June 2001 - Glasgow-haskell-users

Re: GHC core representation
by Josef Svenningsson 20 Jun '01

20 Jun '01

[This is a reply to Andrew Tolmachs message to the Haskell list about the -fext-core flag] Hi! Some comments on the -fext-core facility. First I just want to say that it's a really cool feature. It made me a lot keener to use GHC for program analysis measurements. First a question which is not really specific to the -fext-core flag (which I expect someone of the GHC-people to answer): * I would like to make GHC generate only the core file and the hi file, nothing more. However, I haven't found a way of making GHC stop after outputting the hi file. Being able to do this is useful in some other cases as well; I sometimes look at the raw core that comes is printed with the -ddump* flags. It is often the case that I don't want anything to be generated then. OK, here are the comments about the new feature: * In the grammar: the productions for the nonterminal Arguments seems to be erronious. When used as they are they cause tons of shift/reduce conflicts. Having the following solves to problem (I think this is what you really meant): Argument arg -> @ aty | aexp * Speaking of arguments, it would be a nice thing if one could get the output in A-normal-form. This is done by GHC anyway when converting to STG and in some cases it would be really useful. * Exactly when in the compilation process does ghc spit out this info (when I use -fext-core that is)? What I am really asking is: which transformations is applied to the core before it is printed? Knowing this is pretty important. Being able to control it would be super, and from the documentation this seems to be a goal. * A minor point: The example in the beginning of section 3.6 seems to be incorrect. It should read: f :: %forall a . a -> BinTree (BinTree a) = \ @a (x::a) -> Leaf @(BinTree a) (Leaf @a x) ¯¯¯¯¯¯¯¯¯¯¯ OK, I think that's all for now. I'll probarbly come up with some more questions in a while. Cheers, /Josef

1 0

RE: GHC version 5.00.2 is available
by Julian Seward (Intl Vendor) 20 Jun '01

20 Jun '01

| > What's new in 5.00.2 | > ====================== | > | [..] | | How about Andrew Tolmach's -fext-core flag? Did it make it | into 5.00.2 or do we have to wait for next official release? No, it didn't. The 5.00.X line exists purely to give (hopefully) increasingly stable renditions of what was released as 5.00, and in particular to fix show-stopping bugs. All new development happens in the CVS "HEAD" branch, which will get snapshotted into 5.02 at some point. J

1 0

passing Strings with FFI
by Corneliu Popeea 20 Jun '01

20 Jun '01

Hello, I want to pass a String to a function which is imported from C code: foreign import "pass" prim_pass :: String -> String The declaration above gives me an error from ghc like "String type not supported for imported functions". I thought that String being [Char] should be supported (somehow like a definition with "newtype" keyword). Can you tell me how can I pass strings to and from C code, as simple as possible (I know about HDirect, but I couldn't make it work for me: red-hat 7.1, ghc 5.00)? Best regards, Corneliu

1 0

Re: GHC version 5.00.2 is available
by Juan Carlos Arevalo Baeza 18 Jun '01

18 Jun '01

At 07:46 AM 6/18/2001 -0700, Julian Seward (Intl Vendor) wrote: >The (Interactive) Glasgow Haskell Compiler -- version 5.00.2 > >[...] > >Pre-built >packages for Linux, FreeBSD, and Solaris(sparc) are also available. > >[...] > > What's new in 5.00.1 >====================== > >[...] > >At the moment there is no Win32 build of 5.00 or 5.00.1 available. We >decided to push Win32 support to the 5.02 release, so as not to delay >5.00.1 any further. So, what's up? I've been completely unable to build it (the previous version) on my own... It says it requires Happy, and no matter what I do (Happy build just fine), it doesn't see it. No hurry, but still, I'd like to play with this new version. Salutaciones, JCAB --------------------------------------------------------------------- Juan Carlos "JCAB" Arevalo Baeza | http://www.roningames.com Senior Technology Engineer | mailto:jcab@roningames.com Ronin Entertainment | ICQ: 10913692 (my opinions are only mine) JCAB's Rumblings is so off-line O:-(

1 0

extra_ghc_opts
by S.D.Mechveliani 18 Jun '01

18 Jun '01

To my >> The lack of extra_command_line_opts is not important, >> but I wonder, why make things worse? >> [..] >> I thought the idea of -package was to collect in one place all >> the project-specific settings. For example, for my project docon, >> I need >> doconOpts = -fglasgow-exts -fallow-this-and-this ... Simon Marlow <simonmar(a)microsoft.com> writes > I'm in two minds about this. At the moment, using a -package option > does two well defined things: > > - it brings a set of modules into scope > - it links in the object code for those modules > > According to the Principle of Least Astonishment, we should force the > user to add any extra options himself, because if '-package wibble' > automatically adds a number of language-feature options then it's just > too magical and non-obvious. > > On the other hand, I presume you *need* these options in order to be > able to use Docon, so adding them automatically would seem reasonable. > > Opinions, anyone? To add or not to add automatically the language-feature options, or many other options, this is up to the designer of the package. The designer is always free to set [] in the package. And you were going to desable this choice (concerning package). ----------------- Serge Mechveliani mechvel(a)botik.ru

1 0

RE: Installation
by Simon Peyton-Jones 18 Jun '01

18 Jun '01

We have been labouring over the last few weeks to get a GHC that runs on Windows and does not require you to install anything else does not intefere with anything else you already have installed (notably cygwin) It was a big surprise to me how hard it is to achieve these two objectives simultaneously on a Windows machine. We are now (very) nearly there, after many exciting learning experiences. We're going to put out a thrill-seekers Win2k release of GHC 5 with the above properties. "Thrill-seekers" because it'll be the head of the tree and hence incorporate various things that 5.00.2 doesn't have but which are slightly less stable. So I think the simplest thing is to wait a week or two. If you are in a hurry, install GHC4.08.2, and cygwin, check out the HEAD, and say 'make'. (Works for me...) Simon | -----Original Message----- | From: Amit Garg [mailto:amitji@cs.utexas.edu] | Sent: 17 June 2001 22:39 | To: glasgow-haskell-users(a)haskell.org | Subject: Installation | | | Hello. | | Is there a simple way to get running with ghc over win2k? I | used the ghc4.08 installer ... but it doesnt like my cygwin. | I couldnt use it to compile hdirect or upgrade to ghc5. The | mkdepend script quits with a 'signal 127' message. Any help | would be most appreciated. | | -Amit Garg. | | | _______________________________________________ | Glasgow-haskell-users mailing list | Glasgow-haskell-users(a)haskell.org | http://www.haskell.org/mailman/listinfo/glasgow-| haskell-users |

1 0

Ray Tracing
by Heron 17 Jun '01

17 Jun '01

Hello Haskell & GHC people, I need the source code of a Ray Tracing written in Haskell. This code was used for parallelization using Caliban and Eden. Can you help me on how or where I can obtain it ? Heron de Carvalho, Msc.

2 1

5.00.1 performance test
by S.D.Mechveliani 15 Jun '01

15 Jun '01

Dear ghc-5.00.1 (binary, linux, i386-unknown), Here are some information on performance that may occur useful. Choose n, run test n in various modes and compare the timings. It consists mainly of merge sorting of a list of Int-s combined with reversing. The particular points I find are * ghc -interactive runs the compiled with -O code 1.4 times slower than it is for the batch mode and needs considerably larger minimal heap to obtain the result. * the performace difference between -O and -Onot looks small. ----------------- Serge Mechveliani mechvel(a)botik.ru -- statistics for test 55000 ------------------------- ghc-5.00.1 -interpreted time (under large heap) = 13.3 sec, minimal-needed(Heap -M...) = 12Mb compiled with -O -fvia-C -O2-for-C, loaded and run from -interactive time = 3.1, min(Heap) = 9Mb linked and run in batch mode 2.2 5Mb -Onot, run from -interactive 3.8 10Mb ghc-4.04 -Onot time for -M9m -H8m = 3.5 sec. -------------------------------------------------------- module CompPlatf (test) where -- main = putStr $ shows (test 55000) "\n" test :: Int -> Int test n = max' $ sort' $ zipWith (-) (reverse xs) xs where xs = [1..n] max' :: Ord a => [a] -> a max' [x] = x max' (x:y:xs) = case compare x y of LT -> max' (y:xs) _ -> max' (x:xs) sort' :: Ord a => [a] -> [a] sort' xs = s $ mergePairs [[x] | x <- xs] where s [] = [] s [xs] = xs s xss = s $ mergePairs xss mergePairs (xs:ys:zss) = (merge xs ys):(mergePairs zss) mergePairs xss = xss merge :: Ord a => [a] -> [a] -> [a] merge [] ys = ys merge xs [] = xs merge (x:xs) (y:ys) = case compare x y of GT -> y:(merge (x:xs) ys) _ -> x:(merge xs (y:ys))

1 0

RE: resumeThread cannot find the thread! (GHC/HDirect)
by Reuben Thomas 13 Jun '01

13 Jun '01

You need to use H/Direct 0.17, which unfortunately you have to compile from source. 0.16 (which I presume you're using) doesn't work with GHC 4.08.2. I checked your example against the CVS HEAD version of H/Direct, and it worked fine, except that I had to make the following changes (to correct unintentional errors, I presume): [incr.c] + #include <windows.h> - BOOL WINAPI DllMain(args...) { return TRUE; } + BOOL WINAPI DllMain(args) { return TRUE; } int __declspec(dllexport) incr(int x) { return x + 1; } [incr.idl] module Incr { [pure]int incr([in]int); } [Main.hs] module Main(main) where + import Incr main = print (incr 2)

2 1

RE: Empty Datatypes in GHC 5?
by Simon Peyton-Jones 11 Jun '01

11 Jun '01

Done, in the HEAD at least. Simon | -----Original Message----- | From: Ashley Yakeley [mailto:ashley@semantic.org] | Sent: 08 June 2001 01:14 | To: GHC List | Subject: Empty Datatypes in GHC 5? | | | I just joined this list after finally obtaining an i386 | machine primarily | for the purpose of running GHC (I ssh/ftp to it from my Mac). | | A quick question: is there an option to allow empty datatypes in GHC | 5.00? I couldn't see one in the manual. They are useful for type | parameterisation, for instance: | | data Zero; | data Succ n; | | Unfortunately this is contrary to the Haskell Report. | Nevertheless, Hugs | allows it. But GHC, apparently, does not. | | -- | Ashley Yakeley, Seattle WA | | | _______________________________________________ | Glasgow-haskell-users mailing list | Glasgow-haskell-users(a)haskell.org | http://www.haskell.org/mailman/listinfo/glasgow-| haskell-users |

1 0