- Glasgow-haskell-users

RE: [Linux-basic-compiling]GHC-3.02 build on GHC-2.10
by Simon Marlow 21 Nov '01

21 Nov '01

> I apologize if this is not the correct moment-place to deal with this > old version. > > I'm trying to compile the GHC-3.02 version using ghc-2.10 binaries, > happy-1.6 binaries on a i386-unknown-linux. > In the future I would like to hack with it. [ I presume you have a good reason for needing GHC 3.02 - I'd be interested to know what it is! ] It looks like that the ghc-2.10 binaries you are using don't work well with the version of Linux (in particular the version of the C library) that you're using. Probably the only way to get going is to bootstrap from the .hc files for 3.02. Cheers, Simon

1 0

RE: Binary library
by Simon Marlow 21 Nov '01

21 Nov '01

> Does Malcolm's Binary library exist for ghc? If not, is there > a standard > way (ie using Haskell that works on any compiler) to dump a data > structure into a file that another program or part of the same program > can then pick up and read (cf "serialize" in Java)? I'm > reluctant to go > with a solution that's compiler-specific. Using Show & Read is the only really portable way to do this, unfortunately. Sven Panne did a port of nhc's Binary library a while back, which you can find on haskell.org, but it will certainly need some work to get it going on recent GHC's. We'd really like a good serialisation library, BTW, if anyone out there feels like donating one! Cheers, Simon

1 0

GMP
by Marcin 'Qrczak' Kowalczyk 21 Nov '01

21 Nov '01

I think there exists a problem. I haven't encountered it because I haven't tried to provoke it, but if I'm not mistaken it can cause big trouble. Programs compiled by ghc set gmp's allocation functions to ones which allocate byte arrays on the Haskell heap, instead of the default malloc/free. Now imagine a program or library which uses gmp for its own purposes and is linked with a program or library compiled with ghc. GMP's hooks are global, gmp is shared. Neither of allocation methods is suitable for the other side. So it won't work. Is there a solution? -- __("< Marcin Kowalczyk * qrczak(a)knm.org.pl http://qrczak.ids.net.pl/ \__/ ^^ QRCZAK

2 1

profiling problem
by Andre W B Furtado 20 Nov '01

20 Nov '01

I've installed GHC-5.02 (and the profiling libraries) for Windows platforms (I have Windows98 running on my machine), but everytime I try to check the heap profiling for a program, I always get an empty graph (as the one attached). I'm sure I am doing the right steps (compiling the program with "-prof -auto-all" option, executing it with "+RTS -h" and then calling hp2ps). Any suggestions? -- Andre

2 3

RE: not naming modules Main
by Simon Peyton-Jones 20 Nov '01

20 Nov '01

(Moving this to GHC users.) What exactly do you propose? * Haskell allows you to define a 'main' in any module you please. * GHCi allows you to execute any of these 'main' functions independently. (You get all the speed of compiled code if you compile the module first and load the compiled module into GHCi.) * But if you want to link together a bunch of modules into an executable, there must be some way to decide which 'main' you want to execute. Haskell's way is to say "It's the 'main' from module Main". We like feedback about what would be useful -- GHC has lots of things that aren't exactly Haskell that were generated by such suggestions. But I'm unclear what you would like (let alone how easy it would be to implement). Simon | -----Original Message----- | From: Hal Daume [mailto:hdaume@ISI.EDU] | Sent: 16 November 2001 18:34 | To: haskell(a)haskell.org | Subject: not naming modules Main | | | I'm really frustrated that modules that you want to compile | to executables have to be named Main. I often have a module | with a main method that I use for testing or whatever | (perhaps I want the gained speed of an executable) but is, | for the most part, a module I import into others. I end up | having to constantly change the module name whenever I want | to compile it and I find this terribly frustrating. | | Is there any reason you can't just compile things that simply | export a main method with the proper type? Is this a ghc | specific thing or does nhc also have this restriction? Any | chance ghc will change its policy on this? | | - Hal | | -- | Hal Daume III | | "Computer science is no more about computers | hdaume(a)isi.edu | than astronomy is about telescopes." -Dijkstra | www.isi.edu/~hdaume | | _______________________________________________ | Haskell mailing list | Haskell(a)haskell.org http://www.haskell.org/mailman/listinfo/haskell | |

3 2

(no subject)
by Chen Yu 20 Nov '01

20 Nov '01

hi, there i am compiling ghc with a c library doing cryptology. however i notic there are conflicts on the definitions of these functions: MD5Init MD5Update MD5Final is there any way to remove them from ghc? many thanks! ------ regards, chenyu _________________________________________________________________ Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp

2 1

Binary library
by Amanda Clare 20 Nov '01

20 Nov '01

Does Malcolm's Binary library exist for ghc? If not, is there a standard way (ie using Haskell that works on any compiler) to dump a data structure into a file that another program or part of the same program can then pick up and read (cf "serialize" in Java)? I'm reluctant to go with a solution that's compiler-specific. Amanda

1 0

RE: not naming modules Main
by Simon Marlow 20 Nov '01

20 Nov '01

> Coming up with proposals is always much more difficult that simply > complaining about something. Here is a proposal which seems > reasonable to me (and implementable): > > Allow any module to have a main function. When you compile (with > ghc), and you say, for example: > % ghc --make Foo.hs -o Foo > then it looks in Foo.hs for the main function (regardless of > whether the name of the module appearing in Foo.hs is Main or > Foo). As I never use ghc without the --make option, I'm not sure > how this would roll over when not using it. Perhaps there could > be an option like: > % ghc Foo.hs Bar.hs FooBar.hs -o Foo -main-in Foo.hs > or something where you can specify which module has the main you > want to use. > > Then, say you have two modules, Foo.hs (containing module Foo) > and Bar.hs (containing module Bar), both with main functions, > suppose Foo is the one from which the desire main function comes, > you can always "import Bar hiding (main)". While we could do doubt implement this, I'm not sure the benefits are compelling enough. As Simon points out, you can always do your testing in GHCi which doesn't care whether you have a Main module or not. Furthermore this would be an extension to Haskell which isn't supported by the other compilers, so its usefulness is limited. Would anyone else find this useful? Cheers, Simon

1 0

RE: Posix.sync
by Simon Marlow 20 Nov '01

20 Nov '01

> > > Rather than patching the Posix library to include stuff like > > > O_SYNC, I'd like to see a Haskell binding to the Single > > > UNIX (aka Unix98) instead - it subsumes the > > > POSIX.1 APIs (and you can get at the specs without having > > > to pay the extortionate rates that IEEE charges). Feel free > > > to contribute such a binding! > > > > I'm afraid I will not. There is pretty huge number of > > functions to bind, and I'm afraid to miss some. Then I'm not > > wery good Haskeller. > > > > Moreover, I could not see the Single Unix features in the > > linux-2.2 (at least that nice O_DSYNC flag). Maybe they are > > presented in 2.4, but I have no plan to use it. > > AFAIK no OS implements all of the Single Unix specification. > Nevertheless, I think, Sigbjorn is right. Ideally, we > should have a binding for Single Unix and on any particular > OS, you will have a subset of it available (that's the same > situation as you have with any other programming language). > > As is pointed out in the open(2) man page, O_SYNC, O_DSYNC, > and O_RSYNC are synonymous in Linux 2.2. In 2.4, at least > with some file systems (eg, XFS) you can choose which > behaviour O_SYNC should have. I'd like to see a full SUS binding too. I also think that POSIX is a useful subset, because as Manuel says nobody implements the full SUS but pretty much everyone implements full POSIX (not always with the same semantics though :-). So my preferred solution would be a Unix library, from which Posix re-exports a subset. I started rewriting GHC's Posix library a while back - I got about half way though, and I'll get back onto it when I turn my attention to the new libraries again. I've kept all the suggestions from people on the lists regarding the Posix interface, and I'll put together a proposal for the new interface at some point. Discussion about the interface should probably move to libraries(a)haskell.org. Cheers, Simon

1 0

Posix.sync
by Max A . K . 20 Nov '01

20 Nov '01

Hello. What about include O_SYNC flag in the Posix.OpenFileFlag? Max.

3 4