October 2006 - Glasgow-haskell-users

GHC 6.6 using stale stub.c files
by Juan Carlos Arevalo Baeza 24 Oct '06

24 Oct '06

I tried recompiling a little (big?) Win32 binding library I had made, when the official one wasn't working for me. It had a bunch of generated stubs files in it (I'm not 100% sure what they are for). The problem is that, not even with -fforce-recomp (much better name, thanx!) will GHC regenerate those files (I think that's normal), but still it tries to use them if they are there. I get errors like these: Win32Test_stub.c: In function `Main_d4n0': Win32Test_stub.c:14:0: error: too few arguments to function `rts_mkPtr' Win32Test_stub.c:14:0: warning: passing arg 1 of `rts_apply' from incompatible pointer type Win32Test_stub.c:14:0: error: too few arguments to function `rts_apply' Win32Test_stub.c:14:0: warning: passing arg 1 of `rts_mkWord32' makes pointer from integer without a cast (etc, etc... lots of them). The stub file is old: C:\Documents and Settings\JCAB\My Documents\Haskell>dir Win32Test_stub.c Volume in drive C has no label. Volume Serial Number is A0CD-B8F4 Directory of C:\Documents and Settings\JCAB\My Documents\Haskell 08/18/2005 12:15 AM 637 Win32Test_stub.c 1 File(s) 637 bytes 0 Dir(s) 151,815,680,000 bytes free And yes, I'm on Windows. There are some stubs files that were regenerated: C:\Documents and Settings\JCAB\My Documents\Haskell>dir Win32\Window\Types_stub.c Volume in drive C has no label. Volume Serial Number is A0CD-B8F4 Directory of C:\Documents and Settings\JCAB\My Documents\Haskell\Win32\Window 10/22/2006 05:05 PM 717 Types_stub.c 1 File(s) 717 bytes 0 Dir(s) 151,803,375,616 bytes free So I suppose the no-longer-needed stubs files were related to the modules that made use of this Win32.Window.Types module I had here. I guess GHC doesn't need those any more, but it just tries to use any stub files it finds. It probably shouldn't use them, and it should probably even delete them, unless it generated them, at least when -fforce-recomp is specified. I'm moving on. I made a copy of all the files, so let me know if you need a repro. Deleting the stubs files fixes the problem. I haven't seen anything about this in the documentation, but maybe I just didn't look in the right places. Thanx! JCAB

2 4

Debugging on Mac OS X
by Luke Worth 24 Oct '06

24 Oct '06

Hi all, I just installed GHC 6.6 using the PPC binary installer on Mac OS X. I'm having some trouble getting debuggers to work; I have tried both plargleflarp (buddha) and Hat. Neither of them will compile on GHC 6.6 on Mac yet. For anyone in the same situation, how do you survive? I know of no other method of debugging Haskell programs, aside from manually running chunks of the code from within GHCi - but that makes it very difficult to debug nested 'where' functions. Thanks in advance Luke Worth

3 2

Heap representation evil
by Brandon Moore 24 Oct '06

24 Oct '06

Thinking to take advantage of fortuitous heap layout of some Haskell values for interfacing with C, I've written the following function: addressOf :: a -> Ptr () addressOf x = x `seq` unsafeCoerce# (Box x) data Box x = Box x For example, data A = A {-# UNPACK #-} !(Ptr Word8) {-# UNPACK #-} !CInt main = let a = A nullPtr 12 p = addressOf a `plusPtr` 4 in do x <- peek p :: IO Int y <- peek p :: IO Int print (x, y) prints (0, 12) One thing I don't understand is that this fails if I use Just rather than inventing my box type. I suppose the info table for Just is set up to support a vectored return for pattern matching on Maybe? (the commentary isn't very clear here. The section http://hackage.haskell.org/trac/ghc/wiki/Commentary/Rts/HaskellExecution#Re… says, in full: "Return Convention Direct Returns Vectored Returns" ) The reason I'm messing about with this stuff is that I'm pretty sure passing p to C code would give a usable pointer to struct a {char *; int;}; Obviously my plot will be spoiled if the GC comes along and relocates the value while C code is trying to use it. Are there any other pitfalls with this approach? A different and in all likelihood saner approach is building up more tools for manipulating pointers to C data from Haskell, perhaps along the lines of cmucl's support for "Alien Objects". http://www.math.sunysb.edu/~sorin/online-docs/cmucl/aliens.html The main reason to even think about touching the heap representation of Haskell objects is so that the values can be manipulated by pure code, pattern matched, etc. Brandon

3 4

Benchmarking GHC
by Neil Mitchell 23 Oct '06

23 Oct '06

Hi, I want to benchmark GHC vs some other Haskell compilers, what flags should I use? I just want whichever package of flags is "good optimisation" - nothing unsafe, nothing that changes strictness properties, and a set of flags that is generally applicable - but slow compilation is fine. As far as the user guide says: * At the moment, -O2 is unlikely to produce better code than -O. * When we want to go for broke, we tend to use -O2 -fvia-C >From this I guess the answer is "-O2 -fvia-C"? I just wanted to check this information is still current before I start giving anyone any benchmark numbers. (for reference I'm on Windows, x86 Pentium 4/Athlon, if that makes a difference) Thanks Neil

10 17

ghc error
by Chad Scherrer 23 Oct '06

23 Oct '06

I tried compiling a file today using -O2, and it gave me this: /tmp/ghc10376_0/ghc10376_0.hc:5:20: HsTime.h: No such file or directory It compiled fine without the -O2. Am I missing some files? Where can I find what I need to make this work again? Thanks, -- Chad Scherrer "Time flies like an arrow; fruit flies like a banana" -- Groucho Marx

1 0

Concurrency issue with dynamic linker in GHC
by Vyacheslav Akhmechet 23 Oct '06

23 Oct '06

I am running ghc 6.4.2 on a Win32 machine. I'm using hs-plugins in one thread and a simple getLine loop in another. It appears that getLine blocks the hs-plugins thread on Win32 (this has been verified to work fine on freeBSD). I've tried various combinations of -threaded flag and forkIO/forkOS and always get the undesireable result. Below is the minimal test case. module Main where import Control.Concurrent import System.Plugins import System.IO -- Main loop main = do hSetBuffering stdout NoBuffering forkIO blah test <- getLine putStrLn test blah = do contents <- loadPlugin "Hello" putStrLn contents loadPlugin path = do status <- pdynload (path ++ ".o") [] [] "Prelude.String" "myTestSym" case status of (LoadSuccess _ res) -> return res (LoadFailure errors) -> return $ concat errors --------------------- Hello.hs -------------------- module Hello where myTestSym :: String myTestSym = "Hello 2!" Thanks!

4 17

Re: Timing Stats in ghci-6.6
by Daniel Fischer 23 Oct '06

23 Oct '06

Repost, seems to have been lost last week: Am Dienstag, 17. Oktober 2006 12:35 schrieb Simon PJ: > What timing? Can you be more explicit pls? > Just plain old :set +s in ghci: dafis@linux:~/Downloads/Euler> ghci Euler44 ___ ___ _ / _ \ /\ /\/ __(_) / /_\// /_/ / / | | GHC Interactive, version 6.6, for Haskell 98. / /_\\/ __ / /___| | http://www.haskell.org/ghc/ \____/\/ /_/\____/|_| Type :? for help. Loading package base ... linking ... done. [1 of 3] Compiling Primes ( Primes.hs, interpreted ) [2 of 3] Compiling ContFrac ( ContFrac.hs, interpreted ) [3 of 3] Compiling Euler44 ( Euler44.hs, interpreted ) Ok, modules loaded: Euler44, ContFrac, Primes. *Euler44> :set +s *Euler44> smallest (2167,1020,2395,1912) (0.02 secs, 910618996 bytes) *Euler44> Actually took 23 secs. Built it yesterday from source bundle on linux. Oh, and by the way, I've never managed to get the users-guide built. Is that intentional or would that just require some xml-processing functionality I haven't installed (if so: which)? Cheers, Daniel > Simon > > | -----Original Message----- > | From: glasgow-haskell-users-bounces(a)haskell.org > > [mailto:glasgow-haskell-users- > > | bounces(a)haskell.org] On Behalf Of Daniel Fischer > | Sent: 17 October 2006 10:04 > | To: GHC Users > | Subject: Timing Stats in ghci-6.6 > | > | Hi, > | > | apparently the timing doesn't work properly in ghci 6.6 (at least for > > me), I > > | always get very small timings, usually 0.01 secs, sometimes 0.00 secs, > > once > > | 0.07 secs, no matter how long it actually took. > | > | Is that a common problem or is it only me? > | > | Cheers, > | Daniel > | _______________________________________________ > | Glasgow-haskell-users mailing list > | Glasgow-haskell-users(a)haskell.org > | http://www.haskell.org/mailman/listinfo/glasgow-haskell-users

1 0

getting the -B topdir for a GHC-API app
by Mathew Mills 22 Oct '06

22 Oct '06

What is the "right" way to get the topDir for use with the GHC API. I have a little app that I would like to cabalize and release, but I am not sure what is the approved way of getting the '-B' option set-up. I think it would be great if I could use template-haskell to insert the topDir from the ghc used to compile the app, but I am not sure how to get the topDir from the compiler. Is this possible? Does cabal already provide a means to install a "driver" script like that used for the ghc suite? Also, I am looking for a code-review. If you have some time, I would love some "style" guidance. The work in progress can be found here: http://mathewmills.com/runTests/runTests.lhs

2 1

Installing a library
by jim burton 21 Oct '06

21 Oct '06

A very basic question re using GHC 6.4.1 on ubuntu linux - how do I go about installing a library? I want to install HUnit and suspected that /usr/lib/haskell-packages/ghc6/lib would be the place to untar it (into the dir HUnit-1.0), but that doesn't work. I was reluctant to move the contents of HUnit-1.0 directly into ...ghc6/lib because it includes examples and docs as well as the module files. I notice that MissingH, which I've installed, is in that dir but inside it's top level there are files like HSMissingH-0.13.1.o, installed-pkg-config, libHSMissingH-0.13.1.a ... does that mean that MissingH is a Package (something which I don't understand to be honest) and HUnit isn't? How do I tell what the load path of GHC is and add something to it? Where should I unpack HUnit for it to be accessible? Thanks, -- View this message in context: http://www.nabble.com/Installing-a-library-tf2483332.html#a6924826 Sent from the Haskell - Glasgow-haskell-users mailing list archive at Nabble.com.

2 1

pinned byte arrays
by Bulat Ziganshin 20 Oct '06

20 Oct '06

Hello glasgow-haskell-users, i have program that holds in memory a lot of strings (filenames) and use my own packed-string library to represent these string. this library uses newByteArray# to allocate strings. in my benchmark run (with 300.000 filenames in memory) program prints the following statistics: 37,502,956 bytes maximum residency (16 sample(s)) 2928 collections in generation 0 ( 6.90s) 16 collections in generation 1 ( 12.20s) 65 Mb total memory in use most of this memory occupied by filenames. note that 28 mb of memory is used just for GC procedure (i use compacting GC) then, i thought that using pinned byte arrays should significantly improve memory usage because these arrays can't be moved around and therefore, i thought, these arrays will not be involved in GC. amount of data which should be compacted by GC will decrease and amount of memory wasted by GC should also decrease so i replaced in my packed str lib newByteArray# with newPinnedByteArray# and seen the following: 53,629,344 bytes maximum residency (14 sample(s)) 2904 collections in generation 0 ( 6.97s) 14 collections in generation 1 ( 13.16s) 100 Mb total memory in use why? why memory usage was increased? why another 47 megs is just wasted out? how GC works with pinned byte arrays? and why GC times was almost the same - i thought that GC for pinned data should be very different than GC for unpinned byte arrays -- Best regards, Bulat mailto:Bulat.Ziganshin@gmail.com

3 2