September 2007 - Glasgow-haskell-users

module containing GADTs no longer compiles in ghc 6.8.0
by Daniel Gorín 28 Sep '07

28 Sep '07

Hi I just tried to compile a project of mine that builds fine using ghc 6.6.1 and got many errors like this: src/HyLo/Formula/NNF.hs:247:48: GADT pattern match in non-rigid context for `Opaque' Tell GHC HQ if you'd like this to unify the context In the pattern: Opaque f' In the expression: \ (Opaque f') -> Opaque (Box r f') In the definition of `box': box = \ (Opaque f') -> Opaque (Box r f') I don't know what a "non-rigid context" is, nor if "I like this to unify the context" or not, but I would certainly be happy if I could get this module to compile again! :) For the record, I was using ghc-6.8.0.20070917. Please let me know if you need further information.... Thanks Daniel PS: On a side note, I found this error message to be kind of funny. It seems to indicate no real error but some sort of error-message-driven poll!

3 3

type families + GADT = type unsafety?
by Roberto Zunino 28 Sep '07

28 Sep '07

(Trac reports "database locked", posting here...) Here's unsafeCoerce: > type family Const a > type instance Const a = () > > data T a where T :: a -> T (Const a) > > coerce :: forall a b . a -> b > coerce x = case T x :: T (Const b) of > T y -> y And this indeed "works"... Here's the result with the latest RC: *TypeFam> coerce () 2 <interactive>: internal error: stg_ap_v_ret (GHC version 6.9.20070918 for x86_64_unknown_linux) Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug Aborted Regards, Zun.

6 6

Newbie problem: lexical error when compiling
by David Schonberger 27 Sep '07

27 Sep '07

Hi, I'm new to GHC and am having a problem compiling a simple test program. I recently installed ghc 6.6.1 on my Windows Vista machine. I can ghci ok. For instance, I defined the factorial function and tested it on a few values. So the interpreter seems to work fine. As a next step I created a file called hello.hs (UTF-8 encoding) containing the single line: main = putStrLn "Hello world!" I opened a windows console, navigated to the folder containing this Haskell file and typed 'ghc -o hello hello.hs' and received the following error: hello.hs:1:0 lexical error at character '\65279'. If I type 'ghc -o hello.exe hello.hs' I get the same error. Anybody know what the problem is here and how to fix it? Thanks. Regards, David

3 3

Generics in 6.8.0 RC
by Dirk Kleeblatt 27 Sep '07

27 Sep '07

Dear *, I just failed to get generic classes working with snapshot ghc-6.8.0.20070920-i386-unknown-linux.tar.bz2. Are generics broken in the RC, or am I doing something wrong? I tried the following module: --- module Gen where import Data.Generics class Equal a where equal :: a -> a -> Bool equal {| Unit |} Unit Unit = True equal {| u :+: v |} (Inl x) (Inl y) = equal x y equal {| u :+: v |} (Inr x) (Inr y) = equal x y equal {| u :+: v |} _ _ = False equal {| u :*: v |} (x1 :*: x2) (y1 :*: y2) = equal x1 y1 && equal x2 y2 instance Equal Bool --- and got --- klee@romulus:~/tmp> ghci -fglasgow-exts -XGenerics GHCi, version 6.8.0.20070920: http://www.haskell.org/ghc/ :? for help Loading package base ... linking ... done. Prelude> :l Gen.hs [1 of 1] Compiling Gen ( Gen.hs, interpreted ) Gen.hs:13:0: Can't find interface-file declaration for variable GHC.Base.$gfromBool Probable cause: bug in .hi-boot file, or inconsistent .hi file Use -ddump-if-trace to get an idea of which file caused the error In the first argument of `a', namely `(GHC.Base.$gfromBool b)' In the first argument of `(\ a -> a)', namely `(a (GHC.Base.$gfromBool b))' In the expression: (\ a -> a) (a (GHC.Base.$gfromBool b)) Failed, modules loaded: none. --- Using -ddump-if-trace as suggested gave me no idea of anything, ;-) but if you're interested, I can post the dump, too. Any hints appreciated! Kind regards, Dirk

3 2

where is the full stack trace assigned to its id
by Peter Hercek 27 Sep '07

27 Sep '07

Hi, GHC version: win32 6.6.1 App was compiled with ghc -make -prof -auto-all <files> ... and run with app.exe <appSpecificOptions> +RTS -hc The result is that app.exe.hp contains the sample items in the format like (651)/createAndRunWindows/... 180 The question is where can I find the full stack trace corresponding to 651 identification number? I expected it to be in app.exe.prof (as it is with -hr option), but that file is empty. Do I miss some other RTS option? Peter.

2 2

OSThreads in the RTS
by Cristian Perfumo 26 Sep '07

26 Sep '07

Hi all, I am diving into the RTS, trying to identify where an OSThread is created and where it is destroyed (join). The idea is to read some hardware counters when it is just created and read them again when it is about to be destroyed (for the time being let's ignore, for the sake of understanding, the possibility of thread migration and context switches). Does anyone know where to place these reads? As shown in the code below, I tried "printfing" within "workerStart" (in the places where I thought I should read the counters), but it doesn't seem to print the same throughout several executions. I mean, running with -N2 and forking 2 Haskell threads (plus the main one), sometimes I get three "Scheduling" and two "Unscheduling" and sometimes I get only one "Unscheduling" which discourages me about reading the counters there. Any clue about the reason? Probably "workerStart" is not the place to put it, I don't know. Any light thrown on it or any documentation to read would be welcome. Cheers Cristian Perfumo void workerStart(Task *task) { Capability *cap; // See startWorkerTask(). ACQUIRE_LOCK(&task->lock); cap = task->cap; RELEASE_LOCK(&task->lock); // set the thread-local pointer to the Task: taskEnter(task); printf ("Scheduling %p\n", task); // schedule() runs without a lock. cap = schedule(cap,task); printf ("Unscheduling %p\n", task); // On exit from schedule(), we have a Capability. releaseCapability(cap); workerTaskStop(task); }

2 1

Re[2]: Hackage and GHC 6.8
by Bulat Ziganshin 24 Sep '07

24 Sep '07

Hello Thomas, Monday, September 24, 2007, 6:42:11 PM, you wrote: > This is not a pretty solution and also has the problem of incorrectly > dealing with changes to the base interface. This, IMO, is due to > keeping the name 'base' instead of opting to clarify the > incompatibilities by choosing "base2". it's not yet too late to fix it. i think that this Base2 Hell is serious problem and we need to develop and introduce solution to it before we can issue GHC 6.8.1 from my POV, the best solution is to provide base 1.0.* package that just reexports all the functionality that was available in ghc 6.6; and base 2.0 package - the current one -- Best regards, Bulat mailto:Bulat.Ziganshin@gmail.com

1 0

Re: [Haskell-cafe] GHC 6.7 on Windows / containers-0.1 package?
by Victor Nazarov 21 Sep '07

21 Sep '07

> Is this still up-to-date with the way GHC/GHCi work internally? Then > I'll certainly check it out. I think it isn't since STG is pretty much different with the G-machine describes in the book. The original paper about Spineless Tagless G-machine is here [1]. It is quite verbose. Recent paper "How to make a fast curry: push/enter vs eval/apply" [2] is much smaller, but it represents more recent state of GHC and can be read independently with the first paper. The book from 1987 [3] is a very good and detailed overview of the subject and it is valuable as itself, no matter how it actually implemented now. -- 1. https://research.microsoft.com/users/simonpj/papers/spineless-tagless-gmach… 2. https://research.microsoft.com/users/simonpj/Papers/eval-apply/index.htm 3. https://research.microsoft.com/users/simonpj/Papers/slpj-book-1987/index.htm

1 0

Additional thunk for function
by Victor Nazarov 21 Sep '07

21 Sep '07

What is the purpose for GHC to allocate a thunk for some functions? Why Test.map is not a function, but updatable thunk, which should become equal to the function just after the first call? Here is the details: > % ghc -c test.hs -ddump-stg > > ==================== STG syntax: ==================== > Test.map = > \u [] > let { > map1_sdR = > \r [f_sdN ds_sdI] > case ds_sdI of wild_sdU { > [] -> [] []; > : x_sdM xs_sdQ -> > let { sat_sdT = \u [] map1_sdR f_sdN xs_sdQ; } in > let { sat_sdP = \u [] f_sdN x_sdM; } in : [sat_sdP sat_sdT]; > }; > } in map1_sdR; > SRT(Test.map): [] > > > % cat test.hs > module Test where > > map f [] = [] > map f (x:xs) = f x : Test.map f xs -- Thanks in advance Victor

3 3

Additional thunk for function
by Victor Nazarov 21 Sep '07

21 Sep '07

I have reproduced it on 2 Windows machines with ghc 6.6 and 6.6.1 Here is the version information for ghc used to generate output in original message. % ghc -v Glasgow Haskell Compiler, Version 6.6, for Haskell 98, compiled by GHC version 6.6 Using package config file: C:\Program Files\Visual Haskell\package.conf wired-in package base mapped to base-2.0 wired-in package rts mapped to rts-1.0 wired-in package haskell98 mapped to haskell98-1.0 wired-in package template-haskell mapped to template-haskell-2.0 Hsc static flags: -static -- vir http://vir.comtv.ru/

1 0