- Glasgow-haskell-users

ghci module load path
by Joel Reymont 30 Apr '07

30 Apr '07

Folks, Suppose I have a cabalized project where the source is under src/ and the generated code goes into dist/build/test/test-tmp/. How can I load modules from the generated code directory when I'm in the parent of both src and dist? If I do :set -idist/build/test/test-tmp then ghci tries to look for source files there but they are in src. If I :set -isrc then ghci doesn't look for the already compiled code and just interprets the source files. I'm looking to load the pre-compiled files. Thanks, Joel -- http://wagerlabs.com/

2 3

Re: Release plans
by David Waern 30 Apr '07

30 Apr '07

> Here's a quick summary of the major developments that we already have in > the 6.8 codebase: > > - Associated data types, and the new FC intermediate language > - GHCi debugger (although there's an overhaul of the breakpoint support > almost > ready to go in) > - Coverage (HPC) > - GADTs + typeclasses > - more packages moved from core to extralibs > - GHC API changes, compile to object code inside GHCi > - performance improvements: simplifier rewrite, and SpecConstr improvements > - standalone deriving declarations > - left-to-right impredicative instantiation: runST $ foo It also has the Haddock comment support, so we could do a release of Haddock-GHC. /David, voting for 6.8

2 1

GHC Manuals online
by Neil Mitchell 26 Apr '07

26 Apr '07

Hi, Recently I've been wanting to search for things in the GHC manual, unfortunately that isn't very easy with Google, beacuse lots of people have put up their own copies of the GHC manual - mostly out of date! Try googling for "The Glorious Glasgow Haskell Compilation System User's Guide", I was able to find easily: 6.4 http://web.mit.edu/ghc/www/users_guide/ 3.02 http://www.dcs.gla.ac.uk/fp/software/ghc/3.02/users_guide/users_guide-1.html 6.4.1 http://www.ugcs.caltech.edu/manuals/lang/ghc-6.4.1/users_guide/index.html 4.06 http://www.macs.hw.ac.uk/~dsg/gph/docs/4.06/users_guide/introduction-ghc.ht… The disadvantage of having lots of out of date manuals is that Google rarely finds the correct one, which means when I search for the "Language Pragma GHC" I run into loads of manuals from ages ago. Any chance some GHC person could politely request that some of these get removed? Having the manual easily searchable is a good thing. Thanks Neil

6 8

noinline in where clauses again
by Duncan Coutts 26 Apr '07

26 Apr '07

I winged before about NOINLINE pragma on things defined locally in a where clause not doing what I expected: http://www.mail-archive.com/glasgow-haskell-users@haskell.org/msg10338.html Turns out that one was fixed in ghc HEAD. I've got another similar one which is not working as I expect in 6.6 or in head from 1st March. This example is from an experimental re-implementation of the Put monad in Data.Binary: write :: Int -> (Ptr Word8 -> IO ()) -> Put () write !n body = Put $ \c buf@(Buffer fp o u l) -> if n <= l then write' c fp o u l else write' (flushOld c n fp o u) (newBuffer c n) 0 0 0 where {-# NOINLINE write' #-} write' c !fp !o !u !l = -- warning: this is a tad hardcore B.inlinePerformIO (withForeignPtr fp (\p -> body $! (p `plusPtr` (o+u)))) `seq` c () (Buffer fp o (u+n) (l-n)) {-# INLINE [1] write #-} Then we use it with things like word8 :: Word8 -> Put () word8 !w = write 1 (pokeWord8 w) pokeWord8 :: Word8 -> Ptr Word8 -> IO () pokeWord8 w p = poke p w Then there's a rule so that things like: foo :: Word8 -> Put () foo !n = do word8 n word8 (n+1) word8 (n+17) get turned into a single call to write. Anyway, the point is that when we look at the stg/core we see that in write above, the write' has been inlined at the two call sites where as I want both branches of the if test to make calls to write'. The code is here: http://haskell.org/~duncan/binary/PutMonad.hs http://haskell.org/~duncan/binary/PutTest.hs the stg from ghc-6.6 is: http://haskell.org/~duncan/binary/PutTest.stg with ghc head from 1st March the result is different but still inlines write' in both branches. Duncan

2 2

libgdk_pixbuf-2.0-0.dll Problem
by Mike Matthews 25 Apr '07

25 Apr '07

I wonder if anyone can help me. I have used glade and gtk2hs to create a GUI for my Connect 4 program, using .png images to depict the counters on the screen. Everything was working fine until I tried running my program on another computer. Predictably, there were a number of dll files required, so I copied these into the directory containing my executable (compiled with GHC 6.6). The program then runs correctly, but doesn't display anyof my images. Further investigation showed that the problem seems to be with the libgdk_pixbuf-2.0-0.dll file. Whenever this is in the same folder as my executable (even on my computer) the program displays no images. If anyone can shed any light on this that'd be great. It may be that I'm going about this completely the wrong way. In particular if there's a program out there for easily creating an msi package or something to neatly package the required files please let me know - I haven't been able to find it. Many thanks in advance. Mike

1 0

GHC library profiling
by Pasqualino 'Titto' Assini 25 Apr '07

25 Apr '07

Hi, I have a little library that depends on the 'ghc' api library. I would like to build my library in profiling mode but this is not possible as the 'ghc' library itself seems to be distributed only in non-profiling mode. Is there any way out? How can I get or build a profiling version of 'ghc'? An unrelated comment: I guess that the right command to update the darcs version of ghc plus all its related libraries is: sh darcs-all --extra pull Would it be useful to add this info to the ghc README file? Currently it only explains how to get ghc and libs the very first time Thanks in advance for your help, titto

2 1

Release naming - e.g. "6.8" vs. "6.8.0"
by Isaac Dupree 20 Apr '07

20 Apr '07

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I've gotten confused before between whether GHC "6.6" referred to the whole branch or the point release between 6.4.2 and 6.6.1. It makes it hard to distinguish which is being referred to (although, it also makes it easy to be vague when referring to an upcoming major release, which might be considered an advantage I suppose). I propose we switch to that "X.Y.0"-style major release naming as of 6.8.0 (a.k.a. 6.8). Comments or opinions? Isaac (P.S. GCC's having made this naming change inspired me to suggest this, but I don't actually know what their reasons were) -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFGKOrZHgcxvIWYTTURAtykAJ4s/cSgmcHtGspk1LvNApq+N2SjQQCdFAqH UsNQIR2CXZ6en74MFwNi0HM= =ynkD -----END PGP SIGNATURE-----

1 0

Impressions on GHC debugger
by Alexey Rodriguez Yakushev 19 Apr '07

19 Apr '07

Dear all, Yesterday I darcs-got the GHC HEAD to play with the debugger and I have the following impressions/bug/suggestions: * I would like to know whether I am debugging mode (i.e. in a breakpoint) or not. For example > *Main> -- normal prompt and, for example, > *Main:[1] > -- in breakpoint 1 now, if I evaluate expression "e" via "seq e ()" and it happens to trigger breakpoint 2, I would like to see something like > *Main:[1]:[2] > or some better notation, but you get my point. * The notation for printing a list is non-Haskell, consider the following evaluated head but unevaluated tail: > *Main> :print xs > xs = [True | (_t2::[Bool])] I would have expected "True : (_t2::[Bool])" * Consider the following code: > mymap f [] = [] > mymap f (x:xs) = f x:mymap f xs > > main = mapM_ putStrLn $ mymap ('a':) ["hello","bye"] First, I set a breakpoint on line two. When I invoke "main", the breakpoint is triggered two times before "main" is done evaluating. Afterwards, If I type "main" at the ghci prompt again, the breakpoint is not triggered anymore! Even though ":show breaks" lists it. * Again considering the code above, if I break at line 2: > Stopped at FreeVar2.hs:2:17-19 > x :: GHC.Base.Unknown > f :: t -> a > xs :: [GHC.Base.Unknown] Now I type "f" and the following message appears: > *** Exception: No match in record selector Var.tcTyVarDetails * I was wondering whether it would be nice to have a "break" function/ primitive, so that you can write: > mymap f [] = [] > mymap f (x:xs) = (break (f x):mymap f xs) with "break" having type "a -> a" and inserting a breakpoint around the argument expression. It is handy to not have to recalculate lines. However I am not so sure anymore. Using line numbers is not as bad as I thought. The debugger is a nice addition to GHC, and I look forward to the debugger being integrated into the Emacs Haskell mode! Cheers, Alexey

3 2

RE: Login problems with trac
by C Rodrigues 18 Apr '07

18 Apr '07

The wiki has "edit" links if I login as guest, but not if I login as heatsink. Is that also because of the spam issue? _________________________________________________________________ Get a FREE Web site, company branded e-mail and more from Microsoft Office Live! http://clk.atdmt.com/MRT/go/mcrssaub0050001411mrt/direct/01/

4 5

Release plans
by Simon Marlow 18 Apr '07

18 Apr '07

We'd like to solicit comments from the community on our plans for future GHC releases. The current situation is this: - 6.6.1 is nearly ready to go (perhaps this week, please test the RC!) - 6.6.2 has ~35 outstanding tickets - 6.8 has ~150 outstanding tickets the default option would be to work on 6.6.2 and release it in a couple of months, while continuing to work on 6.8 for a release in ~6 months or so. However, there's another option: we could skip 6.6.2 and head straight for 6.8, the idea being to stabilise 6.8 with the existing feature set (plus a few important developments in progress), so that we can get these features out in a release sooner rather than later. The 6.6.2 tickets would be moved to 6.8, and most of the outstanding 6.8 tickets would be pushed to 6.8.1 or later, retaining only the features/bug-fixes that are essential for the 6.8 release. Here's a quick summary of the major developments that we already have in the 6.8 codebase: - Associated data types, and the new FC intermediate language - GHCi debugger (although there's an overhaul of the breakpoint support almost ready to go in) - Coverage (HPC) - GADTs + typeclasses - more packages moved from core to extralibs - GHC API changes, compile to object code inside GHCi - performance improvements: simplifier rewrite, and SpecConstr improvements - standalone deriving declarations - left-to-right impredicative instantiation: runST $ foo Here are the features in progress that we'd like to get into 6.8: - pointer-tagging (15% perf improvement for compiled code) - get external core working again - base package breakup (see discussion on libraries list) - list fusion We think the above feature set makes for a pretty strong 6.8 release. What do you think of this plan? Are there features/bug-fixes that you really want to see in 6.8? Cheers, Simon & the GHC Team

14 27