and FFI
by Yves Parès 22 Feb '10

22 Feb '10

Hello! I was wondering: is the Haskell default runtime (that which uses only one processor) scheduler able to interrupt a thread which is currently calling to a C function in order to enable another haskell thread to progress? I think it can't, because I have starvation problems when I have a thread which calls inside its loop to a C function which makes a little "sleep". BTW, what is the portable way to sleep a thread in Haskell ? Control.Concurrent.threadDelay is "GHC only". ----- Yves Parès Live long and prosper -- View this message in context: http://old.nabble.com/Threading-and-FFI-tp27611528p27611528.html Sent from the Haskell - Haskell-Cafe mailing list archive at Nabble.com.

8 16

Installing wx via cabal
by Dušan Kolář 22 Feb '10

22 Feb '10

Dear all, Running cabal install --global wx stops very soon with the following error. Building wxcore-0.12.1.2... [ 1 of 22] Compiling Graphics.UI.WXCore.WxcObject ( src/haskell/Graphics/UI/WXCore/WxcObject.hs, dist/build/Graphics/UI/WXCore/WxcObject.o ) [ 2 of 22] Compiling Graphics.UI.WXCore.WxcDefs ( src/haskell/Graphics/UI/WXCore/WxcDefs.hs, dist/build/Graphics/UI/WXCore/WxcDefs.o ) [ 3 of 22] Compiling Graphics.UI.WXCore.WxcClassTypes ( src/haskell/Graphics/UI/WXCore/WxcClassTypes.hs, dist/build/Graphics/UI/WXCore/WxcClassTypes.o ) [ 4 of 22] Compiling Graphics.UI.WXCore.WxcTypes ( src/haskell/Graphics/UI/WXCore/WxcTypes.hs, dist/build/Graphics/UI/WXCore/WxcTypes.o ) [ 5 of 22] Compiling Graphics.UI.WXCore.WxcClassesAL ( src/haskell/Graphics/UI/WXCore/WxcClassesAL.hs, dist/build/Graphics/UI/WXCore/WxcClassesAL.o ) ghc: out of memory (requested 1048576 bytes) cabal: Error: some packages failed to install: It seems quite strange as the machine seems to have 16GB RAM and a free swap, can I somehow put options or so? I have ghc 10.4. installed. Thanks Dusan

3 2

games of chance
by Dupont Corentin 21 Feb '10

21 Feb '10

Hello, are you aware of a framework to find probabilities in a given game of chance, in Haskell? l would like to compute probabilities like: "roll 3 dices, sum the result, then roll 3 other dices, sum the result, what is the probability to obtain the same sum?" Cheers, Corentin

2 2

Re: Parsing of bytestrings with non-String errors?
by Permjacov Evgeniy 21 Feb '10

21 Feb '10

On 02/21/2010 11:57 PM, haskell-cafe-request(a)haskell.org wrote: > Message: 2 > Date: Sun, 21 Feb 2010 12:36:21 +0000 > From: Magnus Therning <magnus(a)therning.org> > Subject: [Haskell-cafe] Parsing of bytestrings with non-String errors? > To: haskell-cafe <haskell-cafe(a)haskell.org> > Message-ID: <4B8128C5.6030105(a)therning.org> > Content-Type: text/plain; charset="utf-8" > > I've looked at polyparse and attoparsec and they seem to have in common that > the error always is a String. My current ideas for a project would be a lot > easier if I could just return some other type, something that I can pattern > match on. > > Is there a parser combinator library out there that works on bytestrings and > allows using a custom error type? > > Or maybe there's some very basic reason why String is so commonly used? > > You can try to play with ParsecT / ErrorT

1 0

Re: [Haskell-cafe] Profiling
by Patai Gergely 21 Feb '10

21 Feb '10

> I tried to compile my program with -prof, but GHC just whines at me > that the packages I'm using haven't been compiled for profiling. Do I > really need to go recompile every single package I'm using with > profiling support before I can profile my program? How do I tell Cabal > to install the necessary code? (I really hope this doesn't involve > uninstalling and reinstalling everything...) Almost. You don't need to uninstall anything, just run cabal install <package> --reinstall -p for every relevant package. Gergely -- http://www.fastmail.fm - mmm... Fastmail...

4 7