- Haskell-Cafe - Haskell.org

STM unperformance
by Wojtek Narczyński 18 Feb '16

18 Feb '16

Dear list, I create four haskell threads, each performs disjoint STM transactions. The more system threads I add, the slower the program runs. With four system threads the performance is 14% of the nonparallel run. What is going on? $ ghc -v Glasgow Haskell Compiler, Version 7.10.3, stage 2 booted by GHC version 7.8.4 $ ghc -O2 -threaded test-stm.hs $ time ./test-stm +RTS -N1 real 1.843s $ time ./test-stm +RTS -N2 real 7.469s $ time ./test-stm +RTS -N3 real 9.640s $ time ./test-stm +RTS -N4 real 13.144s $ cat test-stm.hs {-# LANGUAGE ScopedTypeVariables #-} import Control.Monad import Control.Concurrent import Control.Concurrent.STM import Control.Concurrent.STM.Stats import System.IO.Unsafe -- Copied from GHC docs children :: MVar [MVar ()] children = unsafePerformIO (newMVar []) waitForChildren :: IO () waitForChildren = do cs <- takeMVar children case cs of [] -> return () m:ms -> do putMVar children ms takeMVar m waitForChildren forkChild :: IO () -> IO ThreadId forkChild io = do mvar <- newEmptyMVar childs <- takeMVar children putMVar children (mvar:childs) forkFinally io (\_ -> putMVar mvar ()) -- Test case main = do forkChild $ newTVarIO 0 >>= incrManyTimes "thread1" forkChild $ newTVarIO 0 >>= incrManyTimes "thread2" forkChild $ newTVarIO 0 >>= incrManyTimes "thread3" forkChild $ newTVarIO 0 >>= incrManyTimes "thread4" waitForChildren dumpSTMStats -- Confirms no conflicts incrManyTimes :: String -> TVar Int -> IO () incrManyTimes l = incrRec (1000000 :: Int) where incrRec n v | n == 0 = pure () | otherwise = trackNamedSTM l (modifyTVar v (+1)) >> incrRec (n-1) v -- Wojtek

2 2

GHC worker/wrapper on multiple return values
by Tyson Whitehead 18 Feb '16

18 Feb '16

Recently discovered GHC generates sub-optimal code when you have - two levels of loops in the IO monad - the inner loop doesn't do that much work - the outer loop does a lot of work From dumping core I believe the issue is the worker/wrapper transformation doesn't produce return value unboxing (from the inner loop to the outer in this case) when there are multiple return values (as you windup with the IO monad due to passing around the State# RealWorld behind the scenes). Here is some simple code that demonstrates the issue. With a single return value {-# LANGUAGE BangPatterns #-} module Main (main) where main :: IO () main = do let a = l1 0 100000000 print a l1 :: Int -> Int -> Int l1 !a !n = case n>0 of True -> let m = l2 0 n 3 in l1 (a+m) (n-1) False -> a l2 :: Int -> Int -> Int -> Int l2 !a !n !m = case m>0 of True -> l2 (a+n) (n-1) (m-1) False -> a you get nice clean core that performs no allocations and runs beautifully. Doing the same thing within the IO monad (effectively lifting the return value above into a unboxed tupple) {-# LANGUAGE BangPatterns, MagicHash, UnboxedTuples #-} module Main (main) where main :: IO () main = do a <- l1 0 100000000 print a l1 :: Int -> Int -> IO Int l1 !a !n = do case n>0 of True -> do m <- l2 0 n 3 l1 (a+m) (n-1) False -> return a l2 :: Int -> Int -> Int -> IO Int l2 !a !n !m = do case m>0 of True -> l2 (a+n) (n-1) (m-1) False -> return a you don't get boxing and unboxing the return value between l1 and l2. This has quite a dramatic effect when l2 doesn't do the that much work (massive amounts of additional memory allocation and a 2x slow down). $ time ./Simple-pure 14999999850000000 51,912 bytes allocated in the heap ... real 0m0.322s ... $ time ./Simple-io 14999999850000000 1,600,051,936 bytes allocated in the heap ... real 0m0.634s Should I open a ticket against GHC about this issue? Possibly it wouldn't be that hard to get the work/wrapper stuff to work its unboxing magic on multiple return values too? Cheers! -Tyson [*] https://gist.github.com/twhitehead/5744eee28bcde85b9fa4

1 1

[announcement] filtrable: class of filtrable containers
by M Farkas-Dyck 18 Feb '16

18 Feb '16

I quietly posted this library to Hackage nearly a year ago, but lately learned that some seeking such a package had difficulty finding it, so i announce it now ☺ https://hackage.haskell.org/package/filtrable class Functor f => Filtrable f where mapMaybe :: (a -> Maybe b) -> f a -> f b catMaybes :: f (Maybe a) -> f a filter :: (a -> Bool) -> f a -> f a For laws, see docs on Hackage.

12 19

Haskell.org Committee Financial Statement 2015
by Gershom B 17 Feb '16

17 Feb '16

Dear Haskellers, The Haskell.org Committee [1] manages funds for haskell.org and oversees haskell.org infrastructure. The funds available to Haskell.org generally come from two sources: 1) Mentor payments from the Google Summer of Code program. 2) Since the end of 2013, occasional donations [2] from the Haskell community at large. Our funds are held by Software in the Public Interest, a 501(c)3 US non-profit, through which we also accept donations. In return for its services, SPI receives 5% of donations to Haskell.org. According to our charter, "Each year, the committee will post a statement of the haskell.org assets, and the transactions for that year." Included in this message is a brief statement of Haskell.org assets over the period 31 December 2014 - 31 December 2015. Note that our expenses are nearly entirely hosting, with a small overhead incurred by SPI and payment processing. The bulk of our income this year, over which we did minimal fundraising, is $7,000 in mentor payments from google summer of code 2014 (mentor payments for 2015 have not yet been received). Hosting fees are down from years past, as we have migrated more of our infrastructure away from Hetzner and to Rackspace, which donates to us free hosting. 1. Income and Expenses Total income over 2014: 7751.46 Total expenses over 2014: 892.03 ---- Net income over 2014: 6859.43 3. Total Balance Balance as of 31 December 2014: 26,422.91 Balance as of 31 December 2015: 33,547.34 Note: As of December 2015, we have a moved a large chunk of our balance out of the SPI-held account and to a new account set up directly under the control of Haskell.Org, as Haskell.Org is now registered as a nonprofit 501(c)(3). [1] https://wiki.haskell.org/Haskell.org_committee [2] https://wiki.haskell.org/Donate_to_Haskell.org Best, Gershom Bazerman for the Haskell.org Committee

1 1

Re: [Haskell-cafe] Yesod CSRF token
by Geraldus 17 Feb '16

17 Feb '16

Can you share your code? If you make your POST requests via forms make sure that you generation forms with Yesod functions, in this case it will insert hidden field with CSRF token to resulting form widget automatically. If you sending requests via XHR make sure to use `defaultCsrfMiddleware` from Yesod.Core. Hope this helps. ср, 17 февр. 2016 г. в 15:38, fr33domlover <fr33domlover(a)riseup.net>: > Hello, > > I started a new Yesod web app few days ago. I'm using stack, LTS-5.1, and > the > yesod-postgres scaffolding. > > When I try to log in into my new minimal app, I get an error: A valid CSRF > token wasn't present in HTTP headers or POST parameters. That's because the > POST request doesn't include that token for some reason. But I don't know > why. > > I found a recent PR which removes the CSRF token checking from the > scaffolding, > but I do want CSRF protection to work. As a Yesod beginner, I'm not sure > exactly why the token doesn't get inserted where it should (the Yesod book > says > yesod-form does insert it, so I assume this is a bug here) and how I can > fix > that. Even if I inserted the token manually into the login form, what > about all > the other POST requests my app may use, such as logout? > > I also found a commit that adds the token to the redirectToPost function in > yesod-core, but > > (1) It's in the most recent release, not in LTS > (2) I'm not sure it has anything to do with it because it seems to be some > sort > of JS based form > > Anyone knows whether this is a known issue and how to fix it? > > > > Thanks! > --fr33 > _______________________________________________ > Haskell-Cafe mailing list > Haskell-Cafe(a)haskell.org > http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe >

2 1

Generalising Categories - categories of tuples
by Clinton Mead 17 Feb '16

17 Feb '16

Hi All I've been doing some work on generalising categories, basically so the composition operator (.) works for not just functions, but things like tuples of functions, whilst still retaining type inference. Control.Category is a bit too restrictive for such instances. You can see the details in my blog post here: https://clintonmeadprogramming.wordpress.com/2016/02/16/generalising-catego… Any comments appreciated, I hope people find it interesting. Also, a quick plug, I'm looking for work, so if there's any Haskell (and/or functional programing) work around Sydney, Australia going I'd be interested. :-) If people really think this is useful, tell me, and I'll try to put it up as a package. Alternatively, if it's already been done, I'd appreciate a pointer in that direction. Regards, Clinton Mead

3 4

Yesod CSRF token
by fr33domlover 17 Feb '16

17 Feb '16

Hello, I started a new Yesod web app few days ago. I'm using stack, LTS-5.1, and the yesod-postgres scaffolding. When I try to log in into my new minimal app, I get an error: A valid CSRF token wasn't present in HTTP headers or POST parameters. That's because the POST request doesn't include that token for some reason. But I don't know why. I found a recent PR which removes the CSRF token checking from the scaffolding, but I do want CSRF protection to work. As a Yesod beginner, I'm not sure exactly why the token doesn't get inserted where it should (the Yesod book says yesod-form does insert it, so I assume this is a bug here) and how I can fix that. Even if I inserted the token manually into the login form, what about all the other POST requests my app may use, such as logout? I also found a commit that adds the token to the redirectToPost function in yesod-core, but (1) It's in the most recent release, not in LTS (2) I'm not sure it has anything to do with it because it seems to be some sort of JS based form Anyone knows whether this is a known issue and how to fix it? Thanks! --fr33

1 0

FP Complete developers survey
by Tom Ellis 16 Feb '16

16 Feb '16

Last year FP Complete ran a large survey of Haskell developers. I can only find a synopsis here https://www.fpcomplete.com/blog/2015/05/thousand-user-haskell-survey but my memory tells me they also had a blog post with a more in-depth analysis of the numbers. I can't find it on their website. Does anyone know where it is, or am I imagining it? Tom

3 4

Worked up examples of Simon Marlow's code hot-swapping
by Christopher Allen 16 Feb '16

16 Feb '16

https://twitter.com/raichoo/status/699677135280205826 Mentioned here: https://code.facebook.com/posts/745068642270222/fighting-spam-with-haskell https://phabricator.haskell.org/rGHCbdfefb3b72a71cd0afca6e7766456c0d97c47c86 Are there any out there? Most stuff I've found for this is Plugins or something else. I suggested on Twitter it might be similar to the dynamic linking trick game devs use but I don't really know anything about what Simon specifically did. If anyone could please point to examples or deeper explanation that would be a great help, thank you. --- Chris Allen

1 0

[TFPIE 2016] 1st call for papers
by Peter Achten 16 Feb '16

16 Feb '16

Trends in Functional Programming in Education (TFPIE 2016) Call for papers https://wiki.science.ru.nl/tfpie/TFPIE2016 The 5th International Workshop on Trends in Functional Programming in Education, TFPIE 2016, will be held on June 7, 2016 at the University of Maryland College Park in the USA. It is co-located with the Symposium on Trends in Functional Programming (TFP 2016) which takes place from June 8 - 10. *** Goal *** The goal of TFPIE is to gather researchers, teachers and professionals that use, or are interested in the use of, functional programming in education. TFPIE aims to be a venue where novel ideas, classroom-tested ideas and work-in-progress on the use of functional programming in education are discussed. The one-day workshop will foster a spirit of open discussion by having a review process for publication after the workshop. The program chair of TFPIE 2016 will screen submissions to ensure that all presentations are within scope and are of interest to participants. Potential presenters are invited to submit an extended abstract (4-6 pages) or a draft paper (up to 16 pages) in EPTCS style. The authors of accepted presentations will have their preprints and their slides made available on the workshop's website/wiki. Visitors to the TFPIE 2016 website/wiki will be able to add comments. This includes presenters who may respond to comments and questions as well as provide pointers to improvements and follow-up work. After the workshop, presenters will be invited to submit (a revised version of) their article for review. The PC will select the best articles for publication in the journal Electronic Proceedings in Theoretical Computer Science (EPTCS). Articles rejected for presentation and extended abstracts will not be formally reviewed by the PC. TFPIE workshops have previously been held in St Andrews, Scotland (2012), Provo Utah, USA (2013), Soesterberg, The Netherlands (2014), and Sophia-Antipolis, France (2015). *** Program Committee *** Stephen Chang at Northeastern University in Massachusetts, USA Marc Feeley at Université de Montréal in Québec, Canada Patricia Johann at Appalachian State University in North Carolina, USA Jay McCarthy at University of Massachusetts Lowell in Massachusetts, USA (Chair) Prabhakar Ragde at University of Waterloo in Ontario, Canada Brent Yorgey at Hendrix College in Arkansas, USA *** Submission Guidelines *** TFPIE 2016 welcomes submissions describing techniques used in the classroom, tools used in and/or developed for the classroom and any creative use of functional programming (FP) to aid education in or outside Computer Science. Topics of interest include, but are not limited to: - FP and beginning CS students - FP and Computational Thinking - FP and Artificial Intelligence - FP in Robotics - FP and Music - Advanced FP for undergraduates - Tools supporting learning FP - FP in graduate education - Engaging students in research using FP - FP in Programming Languages - FP in the high school curriculum - FP as a stepping stone to other CS topics - FP and Philosophy *** Best Lectures *** In addition to papers, we request "best lecture" presentations. What is your best lecture topic in an FP related course? Do you have a fun way to present FP concepts to novices or perhaps an especially interesting presentation of a difficult topic? In either case, please consider sharing it. Best lecture topics will be selected for presentation based on a short abstract describing the lecture and its interest to TFPIE attendees. *** Submission *** Papers and abstracts can be submitted via EasyChair at the following link: https://easychair.org/conferences/?conf=tfpie2016 It is expected at at least one author for each submitted paper will attend the workshop. *** Registration & Local Information *** Please see the TFP site for registration and local information: http://tfp2016.org/ *** Important Dates *** April 27, 2016: Submission deadline for draft TFPIE papers and abstracts May 3, 2016: Notification of acceptance for presentation May 13, 2016: Registration for TFP/TFPIE closes June 7, 2016: Presentations in Maryland, USA July 7, 2016: Full papers for EPTCS proceedings due. September 1, 2016: Notification of acceptance for proceedings September 22, 2016: Camera ready copy due for EPTCS Submission of an abstract implies no obligation to submit a full version; abstracts with no corresponding full versions by the full paper deadline will be considered as withdrawn.

1 0