- Haskell-Cafe - Haskell.org

Re: [Haskell-cafe] Re: SQL Database in Haskell?
by Max Desyatov 08 Aug '09

08 Aug '09

As I can say from my experience of usage of hdbc-sqlite3 and happstack-state, the latter covers everything you ever wanted from sqlite3 and more. It you aren't too concerned about performance, you can free yourself from many tedious routines that are imminent when you work with relational database. Elaborated data model design coupled with some generics technique (uniplate with derive, e.g.) gives you a possibility to write down your domain problem directly to haskell. CK Kashyap <ck_kashyap(a)yahoo.com> writes: > I'd be very interested to see a rdbms implementation in Haskell ... > perhaps a port of sqlite > > Regards, > Kashyap

2 1

Linking failing due to Control.Monad.State.Strict?
by phil＠beadling.co.uk 07 Aug '09

07 Aug '09

Hi, I'm getting a linker error when generating makefile dependencies on the below library for a program I'm using. It links fine under 'ghc -- make'. I'm using (more or less) the makefile from the haskell docs. I am using the Strict State Monad in the object files it is complaining about, the compile is fine, just the linker which is blowing up. Any ideas what is causing this? I'm using GHC 6.10.4 on PPC Mac OS X 10.5. I've included the makefile below the error. Cheers! Phil. ghc -o OptionCalculator -O2 -Wall ./FrameworkInterface.o ./Maths/ Prime.o ./Misc/Debug.o ./MonteCarlo/DataStructures.o ./MonteCarlo/ European.o ./MonteCarlo/Framework.o ./MonteCarlo/Interface.o ./ MonteCarlo/Lookback.o ./Normal/Acklam.o ./Normal/BoxMuller.o ./Normal/ Framework.o ./Normal/Interface.o ./OptionCalculator.o ./Random/ Framework.o ./Random/Halton.o ./Random/Interface.o ./Random/Ranq1.o Undefined symbols: "_mtlzm1zi1zi0zi2_ControlziMonadziStateziStrict_a29_info", referenced from: _r1hl_info in Acklam.o _rGL_info in BoxMuller.o "_mtlzm1zi1zi0zi2_ControlziMonadziStateziStrict_a29_closure", referenced from: _r1hl_srt in Acklam.o _rGL_srt in BoxMuller.o "___stginit_mtlzm1zi1zi0zi2_ControlziMonadziStateziStrict_", referenced from: ___stginit_FrameworkInterface_ in FrameworkInterface.o ___stginit_FrameworkInterface_ in FrameworkInterface.o ___stginit_MonteCarloziEuropean_ in European.o ___stginit_MonteCarloziEuropean_ in European.o ___stginit_MonteCarloziFramework_ in Framework.o ___stginit_MonteCarloziFramework_ in Framework.o ___stginit_MonteCarloziLookback_ in Lookback.o ___stginit_MonteCarloziLookback_ in Lookback.o ___stginit_NormalziAcklam_ in Acklam.o ___stginit_NormalziAcklam_ in Acklam.o ___stginit_NormalziBoxMuller_ in BoxMuller.o ___stginit_NormalziBoxMuller_ in BoxMuller.o ___stginit_NormalziFramework_ in Framework.o ___stginit_NormalziFramework_ in Framework.o ___stginit_RandomziFramework_ in Framework.o ___stginit_RandomziFramework_ in Framework.o ___stginit_RandomziHalton_ in Halton.o ___stginit_RandomziHalton_ in Halton.o ___stginit_RandomziRanq1_ in Ranq1.o ___stginit_RandomziRanq1_ in Ranq1.o ld: symbol(s) not found collect2: ld returned 1 exit status make: *** [OptionCalculator] Error 1 Makefile: HC = ghc HC_OPTS = -O2 -Wall $(EXTRA_HC_OPTS) SRCS := $(shell find . -name "*.hs" -print) OBJS = $(SRCS:.hs=.o) PROG = OptionCalculator .SUFFIXES : .o .hs .hi .lhs .hc .s ${PROG} : $(OBJS) rm -f $@ $(HC) -o $@ $(HC_OPTS) $(OBJS) # Standard suffix rules .o.hi: @: .lhs.o: $(HC) -c $< $(HC_OPTS) .hs.o: $(HC) -c $< $(HC_OPTS) .o-boot.hi-boot: @: .lhs-boot.o-boot: $(HC) -c $< $(HC_OPTS) .hs-boot.o-boot: $(HC) -c $< $(HC_OPTS) clean : find . -name "*.hi" -exec rm -f {} \; find . -name "*.o" -exec rm -f {} \; rm -f ${PROG} depend : ghc -M $(HC_OPTS) $(SRCS)

3 5

Re: [Haskell-cafe] Basic questions about concurrency in Haskell
by Sebastian Sylvan 07 Aug '09

07 Aug '09

Bringing the cafe back in. If I remember correctly tuning the GC is one of the things they worked on for the next release (in relation to parallelism).Here's a link to the paper: http://research.microsoft.com/en-us/um/people/simonpj/papers/parallel/multi… You can allocate dynamic memory on the C heap using the Foreign.Marshal.Alloc stuff. The Storable class allows you to specify alignment. However it's not clear you will need it in all cases. Most of the time IME all you really need is for "the business end" of object A and B to end up on different cache lines, padding them with the size of a cache line will accomplish that even if each object starts in the middle of a cache line. In other words, as long as you ensure that there's at least CACHE_LINE bytes difference between the last bit of data in A and the first bit of data in B, they won't exhibit false sharing. I'm not sure if any of the "standard" mutable references etc. respect this. Maybe someone can clarify? If you modify a thread-local IORef will that cause another thread to stall because it has it's own thread-local IORef on the same cache line? On Fri, Aug 7, 2009 at 12:42 PM, Thomas Witzel <witzel.thomas(a)gmail.com>wrote: > I have not yet been able to get the dev version compiled, but I'm > looking into it (basically compiles to 99% and then stops because of a > missing or defective .mk file). Anyhow, I played more with the release > version and one thing I noticed is that the GC time increases almost > linearly with the number of cores. Is there somewhere a document that > explains the overall architecture of the GHC/RTS and how in especially > the GC acts in concurrent situations ? > > As for your comment regarding the padding the data-structures, I'll of > course also need to control the alignment in such a case. Is there > such a thing as explicit dynamic memory allocation in Haskell ? > > Thanks, Thomas > > On Wed, Aug 5, 2009 at 3:16 PM, Sebastian > Sylvan<sebastian.sylvan(a)gmail.com> wrote: > > GHC doesn't have per-thread allocation so it's probably a bit tricky to > get > > that working. Plus, for parallelism it's not clear that a piece of data > is > > necessarily "owned" by one thead, since it could be produced by a spark > and > > consumed by another spark, those two independent sparks may not > necessarily > > occupy the same thread, which means that any *other* data accessed by the > > firs thread could thrash the cache. So really you'd need per-spark > > allocation areas which would probably make sparks very heavy weight. > > In other words, I think there's plenty of research that needs to be done > > w.r.t. scheduling things in time and space so as to avoid false sharing. > You > > could, of course, always chunk your work manually, and make sure that > each > > "chunk" works on a big block that won't share cache lines with anything > else > > (e.g. by padding the data structures). > > Also, while GHC does a fair bit of mutation on its own internal data > (thunks > > etc.), most of the "user data" is read-only, which should help. I.e. once > a > > cache line has been filled up, there won't be any synchronisation needed > on > > that data again. > > > > On Wed, Aug 5, 2009 at 8:04 PM, Thomas Witzel <witzel.thomas(a)gmail.com> > > wrote: > >> > >> I'll try that. I'd like to stick with it. As for the memory, although > >> its probably quite a bit of work, it should be doable to have code > >> generated where the threads have their own, non-overlapping, memory > >> pages, so that the CPUs don't go into a cache-thrashing death-match. > >> I'll spend some more time with Haskell and then go from there. > >> > >> On Wed, Aug 5, 2009 at 3:01 PM, Sebastian > >> Sylvan<sebastian.sylvan(a)gmail.com> wrote: > >> > > >> > > >> > On Wed, Aug 5, 2009 at 6:59 PM, Thomas Witzel < > witzel.thomas(a)gmail.com> > >> > wrote: > >> >> > >> >> 2. I started with the very simple nfib example given in the manual > for > >> >> Control.Parallel (Section 7.18). On my systems using multiple cores > >> >> makes the code actually slower than just using a single core. While > >> >> the manual cautions that this could be the case for certain > >> >> algorithms, I'm wondering whether this is the desired behaviour for > >> >> this example. > >> >> > >> >> I'm using ghc 6.10.4 right now. > >> > > >> > IIRC the development version of GHC has some major work to optimize > >> > concurrency, so it may be worth trying that. In particular I believe > it > >> > executes sparks in batches, to reduce the overhead (which hopefully > >> > fixes > >> > your issue). > >> > > >> > -- > >> > Sebastian Sylvan > >> > > > > > > > > > -- > > Sebastian Sylvan > > > -- Sebastian Sylvan

1 0

FFI: Problem with Signal Handler Interruptions
by Levi Greenspan 07 Aug '09

07 Aug '09

Dear list members, In February this year there was a posting "Why does sleep not work?" (http://www.haskell.org/pipermail/haskell-cafe/2009-February/055400.html) The problem was apparently caused by signal handler interruptions. I noticed the same (not with sleep though) when doing some FFI work and compiled the following test program: {-# LANGUAGE ForeignFunctionInterface #-} module Main where import Foreign.C.Types import Control.Concurrent sleep :: IO () sleep = c_sleep 3 >>= print fails :: IO () fails = sleep works :: IO () works = forkIO sleep >> return () main :: IO () main = fails >> works >> threadDelay 3000000 foreign import ccall unsafe "unistd.h sleep" c_sleep :: CUInt -> IO CUInt When compiled with GHC (using --make -threaded), it will print 3 immediately (from the "fails" function) and after 3 seconds 0 (from "works"), before it finally exits. man sleep(3) tells me that sleep returns 0 on success and if interrupted by a signal the number of seconds left to sleep. Clearly "fails" is interrupted by a signal (which seems to be SIGVTALRM). This was mentioned in the discussion from February. I would like to know why "fails" fails and "works" works, i.e. why is "sleep" not interrupted when run in a separate thread? And what can be done to make "sleep" work in the main thread? It wouldn't be wise to block SIGVTALRM, wouldn't it? Many thanks, Levi

2 4

Seeking for an extention (functional incapsulation)
by Belka 07 Aug '09

07 Aug '09

Hello, cafe visitors! :) This is a double topic: 1. Can't find any good informative resource with descriptions of Haskell extensions. Could anybody please share good one if it exists? The only "good" one I found: http://hackage.haskell.org/trac/haskell-prime/wiki/HaskellExtensions But it's a bit too old and not that full... I undestand, that Haskell is kind of "boiling" language, in a sense of being neverending experiment. It develops all the time, extensions show up and drop out. So it's not that easy to support community with a fresh information about them. But on the other side, the property (of being "boiling" language) makes such information really important for community members... I think. :) 2. Consider situation: ----------------------------------- data SomeDataType = SomeDataType { sdtField1 :: SDT_Field1Type, sdtField2 :: SDT_Field2Type} sdtField3 :: SomeDataType -> SDT_Field3Type sdtField3 sdt = f (sdtField1 sdt) (sdtField2 sdt) ----------------------------------- I induced recently, that it would be very comfortable if I could perform in a way like this: ----------------------------------- data SomeDataType = SomeDataType { sdtField1 :: SDT_Field1Type, sdtField2 :: SDT_Field2Type, sdtField3 :: SDT_Field2Type, sdtField3 = f sdtField1 sdtField2} ----------------------------------- The situation is not that rare, when dealing with nonprimitive data constructions. Moreover would be really comfortable to reduce ----------------------------------- data SomeDataType = SomeDataType_111 { sdtField1 :: SDT_Field1Type, sdtField2 :: SDT_Field2Type} | SomeDataType_222 { sdtField1 :: SDT_Field1Type, sdtField2 :: SDT_Field2Type, sdtField5 :: SDT_Field5Type} sdtField3 :: SomeDataType -> SDT_Field3Type sdtField3 sdt = case sdt of {SomeDataType_111 -> f (sdtField1 sdt) (sdtField2 sdt) ; SomeDataType_222 -> g (sdtField1 sdt) (sdtField2 sdt) (sdtField5 sdt)} \/ \/ \/ \/ \/ \/ \/ \/ \/ \/ data SomeDataType = SomeDataType_111 { sdtField1 :: SDT_Field1Type, sdtField2 :: SDT_Field2Type, sdtField3 :: SDT_Field3Type, sdtField3 = f sdtField1 sdtField2} | SomeDataType_222 { sdtField1 :: SDT_Field1Type, sdtField2 :: SDT_Field2Type, sdtField5 :: SDT_Field5Type, sdtField3 :: SDT_Field3Type, sdtField3 = g sdtField1 sdtField2 sdtField5} ----------------------------------- Usable mechanics for realization would be: 1. Funtion similar to Data.Function.on (example: (*) `on` f = \x y -> f x * f y), but opposite - I called it under. t `under` f = \x y -> (x f) `t` (y f) 2. currying and uncurrying Is there any such extension? Belka -- View this message in context: http://www.nabble.com/Seeking-for-an-extention-%28functional-incapsulation%… Sent from the Haskell - Haskell-Cafe mailing list archive at Nabble.com.

2 6

Looking to check on some capabilities of Data.Colour
by Jeff Heard 06 Aug '09

06 Aug '09

I was wondering if Data.Colour supported Double-valued colour components > 1.0 or less than 0. I'm looking to create an HDR image processing library, and Haskell has one of the most extensive and correct colour models around, thanks to Russell. With 16bpcc or 32bpcc images, however, I need to be sure to be able to correctly calculate colour values that fall outside the usual [0.0,1.0] gamut. Does Data.Colour support this functionality? -- Jeff

2 3

A mistake in haskellwiki
by L.Guo 06 Aug '09

06 Aug '09

Hi haskellers: There is a mistake in http://www.haskell.org/haskellwiki/State_Monad It post two functions like this : evalState :: State s a -> s -> a evalState act = fst $ runState act execState :: State s a -> s -> s execState act = snd $ runState act Both the '$' operators should be '.'. Anyone would correct it ? Regards -------------- L.Guo 2009-08-06

7 7

n00b question: defining datatype
by Iain Barnett 06 Aug '09

06 Aug '09

Hi, I'm trying to get my head around datatypes, and wondering how I might define a simple "Task" datatype in Haskell. data Task = Task { title :: String, completed :: Bool } Ok, that's straightforward, but sometimes tasks become a list of tasks themselves data Task = Task { title :: String, completed :: Bool, subtasks :: [Task] } But that's not really right, because obviously, some tasks don't have subtasks. So I try this: data Task = Task { title :: String, completed :: Bool } | TaskWithSubtasks { title :: String, completed :: Bool, subtasks :: [Task] } It's a bit more accurate, but it's repeating things, which is ok with a simple type. Could anyone suggest a better way to define this? If I was using C#, which I'm far more familiar with, I could overload the constructor and refer to the smaller constructor. Is there a way to do that in Haskell, or am I still thinking too OOP? Iain

8 9

SQL Database in Haskell?
by Günther Schmidt 06 Aug '09

06 Aug '09

Hi all, is there an SQL Database in Haskell or is there a project trying to implement one? Günther

5 8

Do you understand posix well?
by Maurício CA 06 Aug '09

06 Aug '09

I've beeing writing a low-level binding to posix that can be usefull if you want to use posix but has no time to learn FFI: http://hackage.haskell.org/package/bindings-posix However, my understandment of posix is barely nothing, and I see that many of its functionality is enabled or disabled by macros, and I don't know which ones are related to what functionality. So, if you know about that: would you be able to list all macros (or at least most important ones, if there are too many) that enable all (or most) of posix? In exchange, you get a posix binding that is actually comprehensive :) Thanks, Maurício

1 0