From Ralf Hinze Thu Feb 1 10:00:53 2001 Date: Thu, 1 Feb 2001 11:00:53 +0100 (MET) From: Ralf Hinze Ralf Hinze Subject: Documentation 
Dear all,

I just successfully installed GHC 4.08.1 and Happy 1.9
from source under SuSE linux. However, it seems that
the doumentation is not generated. Do I have to take
special actions (NB configure seems to be satisfied with
the DocBook installation)?

Thanks, Ralf
From simonmar@microsoft.com Thu Feb 1 12:06:52 2001 Date: Thu, 1 Feb 2001 04:06:52 -0800 From: Simon Marlow simonmar@microsoft.com Subject: Documentation 
> > Yup, you have to explicitly ask for the documentation.  Go to
> > ghc/docs/set, and type 'make set'.  The full documentation (combined
> > Users' Guide and Library reference) will end up in set/ 
> after a while.
> 
> That does not work. make answers
> 	make: *** No rule to make target `set'.
> Any ideas? Do I have to take special actions when
> configuring?

oops, sorry I meant 'make html'.  You can also do 'make dvi', 'make ps',
'make pdf' etc.

Cheers,
	Simon
From simonmar@microsoft.com Thu Feb 1 11:14:14 2001 Date: Thu, 1 Feb 2001 03:14:14 -0800 From: Simon Marlow simonmar@microsoft.com Subject: Documentation 
> Dear all,
> 
> I just successfully installed GHC 4.08.1 and Happy 1.9
> from source under SuSE linux. However, it seems that
> the doumentation is not generated. Do I have to take
> special actions (NB configure seems to be satisfied with
> the DocBook installation)?
> 

Yup, you have to explicitly ask for the documentation.  Go to
ghc/docs/set, and type 'make set'.  The full documentation (combined
Users' Guide and Library reference) will end up in set/ after a while.

Cheers,
	Simon
From Ralf Hinze Thu Feb 1 12:02:39 2001 Date: Thu, 1 Feb 2001 13:02:39 +0100 (MET) From: Ralf Hinze Ralf Hinze Subject: Documentation 
> Yup, you have to explicitly ask for the documentation.  Go to
> ghc/docs/set, and type 'make set'.  The full documentation (combined
> Users' Guide and Library reference) will end up in set/ after a while.

That does not work. make answers
	make: *** No rule to make target `set'.
Any ideas? Do I have to take special actions when
configuring?

Ralf
From rrt1001@cam.ac.uk Fri Feb 2 14:29:25 2001 Date: Fri, 2 Feb 2001 14:29:25 +0000 (GMT) From: Reuben Thomas rrt1001@cam.ac.uk Subject: Documentation 
> That does not work. make answers
> 	make: *** No rule to make target `set'.
> Any ideas? Do I have to take special actions when
> configuring?

Simon meant "make html" and "make ps". "make set.html" and "make set.ps"
also work, I think.

-- 
http://sc3d.org/rrt/ | free, a.  already paid for (Peyton Jones)
From ger@Informatik.Uni-Bremen.DE Tue Feb 6 13:04:30 2001 Date: Tue, 06 Feb 2001 14:04:30 +0100 From: George Russell ger@Informatik.Uni-Bremen.DE Subject: Semi-applied datatypes in instance declarations 
I apologise if this has been raised before, but the code I am
writing now would look rather nicer if "partially applied 
type constructors" were permitted in instances.  For example:

class Event e where
   sync :: e a -> IO a

data Event extraData a = blah blah . . .

instance (context on extradata) => Event extradata where
   blah blah . . .

Any chance of this?  Or are there reasons why this would be wholly
ridiculous?
From v-julsew@microsoft.com Tue Feb 6 13:57:57 2001 Date: Tue, 6 Feb 2001 05:57:57 -0800 From: Julian Seward (Intl Vendor) v-julsew@microsoft.com Subject: ANNOUNCE: The Glasgow Haskell Compiler -- version 4.08.2 
           The Glasgow Haskell Compiler -- version 4.08.2
          ================================================

We are pleased to announce a new release of the Glasgow Haskell
Compiler (GHC), version 4.08.2.  The source distribution is freely
available via the World-Wide Web and through anon. FTP, under a
BSD-style license.  See below for download details.  Pre-built
packages for Linux, FreeBSD, Solaris and Win32 are also available.

Haskell is "the" standard lazy functional programming language; the
current language version is Haskell 98, agreed in December 1998.

GHC is a state-of-the-art optimising compiler for Haskell, generating
good code for a variety of platforms.  The distribution includes space
and time profiling facilities, a large collection of libraries, and
support for various language extensions, including concurrency,
exceptions, and foreign language interfaces (C, C++, whatever).

A wide variety of Haskell related resources (tutorials, libraries,
specifications, documentation, compilers, interpreters, references,
contact information, links to research groups) are available from the
Haskell home page at

        http://www.haskell.org/

GHC's Web page lives at

        http://www.haskell.org/ghc/


+ What's new in 4.08.2
=======================

No new features, just a few minor bug fixes.


+ What's new in 4.08
=====================

This should be a stable release.  There have been many enhancements
since 4.06, and shed-loads of bug-fixes (one shed (imperial) ~ one ton
(US)).

There are the following changes

   - New profiling subsystem, based on cost-centre stacks.

   - Working x86 native code generator: now it works properly, runs
     about twice as fast as compiling via C, and is on a par for
     run-time speed (except in FP-intensive programs).

   - Implicit parameters (i.e. dynamic scoping without the pain).

   - DEPRECATED pragma for marking obsolescent interfaces.

   - In the wake of hslibs, a new package system for
     libraries. -package should now be used instead of -syslib.

   - Result type signatures work.

   - Many tiresome long-standing bugs and problems (e.g. the trace
     problem) have been fixed.

   - Many error messages have been made more helpful and/or
     accurate.

For full details see the release notes:

        http://www.haskell.org/ghc/docs/4.08/users_guide/release-4-08.html


+ Important Info For Win32 users
=================================

ALERT: For reasons as yet not understood, as of January 2001 Cygwin's
mingw package has a problem that causes GHC-compiled binaries to be
built incorrectly. The fix is to ensure that your mingw package is
dated 20001111 (you can set this in the Cygwin installer). See the
installation guide for more details.


+ Mailing lists
================

We run mailing lists for GHC users and bug reports; to subscribe, use
the web interfaces at

        http://www.haskell.org/mailman/listinfo/glasgow-haskell-users
        http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs

There are several other haskell and ghc-related mailing lists on
www.haskell.org; for the full list, see

        http://www.haskell.org/mailman/listinfo/

Please send bug reports about GHC to glasgow-haskell-bugs@haskell.org;
GHC users hang out on glasgow-haskell-users@haskell.org.  Bleeding
edge CVS users party on cvs-ghc@haskell.org.


+ On-line GHC-related resources
================================

Relevant URLs on the World-Wide Web:

GHC home page             http://www.haskell.org/ghc/
Haskell home page         http://www.haskell.org/
comp.lang.functional FAQ  http://www.cs.nott.ac.uk/~gmh/faq.html


+ How to get it
================

The easy way is to go to the WWW page, which should be
self-explanatory:

        http://www.haskell.org/ghc/

Once you have the distribution, please follow the pointers in the
README file to find all of the documentation about this release.  NB:
preserve modification times when un-tarring the files (no `m' option
for tar, please)!


+ System requirements
======================

To compile the sources, you need a machine with 32+MB memory, GNU C
(`gcc'), `perl' plus a version of GHC installed (3.02 at least).  This
release is known to work on the following platforms:

  * i386-unknown-{linux,freebsd,netbsd,cygwin32,mingw32}
  * sparc-sun-solaris2
  * hppa1.1-hp-hpux{9,10}

Ports to the following platforms should be relatively easy (for a
wunderhacker), but haven't been tested due to lack of time/hardware:

  * i386-unknown-solaris2
  * alpha-dec-osf{2,3}
  * mips-sgi-irix{5,6}
  * {rs6000,powerpc}-ibm-aix

The builder's guide included in distribution gives a complete
run-down of what ports work; an on-line version can be found at

   http://www.haskell.org/ghc/docs/4.08/building/building-guide.html
From mk167280@students.mimuw.edu.pl Tue Feb 6 14:29:31 2001 Date: Tue, 6 Feb 2001 15:29:31 +0100 (CET) From: Marcin 'Qrczak' Kowalczyk mk167280@students.mimuw.edu.pl Subject: Semi-applied datatypes in instance declarations 
On Tue, 6 Feb 2001, George Russell wrote:

> I apologise if this has been raised before, but the code I am
> writing now would look rather nicer if "partially applied 
> type constructors" were permitted in instances.

They are. For example monads.

-- 
Marcin 'Qrczak' Kowalczyk
From ger@Informatik.Uni-Bremen.DE Tue Feb 6 14:34:28 2001 Date: Tue, 06 Feb 2001 15:34:28 +0100 From: George Russell ger@Informatik.Uni-Bremen.DE Subject: Semi-applied datatypes in instance declarations 
Marcin 'Qrczak' Kowalczyk wrote:
> 
> On Tue, 6 Feb 2001, George Russell wrote:
> 
> > I apologise if this has been raised before, but the code I am
> > writing now would look rather nicer if "partially applied
> > type constructors" were permitted in instances.
> 
> They are. For example monads.
Yes you are right.  I'm sorry for wasting your time - I made a mistake
in constructing my test case.
From Malcolm.Wallace@cs.york.ac.uk Tue Feb 6 14:31:02 2001 Date: Tue, 6 Feb 2001 14:31:02 +0000 From: Malcolm Wallace Malcolm.Wallace@cs.york.ac.uk Subject: GHC-4.08.2 
The download links on
    http://www.mirror.ac.uk/sites/www.haskell.org/ghc/
for the new GHC 4.08.2 have somewhat (ahem) strange referents.

(The RedHat 6 binary package gives me a picture of Haskell B Curry,
the RH6 profiling package gives me a .dvi file of the dynamic semantics
of the language, RH7 source RPM gives me an Emacs hugs-mode....)

The original haskell.org site seems ok, just mirror.ac.uk is broken.

    Malcolm
From Malcolm.Wallace@cs.york.ac.uk Tue Feb 6 14:33:22 2001 Date: Tue, 6 Feb 2001 14:33:22 +0000 From: Malcolm Wallace Malcolm.Wallace@cs.york.ac.uk Subject: GHC-4.08.2 
> The original haskell.org site seems ok, just mirror.ac.uk is broken.

Oops, retract that.  The RedHat 6 packages all seem to be 4.08.1, not
the advertised 4.08.2.

Regards,
    Malcolm
From simonmar@microsoft.com Tue Feb 6 16:21:41 2001 Date: Tue, 6 Feb 2001 08:21:41 -0800 From: Simon Marlow simonmar@microsoft.com Subject: GHC on Sourceforge 
Hi folks,

A while ago we opened a SourceForge project for GHC, mainly to make use
of the bug tracking facilities.  It's been working quite happily for
some time now, so I think it's time to start using the bug tracking
system for all bug reports.

To report a bug, go to

	http://sourceforge.net/projects/ghc/

and follow the "bugs" link, then "submit a bug".  Please include all the
usual information you'd normally give with a bug report on the mailing
list.

We also plan to start using the task manager (as a kind of wish list /
job adverts for GHC), and perhaps some of the other cool gadgets that
SourceForge provides.

Cheers,
	Simon
From chak@cse.unsw.edu.au Wed Feb 7 02:57:23 2001 Date: Wed, 07 Feb 2001 13:57:23 +1100 From: Manuel M. T. Chakravarty chak@cse.unsw.edu.au Subject: GHC-4.08.2 
Malcolm Wallace <Malcolm.Wallace@cs.york.ac.uk> wrote,

> > The original haskell.org site seems ok, just mirror.ac.uk is broken.
> 
> Oops, retract that.  The RedHat 6 packages all seem to be 4.08.1, not
> the advertised 4.08.2.

There shouldn't be any links to RedHat 6 packages.  I don't
have a RedHat 6.x machine anymore, and so, can't make any
packages for it.  If you have a RedHat 6.x box with an older
version of GHC already installed, just run

  rpm --rebuild ghc-4.08.2-1.src.rpm

This should (with a little waiting ;-)  produce binary
packages for 4.08.2 running on your flavour of RedHat.

Cheers,
Manuel
From rrt1001@cam.ac.uk Wed Feb 7 10:27:11 2001 Date: Wed, 7 Feb 2001 10:27:11 +0000 (GMT) From: Reuben Thomas rrt1001@cam.ac.uk Subject: GHC-4.08.2 
> Malcolm Wallace <Malcolm.Wallace@cs.york.ac.uk> wrote,
>
> > > The original haskell.org site seems ok, just mirror.ac.uk is broken.
> >
> > Oops, retract that.  The RedHat 6 packages all seem to be 4.08.1, not
> > the advertised 4.08.2.
>
> There shouldn't be any links to RedHat 6 packages.

We're providing the links for 4.08.1 since that's all we have, and a lot of
people still seem to use RH6.

> packages for it.  If you have a RedHat 6.x box with an older
> version of GHC already installed, just run
>
>   rpm --rebuild ghc-4.08.2-1.src.rpm

This will only work if you install RPM 4, no? I couldn't find a way to make
RPM 4 produce v3 RPMs, either. But never mind, I installed RPM 4, installed
the source RPM, downgraded again to RPM 3, and am now building RH 6 RPMs,
which I'll make available.

One problem I had: the spec file doesn't have a dependency on the jade
package, but it needs it for some DSSSL files without which the docs won't
build.

-- 
http://sc3d.org/rrt/ | RSA, n.  safety in numbers
From chak@cse.unsw.edu.au Thu Feb 8 06:09:19 2001 Date: Thu, 08 Feb 2001 17:09:19 +1100 From: Manuel M. T. Chakravarty chak@cse.unsw.edu.au Subject: GHC-4.08.2 
Reuben Thomas <rrt1001@cam.ac.uk> wrote,

> > packages for it.  If you have a RedHat 6.x box with an older
> > version of GHC already installed, just run
> >
> >   rpm --rebuild ghc-4.08.2-1.src.rpm
> 
> This will only work if you install RPM 4, no? I couldn't find a way to make
> RPM 4 produce v3 RPMs, either. But never mind, I installed RPM 4, installed
> the source RPM, downgraded again to RPM 3, and am now building RH 6 RPMs,
> which I'll make available.

True.  I didn't know whether the src.rpm format also changed.

> One problem I had: the spec file doesn't have a dependency on the jade
> package, but it needs it for some DSSSL files without which the docs won't
> build.

Right - there should be a build dependency.

Cheers,
Manuel
From qrczak@knm.org.pl Thu Feb 8 07:36:38 2001 Date: 8 Feb 2001 07:36:38 GMT From: Marcin 'Qrczak' Kowalczyk qrczak@knm.org.pl Subject: GHC-4.08.2 
Thu, 08 Feb 2001 17:09:19 +1100, Manuel M. T. Chakravarty <chak@cse.unsw.edu.au> pisze:

> > This will only work if you install RPM 4, no?
> 
> True.  I didn't know whether the src.rpm format also changed.

AFAIK it did not change.

-- 
 __("<  Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZASTĘPCZA
QRCZAK
From chak@cse.unsw.edu.au Sun Feb 11 12:20:07 2001 Date: Sun, 11 Feb 2001 23:20:07 +1100 From: Manuel M. T. Chakravarty chak@cse.unsw.edu.au Subject: ANN: C->Haskell 0.8.1 
I am pleased to announce the availability of version 0.8.1
of the interface generator C->Haskell.  It works with the
current stable release series 4.08.x of GHC as well as the
current development series 4.11.  For both versions of GHC,
it supports the *same* FFI library that GHC natively only
supports in the development version 4.11 and which
constitutes the result of the work of the FFI Task Force
over the last couple of months.  The interface specification
of the library is online available at

  http://www.cse.unsw.edu.au/~chak/haskell/c2hs/docu/c2hs-4.html

For more information on C->Haskell and for downloading, see

  http://www.cse.unsw.edu.au/~chak/haskell/c2hs/

The main feature in this release is the new FFI library.  It
should allow users of the GHC stable series to use the new
FFI library interface, which we hope to keep stable from now
on.  An update of the C->Haskell tool proper is being worked
at.

Happy Hacking,
Manuel
From qrczak@knm.org.pl Sun Feb 11 13:36:31 2001 Date: 11 Feb 2001 13:36:31 GMT From: Marcin 'Qrczak' Kowalczyk qrczak@knm.org.pl Subject: ANN: C->Haskell 0.8.1 
Sun, 11 Feb 2001 23:20:07 +1100, Manuel M. T. Chakravarty <chak@cse.unsw.edu.au> pisze:

> I am pleased to announce the availability of version 0.8.1
> of the interface generator C->Haskell.

Wow!

A problem: c2hs/gen/CInfo.lhs imports class Storable from module
C2HSConfig, which does not export it.

Also, the compilation fails (on ghc-4.11):
/usr/local/bin/ghc -c -syslib lang -syslib posix -O -recomp -fno-warn-incomplete-patterns  -i. -fglasgow-exts   C2HSMarsh.hs

C2HSMarsh.hs:43:
    Bad interface file: ./Ptr.hi
        does not exist

and similarly for other files. This is because there exists Ptr.hs
in this directory - but it should not be taken from there, because
the right version is in ghc's own libraries.

I don't know how to solve such problem in general, except by splitting
modules into many directories playing with -i options (that's why
QForeign had to use different module names for modules it brings
which may or may not come with the compiler, except some modules not
present in nhc98 which are handled by -i options).

-- 
 __("<  Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZASTĘPCZA
QRCZAK
From qrczak@knm.org.pl Sun Feb 11 13:38:18 2001 Date: 11 Feb 2001 13:38:18 GMT From: Marcin 'Qrczak' Kowalczyk qrczak@knm.org.pl Subject: ANN: C->Haskell 0.8.1 
11 Feb 2001 13:36:31 GMT, Marcin 'Qrczak' Kowalczyk <qrczak@knm.org.pl> pisze:

> A problem:
[...]

These problems apply to cvs' HEAD.

-- 
 __("<  Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZASTĘPCZA
QRCZAK
From chak@cse.unsw.edu.au Mon Feb 12 03:07:01 2001 Date: Mon, 12 Feb 2001 14:07:01 +1100 From: Manuel M. T. Chakravarty chak@cse.unsw.edu.au Subject: ANN: C->Haskell 0.8.1 
qrczak@knm.org.pl (Marcin 'Qrczak' Kowalczyk) wrote,

> Sun, 11 Feb 2001 23:20:07 +1100, Manuel M. T. Chakravarty <chak@cse.unsw.edu.au> pisze:
> 
> > I am pleased to announce the availability of version 0.8.1
> > of the interface generator C->Haskell.
> 
> Wow!
> 
> A problem: c2hs/gen/CInfo.lhs imports class Storable from module
> C2HSConfig, which does not export it.
> 
> Also, the compilation fails (on ghc-4.11):
> /usr/local/bin/ghc -c -syslib lang -syslib posix -O -recomp -fno-warn-incomplete-patterns  -i. -fglasgow-exts   C2HSMarsh.hs
> 
> C2HSMarsh.hs:43:
>     Bad interface file: ./Ptr.hi
>         does not exist
> 
> and similarly for other files. This is because there exists Ptr.hs
> in this directory - but it should not be taken from there, because
> the right version is in ghc's own libraries.

This is very strange, because I have tested this and it
worked for me.  In fact, if you look at the Makefile in the
c2hs/lib/ directory, then you will see that the `depend'
target moves these troublesome files into a subdirectory
before compilation (called ghc411hackdir).  

Is it possible that you used the same build tree to compile
with 4.08 first and then re-compile with 4.11?  This won't
work, because of the hack mentioned above.  

In fact, I think, this is a bug in ghc 4.11 - I have filled
it as bug #131631 in the bug tracker.

Cheers,
Manuel
From info@tips2you.co.uk Mon Feb 12 07:51:46 2001 Date: Mon, 12 Feb 2001 07:51:46 From: * TIPS 4 U * info@tips2you.co.uk Subject: * TIPS 4 U * 
This is the first time we have advertised this service anywhere as most of our 
clients/participants have been referred to us by word of mouth or are friends of the family 
who have benefited from our information. 

We feel it only fair, bearing in mind the enormous profit generating potential here, that we 
share this information with others and enable everyone to enjoy a better standard of living. 
This is why we have kept the fees for providing such information as low as Ł10. 

Our horses generate a steady profit week in week out and month in and month out. We 
have never shown a month of loss. Many tipsters nowadays have runs of good luck in 
accordance with seasonal changes. Some do well in winter others in summer, some on the 
flat and others on the jumps. We have maintained a consistent profit stream, which ensures 
that we do not have to charge exorbitant prices in order to offset any losses. We are sure 
that you will find us one of the most competitive tipping services anywhere. 

The service is launched for members only now. For more information e-mail us at 
info@tips2you.co.uk

*************************************************************************************************
To be removed please state remove in the header and your details will be deleted 
immediately.
**************************************************************************************************
From qrczak@knm.org.pl Mon Feb 12 14:48:09 2001 Date: 12 Feb 2001 14:48:09 GMT From: Marcin 'Qrczak' Kowalczyk qrczak@knm.org.pl Subject: ANN: C->Haskell 0.8.1 
Mon, 12 Feb 2001 14:07:01 +1100, Manuel M. T. Chakravarty <chak@cse.unsw.edu.au> pisze:

> This is very strange, because I have tested this and it
> worked for me.

It could be my fault: I ran ./configure before updating it with
autoconf, and perhaps did not clean afterwards.

> In fact, if you look at the Makefile in the c2hs/lib/ directory,
> then you will see that the `depend' target moves these troublesome
> files into a subdirectory before compilation

So they will be downloaded each time from the CVS... Perhaps it
would be better to store them under different names and make symlinks
if needed?

-- 
 __("<  Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZASTĘPCZA
QRCZAK
From dongen@cs.ucc.ie Mon Feb 12 21:05:15 2001 Date: Mon, 12 Feb 2001 21:05:15 +0000 From: Marc van Dongen dongen@cs.ucc.ie Subject: lreadline 
Hello all,


When linking, ghc-4.08.1 fails due to a
 ld: fatal: library -lreadline: not found
error. Is this a ghc error or should I get
that library from somewhere else?

Thanks in advance.


Marc van Dongen
-- 
     Marc van Dongen, CS Dept | phone:  +353 21 4903578
University College Cork, NUIC | Fax:    +353 21 4903113
  College Road, Cork, Ireland | Email: dongen@cs.ucc.ie
From chak@cse.unsw.edu.au Tue Feb 13 07:46:58 2001 Date: Tue, 13 Feb 2001 18:46:58 +1100 From: Manuel M. T. Chakravarty chak@cse.unsw.edu.au Subject: ANN: C->Haskell 0.8.1 
qrczak@knm.org.pl (Marcin 'Qrczak' Kowalczyk) wrote,

> Mon, 12 Feb 2001 14:07:01 +1100, Manuel M. T. Chakravarty <chak@cse.unsw.edu.au> pisze:
> 
> > In fact, if you look at the Makefile in the c2hs/lib/ directory,
> > then you will see that the `depend' target moves these troublesome
> > files into a subdirectory before compilation
> 
> So they will be downloaded each time from the CVS... Perhaps it
> would be better to store them under different names and make symlinks
> if needed?

They are only moved away in the build tree (where they are
only sym links anyway).  if you clean the whole tree and
start again with configuration and build, you should be
fine.

Manuel
From andreas.marth@daimlerchrysler.com Tue Feb 13 12:01:19 2001 Date: Tue, 13 Feb 2001 13:01:19 +0100 From: andreas.marth@daimlerchrysler.com andreas.marth@daimlerchrysler.com Subject: mingw-20001111-1.tar.gz still needed for cygwin WinNT? 
Hi is the mingw-20001111-1.tar.gz still needed? How do I install it wit=
h the=20
actual cygwin (just downloaded)?
I put it in the ../latest/mingw/ directory but can not chose it in the =

installation program.
Any help appreciated.
Thanks,
Andreas=
From andreas.marth@daimlerchrysler.com Tue Feb 13 13:08:39 2001 Date: Tue, 13 Feb 2001 14:08:39 +0100 From: andreas.marth@daimlerchrysler.com andreas.marth@daimlerchrysler.com Subject: Antwort: Re: mingw-20001111-1.tar.gz still needed for cygwin WinNT? 
It cycles throug 2 alternatives: "20010130-1" and "20001225-1".
Even the choices what to downlod is to this mingw packages restricted. =

So I see only two alternatives:
1.) Use 20010130-1 or 20001225-1
2.) Use an older cygwin
I have not found an old cygwin1.1 at http://sources.redhat.com/cygwin/.=


Any comments?

Bye,
Andreas




rrt1001@cam.ac.uk@cl.cam.ac.uk am 13.02.2001 13:21:34
Gesendet von:	reuben.thomas@cl.cam.ac.uk
An:	Andreas Marth/FT/DCAG/DCX@WK-EMEA2
Kopie:	 =20

Thema:	Re: mingw-20001111-1.tar.gz still needed for cygwin WinNT?

> Hi is the mingw-20001111-1.tar.gz still needed?

Yes, as far as I know.

> How do I install it with the
> actual cygwin (just downloaded)?

You should be able to select it in the installer even without downloadi=
ng
it: if you click on the package date, it should cycle through several c=
hoices.

--=20
http://sc3d.org/rrt/
"Reality is what refuses to disappear when you stop believing in it" -
Philip K. Dick




=
From christian.lescher@icn.siemens.de Tue Feb 13 16:03:03 2001 Date: Tue, 13 Feb 2001 17:03:03 +0100 From: Lescher Christian christian.lescher@icn.siemens.de Subject: mingw-20001111-1.tar.gz still needed for cygwin WinNT? 
I had the same problem, and I solved it by changing the setup.ini file after downloading & putting the old mingw archive in latest/mingw:

@ mingw
version: 20001111-1
install: latest/mingw/mingw-20001111-1.tar.gz 130113
source:  latest/mingw/mingw-20001111-1-src.tar.gz 121498

Cheers, Christian
From christian.lescher@icn.siemens.de Tue Feb 13 16:11:14 2001 Date: Tue, 13 Feb 2001 17:11:14 +0100 From: Lescher Christian christian.lescher@icn.siemens.de Subject: Message: "DEFAULT_TMPDIR has not been set to anything useful!" 
After installing the newest version of GHC on WinNT/Cygwin, I get the message 

  "DEFAULT_TMPDIR has not been set to anything useful!"

when executing GHC (however, compilation is ok.)

Setting DEFAULT_TMPDIR to "c:/tmp" didn't change anything. What may be wrong with my settings?

Christian
From andreas.marth@daimlerchrysler.com Tue Feb 13 20:06:41 2001 Date: Tue, 13 Feb 2001 21:06:41 +0100 From: andreas.marth@daimlerchrysler.com andreas.marth@daimlerchrysler.com Subject: questions about the installation instructions for WinNT and ghc-4.08x 
I have some question relating the intsallation of ghc-4.08x on a WinNT/=
cygwin=20
system.

In the installation instruction is under 2.2.2.1 paragraph "Here's how =
to=20
install Cygwin" written:
	After installation, start up a Cygwin shell and issue the following=20
command:=20
		mount -f c: /
 	assuming you installed Cygwin at C:\cygwin; otherwise change the driv=
e=20
and directory as appropriate.
(So it definitely says / is at c: not at c:\cygwin. But later, when set=
ting the=20
environment variables it seems like it is assumed
that / is at c:\cygwin.)
Is it right to mount / at c: or should it be c:\cygwin?

In "2.2.2.2 Environment variables" is a table that says I should add C:=
\usr\bin=20
to PATH.
That directory does not exist. the directory C:\cygwin\usr\bin exists (=
assuming=20
you installed Cygwin at C:\cygwin) but is empty.
The C:\cygwin\bin directory seems to be the one meant.
Should I add C:\cygwin\bin to PATH?

The same table says that SHELL should be C:/usr/bin/bash. (Observe the =
'/'=20
signs instead of '\'.)
But bash is in C:\cygwin\bin (assuming you installed Cygwin at C:\cygwi=
n).
Should I set SHELL to C:\cygwin\bin\bash (does it matter if I write=20
C:\cygwin\bin\bash or C:/cygwin/bin/bash?)?


Okay that are my questions at the moment. Any help would be appreciated=
.

PS:Is there still something else to observe than putting ghc into scope=
 of PATH?
=
From rrt1001@cam.ac.uk Wed Feb 14 11:07:56 2001 Date: Wed, 14 Feb 2001 11:07:56 +0000 (GMT) From: Reuben Thomas rrt1001@cam.ac.uk Subject: questions about the installation instructions for WinNT and ghc-4.08x 
> In the installation instruction is under 2.2.2.1 paragraph "Here's how to
> install Cygwin" written:
> 	After installation, start up a Cygwin shell and issue the following
> command:
> 		mount -f c: /
>  	assuming you installed Cygwin at C:\cygwin; otherwise change the drive
> and directory as appropriate.
> (So it definitely says / is at c: not at c:\cygwin. But later, when setting the
> environment variables it seems like it is assumed
> that / is at c:\cygwin.)

There are two things going on here:

1. You're mounting C: at / so you can refer to the whole drive with
UNIX-style paths.

2. Cygwin automatically mounts parts of its directory tree under /usr.

> In "2.2.2.2 Environment variables" is a table that says I should add C:\usr\bin
> to PATH.
> That directory does not exist. the directory C:\cygwin\usr\bin exists (assuming
> you installed Cygwin at C:\cygwin) but is empty.
> The C:\cygwin\bin directory seems to be the one meant.
> Should I add C:\cygwin\bin to PATH?

No. Add /usr/bin. Cygwin mounts \cygwin\usr\bin at /usr/bin.

> The same table says that SHELL should be C:/usr/bin/bash. (Observe the '/'
> signs instead of '\'.)
> But bash is in C:\cygwin\bin (assuming you installed Cygwin at C:\cygwin).
> Should I set SHELL to C:\cygwin\bin\bash (does it matter if I write
> C:\cygwin\bin\bash or C:/cygwin/bin/bash?)?

As above.

I think these instructions have worked for other people; my own system is in
a slightly different configuration (which for various reasons we used to
recommend).

-- 
http://sc3d.org/rrt/ | humour, n.  unexpected recognition
From qrczak@knm.org.pl Wed Feb 14 23:14:24 2001 Date: 14 Feb 2001 23:14:24 GMT From: Marcin 'Qrczak' Kowalczyk qrczak@knm.org.pl Subject: ANN: C->Haskell 0.8.1 
12 Feb 2001 14:48:09 GMT, Marcin 'Qrczak' Kowalczyk <qrczak@knm.org.pl> pisze:

> It could be my fault: I ran ./configure before updating it with
> autoconf, and perhaps did not clean afterwards.

Ok, it was my fault. But the error in CInfo is still there. And
C2HS does not export most of cast*Ptr functions, freeHaskellFunPtr
and newStablePtr, which leads to errors in compilig gtk+hs. And
GtkCList (from the CVS version of gtk+hs) does not import C2HS but
uses newStablePtr etc. And ghc version check in gtk+hs fails for 4.11.

-- 
 __("<  Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZASTĘPCZA
QRCZAK
From trebla@vex.net Thu Feb 15 04:35:36 2001 Date: 14 Feb 2001 23:35:36 -0500 From: Albert Lai trebla@vex.net Subject: Confused about enabling parallel ghc 
I installed ghc 4.08.2 binaries for sparc solaris (and those for intel
linux, too), then tried compiling a sample program (the fibonacci one
in the user guide) with ghc -parallel.  The compiler complained:

    Could not find interface file for `Prelude'
    in the directories ./*.hi
                       /u/trebla/lib/ghc-4.08.2/imports/lang/*.mp_hi
                       /u/trebla/lib/ghc-4.08.2/imports/concurrent/*.mp_hi
                       /u/trebla/lib/ghc-4.08.2/imports/lang/*.mp_hi
                       /u/trebla/lib/ghc-4.08.2/imports/concurrent/*.mp_hi
                       /u/trebla/lib/ghc-4.08.2/imports/std/*.mp_hi

(and similarly for Parallel.)

Does it mean I must build ghc myself from source?

The user guide promises a lot of binary bundles such as par.  Ever
since 4.0x, the ghc download page ceases to mention them.  Where
are they?
From chak@cse.unsw.edu.au Thu Feb 15 07:44:56 2001 Date: Thu, 15 Feb 2001 18:44:56 +1100 From: Manuel M. T. Chakravarty chak@cse.unsw.edu.au Subject: ANN: C->Haskell 0.8.1 
qrczak@knm.org.pl (Marcin 'Qrczak' Kowalczyk) wrote,

> 12 Feb 2001 14:48:09 GMT, Marcin 'Qrczak' Kowalczyk <qrczak@knm.org.pl> pisze:
> 
> > It could be my fault: I ran ./configure before updating it with
> > autoconf, and perhaps did not clean afterwards.
> 
> Ok, it was my fault. But the error in CInfo is still
> there. 

That's strange.  Could you try this with either the release
of 0.8.2 that I put up yesterday or with the CVS version
again?  I don't have any trouble with CInfo with both 4.08
and 4.11.

> And
> C2HS does not export most of cast*Ptr functions, freeHaskellFunPtr
> and newStablePtr, which leads to errors in compilig gtk+hs. And
> GtkCList (from the CVS version of gtk+hs) does not import C2HS but
> uses newStablePtr etc. And ghc version check in gtk+hs fails for 4.11.

As noted on the Web page, 0.8.1 didn't work with gtk+hs.
You need c2hs 0.8.2 and the new version 0.10.2 of gtk+hs,
which I released yesterday.  (Alternatively, the versions of
both packages in CVS should work, too.)

Cheers,
Manuel
From qrczak@knm.org.pl Thu Feb 15 08:20:00 2001 Date: 15 Feb 2001 08:20:00 GMT From: Marcin 'Qrczak' Kowalczyk qrczak@knm.org.pl Subject: ANN: C->Haskell 0.8.1 
Thu, 15 Feb 2001 18:44:56 +1100, Manuel M. T. Chakravarty <chak@cse.unsw.edu.au> pisze:

> > But the error in CInfo is still there. 
> 
> That's strange.  Could you try this with either the release of
> 0.8.2 that I put up yesterday or with the CVS version again?

Hmm, it works now. There are two files called C2HSConfig.hs (in
c2hs/lib and c2hs/toplevel). I don't know why the wrong one was
being imported.

> As noted on the Web page, 0.8.1 didn't work with gtk+hs.
> You need c2hs 0.8.2 and the new version 0.10.2 of gtk+hs,
> which I released yesterday.  (Alternatively, the versions of
> both packages in CVS should work, too.)

Both are current versions from CVS, with version numbers as you say.

Does C2HSDeprecated export newStablePtr and freeHaskellFunPtr?
Currently it does not, but GtkCList assumes it does.

Does C2HS export castPtrToFunPtr? Currently it does not, but GMarsh
assumes it does.

-- 
 __("<  Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZASTĘPCZA
QRCZAK
From simonmar@microsoft.com Thu Feb 15 11:00:55 2001 Date: Thu, 15 Feb 2001 03:00:55 -0800 From: Simon Marlow simonmar@microsoft.com Subject: Confused about enabling parallel ghc 
> I installed ghc 4.08.2 binaries for sparc solaris (and those for intel
> linux, too), then tried compiling a sample program (the fibonacci one
> in the user guide) with ghc -parallel.  The compiler complained:
> 
>     Could not find interface file for `Prelude'
>     in the directories ./*.hi
>                        /u/trebla/lib/ghc-4.08.2/imports/lang/*.mp_hi
>                        
> /u/trebla/lib/ghc-4.08.2/imports/concurrent/*.mp_hi
>                        /u/trebla/lib/ghc-4.08.2/imports/lang/*.mp_hi
>                        
> /u/trebla/lib/ghc-4.08.2/imports/concurrent/*.mp_hi
>                        /u/trebla/lib/ghc-4.08.2/imports/std/*.mp_hi
> 
> (and similarly for Parallel.)
> 
> Does it mean I must build ghc myself from source?
> 
> The user guide promises a lot of binary bundles such as par.  Ever
> since 4.0x, the ghc download page ceases to mention them.  Where
> are they?

This just means that Parallel support isn't included in the binary
bundle you have.  In fact, parallel support is developed and distributed
separately.  Contact the GPH (Glasgow Parallel Haskell) folks for more
information:

	http://www.cee.hw.ac.uk/~dsg/gph/

Cheers,
	Simon
From qrczak@knm.org.pl Thu Feb 15 15:35:45 2001 Date: 15 Feb 2001 15:35:45 GMT From: Marcin 'Qrczak' Kowalczyk qrczak@knm.org.pl Subject: ANN: C->Haskell 0.8.1 
15 Feb 2001 08:20:00 GMT, Marcin 'Qrczak' Kowalczyk <qrczak@knm.org.pl> pisze:

> Does C2HSDeprecated export newStablePtr and freeHaskellFunPtr?
> Currently it does not, but GtkCList assumes it does.
> 
> Does C2HS export castPtrToFunPtr? Currently it does not, but GMarsh
> assumes it does.

Now I see: these functions in module C2HS are present in the released
c2hs-0.8.2, but not in the CVS. I guess you have not committed changes.

gtk+hs' examples which compile with the present interface don't link on
ghc-4.08.2, because of the ghc's bug in handling newtypes in foreign
exports which references rts_mkPtr. This is because you made Addr a
type synonym for Ptr ().

It can be "fixed" without making Addr incompatible with Ptr (which
I guess is needed because c2hs generates Addr and code uses Ptr)
by something like this:
    module PtrHack where
        import qualified Addr
        newtype Addr a = Ptr Addr.Addr
    module C2HSSomething where
        import qualified PtrHack
        type Ptr = PtrHack.Addr
        type Addr = Ptr ()
This ensures that the "real" name of the Ptr type is Addr.

I'll try this hack for QForeign to see if it can reduce the amount
of #ifdefs for broken compilers. It applies to newtypes in arguments
and results of functions in foreign export and foreign export dynamic
in ghc-4.08*.

It should be applied to CInt etc. too, to let them work there. I can
provide my own CTypes for ghc-4.08*, but at least I will get rid of
many of those stupid #ifdefs.

Unfortunately it does not help for Ptr in the result of foreign export
dynamic (ghc-4.08) nor in the argument of foreign import dynamic
(ghc-4.08*), where newtypes don't work. This means that gtk+hs does
not compile on 4.08 because of Ptr () (spelled as Addr) in the result
of foreign export dynamic.

Here is which ghc versions are broken in which ways:

            |                 newtypes work in foreign...                 |
            |                                                             |
            |   export   |  export   |   import   |  import   |           |
            | stat.& dyn.|  dynamic  | stat.& dyn.|  dynamic  |   label   |
            | (function) | (pointer) | (function) | (pointer) | (pointer) |
------------+------------+-----------+------------+-----------+-----------+
 ghc-4.08   |   hacked   |    no     |    yes     |    no     |    no     |
 ghc-4.08.1 |   hacked   |    yes    |    yes     |    no     |    yes    |
 ghc-4.08.2 |   hacked   |    yes    |    yes     |    no     |    yes    |
 ghc-4.11   |    yes     |    yes    |    yes     |    yes    |    yes    |

"Hacked" means that they work as long as the type name after expanding
type synonyms is recognized by the rts (and there is no way to #include
something in stubs I think).

-- 
 __("<  Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZASTĘPCZA
QRCZAK
From simonmar@microsoft.com Thu Feb 15 17:31:36 2001 Date: Thu, 15 Feb 2001 09:31:36 -0800 From: Simon Marlow simonmar@microsoft.com Subject: Haskell threads & pipes & UNIX processes 
[ moved to glasgow-haskell-users@haskell.org... ]

> I need to compress the output of my Hakell program. To avoid 
> the creation
> of huge files, I want to compress before writing by means of gzip or
> bzip2. However, this seems to be quite involved.
> 
> I decided to run the compressor with 
> 
> runProcess :: FilePath                    -- Command
>            -> [String]                    -- Arguments
>             -> Maybe [(String, String)]    -- 
> Environment(Nothing -> Inherited)
>             -> Maybe FilePath              -- Working 
> directory (Nothing -> inherited)
>             -> Maybe Handle                -- stdin (Nothing 
> -> inherited)
>             -> Maybe Handle                -- stdout (Nothing 
> -> inherited)
>             -> Maybe Handle                -- stderr (Nothing 
> -> inherited)
>             -> IO ()
> 
> First I wanted to use Posix.createPipe to connect to the 
> compressor but
> Posix.createPipe returns a Fd. So this does not fit.

Aha, you want the (unadvertised) function
	
	PosixIO.fdToHandle :: Fd -> Handle

> I ended up with using Posix.createNamedPipe. This creates a named pipe
> that can be openend with IO.openFile (which again yields the 
> Handle needed
> for runProcess).
> 
> At this point, it turned out that the pipe has to be opened 
> for reading
> before it can be opened for writing (ghc-4.08.1). This seems 
> to be a bug.
> (In a shell, the order does not matter; the processes are suspended 
> as expected.)

We've had some problems with named pipes - check the archives of
glasgow-haskell-bugs.  It seems certain operating systems disagree on
the semantics of a non-blocking open of a named pipe.  I don't recommend
using named pipes if you want portable code.

> Then, as expected, the problem occured that writing to the pipe and
> compressing cannot be sequenced. If the producer is started first, the
> producer blocks because there is no consumer. If the consumer 
> is started
> first, the consumer blocks because there is no input.
> 
> So I played around with threads. If both consumer and producer are
> executed as threads, the program terminates immediately without any
> output. There seems to be no need to start executing any of 
> the threads.

In GHC, the main program terminates as soon as the main thread
terminates.  It doesn't wait for any child threads to terminate - if you
want this behaviour, you can program it using MVars.

We'll be happy to look at any of the other problems you mentioned:
please some code demonstrating the problem to
glasgow-haskell-bugs@haskell.org.

Cheers,
	Simon
From chak@cse.unsw.edu.au Fri Feb 16 07:42:08 2001 Date: Fri, 16 Feb 2001 18:42:08 +1100 From: Manuel M. T. Chakravarty chak@cse.unsw.edu.au Subject: ANN: C->Haskell 0.8.1 
qrczak@knm.org.pl (Marcin 'Qrczak' Kowalczyk) wrote,

> 15 Feb 2001 08:20:00 GMT, Marcin 'Qrczak' Kowalczyk <qrczak@knm.org.pl> pisze:
> 
> > Does C2HSDeprecated export newStablePtr and freeHaskellFunPtr?
> > Currently it does not, but GtkCList assumes it does.
> > 
> > Does C2HS export castPtrToFunPtr? Currently it does not, but GMarsh
> > assumes it does.
> 
> Now I see: these functions in module C2HS are present in the released
> c2hs-0.8.2, but not in the CVS. I guess you have not committed changes.

Oops, you are right, I did only partially commit my last
changes.  I am sorry for that.  Now, everything should be
checked in. 

> gtk+hs' examples which compile with the present interface don't link on
> ghc-4.08.2, because of the ghc's bug in handling newtypes in foreign
> exports which references rts_mkPtr. This is because you made Addr a
> type synonym for Ptr ().
> 
> It can be "fixed" without making Addr incompatible with Ptr (which
> I guess is needed because c2hs generates Addr and code uses Ptr)
> by something like this:
>     module PtrHack where
>         import qualified Addr
>         newtype Addr a = Ptr Addr.Addr
>     module C2HSSomething where
>         import qualified PtrHack
>         type Ptr = PtrHack.Addr
>         type Addr = Ptr ()
> This ensures that the "real" name of the Ptr type is Addr.
> 
> I'll try this hack for QForeign to see if it can reduce the amount
> of #ifdefs for broken compilers. It applies to newtypes in arguments
> and results of functions in foreign export and foreign export dynamic
> in ghc-4.08*.
> 
> It should be applied to CInt etc. too, to let them work there. I can
> provide my own CTypes for ghc-4.08*, but at least I will get rid of
> many of those stupid #ifdefs.
> 
> Unfortunately it does not help for Ptr in the result of foreign export
> dynamic (ghc-4.08) nor in the argument of foreign import dynamic
> (ghc-4.08*), where newtypes don't work. This means that gtk+hs does
> not compile on 4.08 because of Ptr () (spelled as Addr) in the result
> of foreign export dynamic.

I thought that I had fixed all this for Gtk+HS.  (In fact,
all Gtk+HS examples are running fine with GHC 4.08 on my
machine.)  Have a look at the file gtk+hs/gtk/ghcRtsAux.c.
It defines rts_mkPtr in a somewhat nasty way, but it works :-)
It's a bit like your hack, but on the C level.

Maybe you forgot to run autoconf and ./configure after your
last cvs update for Gtk+HS?

> Here is which ghc versions are broken in which ways:
> 
>             |                 newtypes work in foreign...                 |
>             |                                                             |
>             |   export   |  export   |   import   |  import   |           |
>             | stat.& dyn.|  dynamic  | stat.& dyn.|  dynamic  |   label   |
>             | (function) | (pointer) | (function) | (pointer) | (pointer) |
> ------------+------------+-----------+------------+-----------+-----------+
>  ghc-4.08   |   hacked   |    no     |    yes     |    no     |    no     |
>  ghc-4.08.1 |   hacked   |    yes    |    yes     |    no     |    yes    |
>  ghc-4.08.2 |   hacked   |    yes    |    yes     |    no     |    yes    |
>  ghc-4.11   |    yes     |    yes    |    yes     |    yes    |    yes    |
> 
> "Hacked" means that they work as long as the type name after expanding
> type synonyms is recognized by the rts (and there is no way to #include
> something in stubs I think).

Yes, I agree.  I also tried to get something #include'ed in
stubs, but failed.  That would have been the easiest
solution.

Anyway, thanks for checking that stuff.

Cheers,
Manuel

PS: With the current Gtk+HS source in CVS, all Gtk+HS
    examples as well as the iHaskell library and its three
    examples should now all work again.  I tested it all on
    my machine.
From qrczak@knm.org.pl Fri Feb 16 09:21:51 2001 Date: 16 Feb 2001 09:21:51 GMT From: Marcin 'Qrczak' Kowalczyk qrczak@knm.org.pl Subject: ANN: C->Haskell 0.8.1 
Fri, 16 Feb 2001 18:42:08 +1100, Manuel M. T. Chakravarty <chak@cse.unsw.edu.au> pisze:

> Now, everything should be checked in.

Seems OK, thanks.

ghc411hack_dir does not work for me again: ghc -M in build/ghc4/chs/lib
can't find NewStablePtr.hs (because there is only NewStablePtr.hs.in)
and the whole 'make depend' there fails, without removing files
conflicing with ghc' libraries. The failure of 'make depend' is
ignored and finally 'make' there fails to compile C2HS.hs.

> I thought that I had fixed all this for Gtk+HS.  (In fact,
> all Gtk+HS examples are running fine with GHC 4.08 on my
> machine.)  Have a look at the file gtk+hs/gtk/ghcRtsAux.c.
> It defines rts_mkPtr in a somewhat nasty way, but it works :-)

I've seen the hack, but on another box freshly installed ghc-4.08.2,
c2hs from tarball and gtk+hs from tarball did not work (linker can't
find rts_mkPtr). I must see if ghcRtsAux.c is compied at all there.

> PS: With the current Gtk+HS source in CVS, all Gtk+HS
>     examples as well as the iHaskell library and its three
>     examples should now all work again.  I tested it all on
>     my machine.

ghc version check fails on 4.11 (it should be lexicographic comparison
of version number components, not conjunction of independent
comparisons).

Checking for buggy readXXXOffAddr (for ghc-4.03..4.06) is now
unnecessary as gtk+hs requires ghc-4.08.1 anyway.

-- 
 __("<  Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZASTĘPCZA
QRCZAK
From qrczak@knm.org.pl Fri Feb 16 09:23:28 2001 Date: 16 Feb 2001 09:23:28 GMT From: Marcin 'Qrczak' Kowalczyk qrczak@knm.org.pl Subject: ANN: C->Haskell 0.8.1 
16 Feb 2001 09:21:51 GMT, Marcin 'Qrczak' Kowalczyk <qrczak@knm.org.pl> pisze:

> ghc411hack_dir does not work for me again:

Ah, I see: haven't run autoconf in c2hs subdirectory. Now it will
hopefully work.

-- 
 __("<  Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZASTĘPCZA
QRCZAK
From qrczak@knm.org.pl Fri Feb 16 09:40:02 2001 Date: 16 Feb 2001 09:40:02 GMT From: Marcin 'Qrczak' Kowalczyk qrczak@knm.org.pl Subject: ANN: C->Haskell 0.8.1 
Fri, 16 Feb 2001 18:42:08 +1100, Manuel M. T. Chakravarty <chak@cse.unsw.edu.au> pisze:

> PS: With the current Gtk+HS source in CVS, all Gtk+HS
>     examples as well as the iHaskell library and its three
>     examples should now all work again.  I tested it all on
>     my machine.

gtk+hs under ghc-4.11 needs -package lang in HCFLAGS in configure.in.

-- 
 __("<  Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZASTĘPCZA
QRCZAK
From qrczak@knm.org.pl Fri Feb 16 19:12:14 2001 Date: 16 Feb 2001 19:12:14 GMT From: Marcin 'Qrczak' Kowalczyk qrczak@knm.org.pl Subject: ANN: C->Haskell 0.8.1 
Fri, 16 Feb 2001 18:42:08 +1100, Manuel M. T. Chakravarty <chak@cse.unsw.edu.au> pisze:

> PS: With the current Gtk+HS source in CVS, all Gtk+HS
>     examples as well as the iHaskell library and its three
>     examples should now all work again.

They compile but they don't run correctly on ghc-4.11. It's because
GMarsh.writeCharOffAddr does not simulate Addr.writeCharOffAddr when
Char is wide, because instance Storable Char treats Char as wide.
You need to cast to CChar, Word8, or Int8 (be careful with sign
extension).

-- 
 __("<  Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZASTĘPCZA
QRCZAK
From chak@cse.unsw.edu.au Sun Feb 18 02:25:24 2001 Date: Sun, 18 Feb 2001 13:25:24 +1100 From: Manuel M. T. Chakravarty chak@cse.unsw.edu.au Subject: ANN: C->Haskell 0.8.1 
qrczak@knm.org.pl (Marcin 'Qrczak' Kowalczyk) wrote,

> > ghc411hack_dir does not work for me again:
> 
> Ah, I see: haven't run autoconf in c2hs subdirectory. Now it will
> hopefully work.

Automating the invocation fo autoconf is on the TODO list,
but I didn't get around to it yet.

Manuel
From chak@cse.unsw.edu.au Mon Feb 19 04:53:41 2001 Date: Mon, 19 Feb 2001 15:53:41 +1100 From: Manuel M. T. Chakravarty chak@cse.unsw.edu.au Subject: ANN: C->Haskell 0.8.1 
qrczak@knm.org.pl (Marcin 'Qrczak' Kowalczyk) wrote,

> Fri, 16 Feb 2001 18:42:08 +1100, Manuel M. T. Chakravarty <chak@cse.unsw.edu.au> pisze:
> 
> > I thought that I had fixed all this for Gtk+HS.  (In fact,
> > all Gtk+HS examples are running fine with GHC 4.08 on my
> > machine.)  Have a look at the file gtk+hs/gtk/ghcRtsAux.c.
> > It defines rts_mkPtr in a somewhat nasty way, but it works :-)
> 
> I've seen the hack, but on another box freshly installed ghc-4.08.2,
> c2hs from tarball and gtk+hs from tarball did not work (linker can't
> find rts_mkPtr). I must see if ghcRtsAux.c is compied at all there.

Strange - meanwhile, I have put release 0.10.4 out.  Among
other things, it finally has an `install' target.

> > PS: With the current Gtk+HS source in CVS, all Gtk+HS
> >     examples as well as the iHaskell library and its three
> >     examples should now all work again.  I tested it all on
> >     my machine.
> 
> ghc version check fails on 4.11 (it should be lexicographic comparison
> of version number components, not conjunction of independent
> comparisons).
> 
> Checking for buggy readXXXOffAddr (for ghc-4.03..4.06) is now
> unnecessary as gtk+hs requires ghc-4.08.1 anyway.

Ok, I fixed these.

> > PS: With the current Gtk+HS source in CVS, all Gtk+HS
> >     examples as well as the iHaskell library and its three
> >     examples should now all work again.
> 
> They compile but they don't run correctly on ghc-4.11. It's because
> GMarsh.writeCharOffAddr does not simulate Addr.writeCharOffAddr when
> Char is wide, because instance Storable Char treats Char as wide.
> You need to cast to CChar, Word8, or Int8 (be careful with sign
> extension).

`castCCharToChar' and `castCharToCChar' should be sufficient
here, right?  I have added them in the 0.10.4 release.

Thanks,
Manuel
From rrt1001@cam.ac.uk Mon Feb 19 15:35:24 2001 Date: Mon, 19 Feb 2001 15:35:24 +0000 (GMT) From: Reuben Thomas rrt1001@cam.ac.uk Subject: [Bug #133086] complete failure (fwd) 
[This sort of thing is probably better on the list, because it doesn't
really isolate a bug]

> Date: Mon, 19 Feb 2001 07:25:29 -0800
> From: noreply@sourceforge.net
> To: noreply@sourceforge.net, noreply@sourceforge.net, cvs-ghc@haskell.org
> Subject: [Bug #133086] complete failure

> Bug #133086, was updated on 2001-Feb-19 07:25
> Here is a current snapshot of the bug.

> Details: I have tried to install ghc-4.08.2 under Windows NT
> and (as usal with the last few releases) failed.

Sorry to hear you've had difficulties with recent releases; for most people
things seem to have improved.

> I get an immediate warning that DEFAULT_TMPDIR is not set to anything
> useful.

Check you've got the most recent InstallShield (same date as the current
one); there's a bug in previous versions.

> Although files will compile they run without producing
> any output (even Hello World).

This is the mingwin problem.

> By the way,
> I note that mingw from Cygwin has no version dated
> 20001111 but the only version before this January
> is listed as 20001225 (yes really Christmas day).

There is a 20001111, but since it's now the last version but one it's not
available directly from the installer; you have to get it by hand. Sorry
about that. It looks as though the latest version (from this year) has cured
the problem, so you could try using that instead. As I write I'm building
GHC 4.08.2 with the new mingwin package installed; if that works, I'll ship
a new InstallShield shortly.
From qrczak@knm.org.pl Mon Feb 19 19:02:29 2001 Date: 19 Feb 2001 19:02:29 GMT From: Marcin 'Qrczak' Kowalczyk qrczak@knm.org.pl Subject: ANN: C->Haskell 0.8.1 
Mon, 19 Feb 2001 15:53:41 +1100, Manuel M. T. Chakravarty <chak@cse.unsw.edu.au> pisze:

> Strange - meanwhile, I have put release 0.10.4 out.
> Among other things, it finally has an `install' target.

Thanks, everything compiles and mostly works. Examples still behave
strangely: rngtest goes crazy when changing the number of digits,
all rngtest's widgets are displayed with some black border unless the
mouse cursor is over them, the scale in ih/examples/Counter does not
work, threadtest does not work etc.

> `castCCharToChar' and `castCharToCChar' should be sufficient
> here, right?

Yes. Same in C2HSDeprecated.

-- 
 __("<  Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZASTĘPCZA
QRCZAK
From janna@cs.chalmers.se Tue Feb 20 10:31:11 2001 Date: Tue, 20 Feb 2001 11:31:11 +0100 From: Janna Khegai janna@cs.chalmers.se Subject: Exposing Haskell functions with String arguments in dll 
This is a multi-part message in MIME format.

------=_NextPart_000_000F_01C09B30.A07791A0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Hi,

I want to expose the following Haskell function in dll to call it from =
Java application:=20
=20
trans: String -> IO String=20

I read that I shoud use tools like GreenCard, HDirect or some =
marshalling libraries, but not sure, which way is better.
If anyone has experience in this, I will be very grateful for an =
example, which shows the steps of bilding such dll.

Thanks in advance, Janna Khegai.

------=_NextPart_000_000F_01C09B30.A07791A0
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 5.50.4611.1300" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>Hi,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>I&nbsp;want to&nbsp;expose the=20
following&nbsp;Haskell function in dll to call it from Java=20
application:&nbsp;</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>&nbsp;</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>trans: String -&gt; IO =
String&nbsp;</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>I read that I shoud use tools like =
GreenCard,=20
HDirect or some marshalling libraries, but not sure, which way is=20
better.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>If anyone has&nbsp;experience in this, =
I will be=20
very grateful for an example, which shows the steps of bilding such=20
dll.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Thanks in advance, Janna=20
Khegai.</FONT></DIV></BODY></HTML>

------=_NextPart_000_000F_01C09B30.A07791A0--
From bgregor@BUPHY.bu.edu Tue Feb 20 13:25:46 2001 Date: Tue, 20 Feb 2001 08:25:46 -0500 From: Brian Gregor bgregor@BUPHY.bu.edu Subject: language shootout 
There's a page with microbenchmarks for a large array of
languages, including Haskell:
	http://www.bagley.org/~doug/shootout/

and please note his disclaimer :-) :
	http://www.lib.uchicago.edu/keith/crisis/disclaimer.html

Anyway, many of the categories are lacking a Haskell
program.  I've sent in 2.  I thought that this could also
serve as a nice little collection of examples for style
and the use of some of the libraries in HSlibs for those
of us (particularly me :) ) who are still learning the
language.

Any takers?



-brian
From byron.hale@einfo.com Wed Feb 21 09:35:09 2001 Date: Wed, 21 Feb 2001 01:35:09 -0800 From: Byron Hale byron.hale@einfo.com Subject: Installation of mingw-20001111-1.tar.gz 
Hi,

I'm trying to install mingw-20001111-1.tar.gz for ghc4.08.2. The 
instructions on Win2K. The Web instructions seem a bit cavalier, but 
perhaps it is just my ignorance.

..."directory called mingw, place that in a directory called latest, then 
run the Cygwin installer "...

Are these intended to be sub directories of Cygwin's "bin" directory, or what?
Is the Cygwin "installer" "bin/install.exe?" If so, install with what options?

Best Regards,

Byron Hale
byron.hale@einfo.com
From rrt1001@cam.ac.uk Wed Feb 21 10:07:59 2001 Date: Wed, 21 Feb 2001 10:07:59 +0000 (GMT) From: Reuben Thomas rrt1001@cam.ac.uk Subject: Installation of mingw-20001111-1.tar.gz 
> I'm trying to install mingw-20001111-1.tar.gz for ghc4.08.2. The
> instructions on Win2K. The Web instructions seem a bit cavalier, but
> perhaps it is just my ignorance.

Neither, really, it's just that we can't document our entire world
(documenting our system is hard enough). We are forced just to skate over
the externals. If you don't know about one or more of the pieces we rely on,
you should read their documentation (if it exists).

However, within an hour or so I should have uploaded a fixed InstallShield
that works around what does indeed turn out to be a bug in mingwin, so you
won't need to bother with all this junk.

-- 
http://sc3d.org/rrt/ | Careful Cyclists Approaching From Right
From rrt1001@cam.ac.uk Wed Feb 21 10:10:45 2001 Date: Wed, 21 Feb 2001 10:10:45 +0000 (GMT) From: Reuben Thomas rrt1001@cam.ac.uk Subject: Installation of mingw-20001111-1.tar.gz 
> Are these intended to be sub directories of Cygwin's "bin" directory, or what?
> Is the Cygwin "installer" "bin/install.exe?" If so, install with what options?

...and just to clear these up: the directories I mention are not supposed to
be subdirectories of anywhere in particular; it doesn't matter where you put
them. The Cygwin installer is the program you run to install Cygwin
(setup.exe, as downloaded from http://cygwin.com/ or one of its aliases).

-- 
http://sc3d.org/rrt/ | certain, a.  insufficiently analysed
From rrt1001@cam.ac.uk Wed Feb 21 15:59:45 2001 Date: Wed, 21 Feb 2001 15:59:45 +0000 (GMT) From: Reuben Thomas rrt1001@cam.ac.uk Subject: New GHC InstallShield: mingwin problem fixed 
I have just uploaded a new GHC InstallShield. It fixes the recent problem
with needing a particular version of the mingw package. Unfortunately the
situation here is still unstable (roll on Cygwin 1.2), so it may break
again, but it seems to work for now.

-- 
http://sc3d.org/rrt/ | certain, a.  insufficiently analysed
From byron.hale@einfo.com Wed Feb 21 21:30:18 2001 Date: Wed, 21 Feb 2001 13:30:18 -0800 From: Byron Hale byron.hale@einfo.com Subject: Installation of mingw-20001111-1.tar.gz 
Thanks, Reuben. You're very prompt.

Byron
byron.hale@einfo.com

At 10:07 AM 2/21/2001 +0000, you wrote:
> > I'm trying to install mingw-20001111-1.tar.gz for ghc4.08.2. The
> > instructions on Win2K. The Web instructions seem a bit cavalier, but
> > perhaps it is just my ignorance.
>
>Neither, really, it's just that we can't document our entire world
>(documenting our system is hard enough). We are forced just to skate over
>the externals. If you don't know about one or more of the pieces we rely on,
>you should read their documentation (if it exists).
>
>However, within an hour or so I should have uploaded a fixed InstallShield
>that works around what does indeed turn out to be a bug in mingwin, so you
>won't need to bother with all this junk.
>
>--
>http://sc3d.org/rrt/ | Careful Cyclists Approaching From Right
>
>
>
>_______________________________________________
>Glasgow-haskell-users mailing list
>Glasgow-haskell-users@haskell.org
>http://www.haskell.org/mailman/listinfo/glasgow-haskell-users
From qrczak@knm.org.pl Fri Feb 23 00:23:24 2001 Date: 23 Feb 2001 00:23:24 GMT From: Marcin 'Qrczak' Kowalczyk qrczak@knm.org.pl Subject: --add-package 
In practice installation script of an external package does
--del-package before --add-package, because if it was installed before,
--add-package would cause an error.

Wouldn't be reasonable then to let --add-package overwrite any existing
package of that name? There is no real safety, because the package
is explicitly deleted anyway.

-- 
 __("<  Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZASTĘPCZA
QRCZAK
From simonmar@microsoft.com Fri Feb 23 10:50:27 2001 Date: Fri, 23 Feb 2001 10:50:27 -0000 From: Simon Marlow simonmar@microsoft.com Subject: --add-package 
> In practice installation script of an external package does
> --del-package before --add-package, because if it was=20
> installed before,
> --add-package would cause an error.
>=20
> Wouldn't be reasonable then to let --add-package overwrite=20
> any existing
> package of that name? There is no real safety, because the package
> is explicitly deleted anyway.

I don't think I agree.  Installation of an external package shouldn't be
doing --delete-package first: that should be left to the user (or rpm
-e, or whatever).  If you want to install a new package that replaces an
existing one, you have to remove the existing one first.

Cheers,
	Simon
From Dominic.J.Steinitz@BritishAirways.com Fri Feb 23 11:09:34 2001 Date: 23 Feb 2001 11:09:34 Z From: Steinitz, Dominic J Dominic.J.Steinitz@BritishAirways.com Subject: Sockets on Windows 
--PART.BOUNDARY.bawhub1.4eaf.3a9645c6.0001
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: Quoted-Printable
Content-Disposition: inline

I am was trying out sockets. These worked fine under linux. Under windows I=
 get the following error. Does anyone know what it means and how I can fix =
it? I can telnet from within the cygwin window so I assume tcp/ip is runnin=
g.

Thanks, Dominic.

administrator@DEFAULT /cygdrive/d/dom/home/socketTest
$ ./server1 8000

Fail: does not exist
Action: getProtocolByName
Reason: no such protocol entry




---------------------------------------------------------------------------=
----------------------
21st century air travel     http://www.britishairways.com

--PART.BOUNDARY.bawhub1.4eaf.3a9645c6.0001
Content-Type: application/octet-stream; name="ATTKHVFA"
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment; filename="ATTKHVFA"
Content-Description: ATTKHVFA
FTBP-Modification-Date: 23 Feb 2001 11:09:00 Z
FTBP-Object-Size: 1390

--------------------------------------------------

-- Simple Server

-- Chris Reade, Nov 2000

-- Based on example by  Hermann Oliveira Rodrigues

--------------------------------------------------



module Main (main) where



import System

import IO

import Time

import Socket



main :: IO ()

main = do prog <- getProgName

          args <- getArgs

          if (length args /= 1)

              then do putStrLn ("Use: " ++ prog ++ " <port>")

                      exitWith (ExitFailure (-1))

              else return ()

          let port = read (args !! 0) :: Int in

              server (PortNumber (mkPortNumber port))



-- The server function creates a socket to listen on the port and

-- loops to service requests continuously



server :: PortID -> IO ()

server port = do socket <- listenOn port 

                 -- Fique em `loop' atendendo requisicoes.

                 let loop = do (sh, host, portid) <- accept socket

                               procRequest sh

                               loop in

                     loop 



-- The procRequest function deals with an individual request



procRequest :: Handle -> IO ()

procRequest hd = do clock <- getClockTime

                    calendar <- toCalendarTime clock

                    hPutStrLn hd (calendarTimeToString calendar)

                    hFlush hd

                    hClose hd


--PART.BOUNDARY.bawhub1.4eaf.3a9645c6.0001--
From rrt1001@cam.ac.uk Fri Feb 23 11:41:56 2001 Date: Fri, 23 Feb 2001 11:41:56 +0000 (GMT) From: Reuben Thomas rrt1001@cam.ac.uk Subject: Sockets on Windows 
> administrator@DEFAULT /cygdrive/d/dom/home/socketTest
> $ ./server1 8000
>
> Fail: does not exist
> Action: getProtocolByName
> Reason: no such protocol entry

Just from looking at the manpage for getprotobyname, I surmise that the
problem is that you don't have an /etc/protocols file (or equivalent) under
Windows. I'm not sure what's supposed to happen (or for that matter, what's
supposed to work). Look in hslibs/net/BSD.lhs; I couldn't work out what was
going on in a minute; perhaps someone else knows.

The other thing I couldn't work out was where getProtocolByName was being
called in the first place, but presumably it's to find the protocol number
of tcp.

-- 
http://sc3d.org/rrt/ | Slow Pedestrian Crossing
From mwr22@cam.ac.uk Fri Feb 23 12:19:41 2001 Date: Fri, 23 Feb 2001 12:19:41 -0000 From: Matthew Richards mwr22@cam.ac.uk Subject: Sockets on Windows 
You need to wrap up the use of sockets in a "withSocketsDo", as in:

> server port = withSocketsDo $
>               do socket <- listenOn port
>			-- etc

I seem to recall the documentation says you always need to use that, so I
don't know why it works without on Linux :-)

Hope this helps,
Matthew


-----Original Message-----
From: glasgow-haskell-users-admin@haskell.org
[mailto:glasgow-haskell-users-admin@haskell.org]On Behalf Of Steinitz,
Dominic J
Sent: Friday, February 23, 2001 11:10 AM
To: glasgow-haskell-users
Subject: Sockets on Windows


I am was trying out sockets. These worked fine under linux. Under windows I
get the following error. Does anyone know what it means and how I can fix
it? I can telnet from within the cygwin window so I assume tcp/ip is
running.

Thanks, Dominic.

administrator@DEFAULT /cygdrive/d/dom/home/socketTest
$ ./server1 8000

Fail: does not exist
Action: getProtocolByName
Reason: no such protocol entry




----------------------------------------------------------------------------
---------------------
21st century air travel     http://www.britishairways.com
From mk167280@students.mimuw.edu.pl Fri Feb 23 12:37:32 2001 Date: Fri, 23 Feb 2001 13:37:32 +0100 (CET) From: Marcin 'Qrczak' Kowalczyk mk167280@students.mimuw.edu.pl Subject: --add-package 
On Fri, 23 Feb 2001, Simon Marlow wrote:

> I don't think I agree.  Installation of an external package shouldn't be
> doing --delete-package first: that should be left to the user (or rpm
> -e, or whatever).

rpm is a different story. I would expect 'make install' to be idempotent.
With your scheme one has to write 'make uninstall install'.

-- 
Marcin 'Qrczak' Kowalczyk
From simonmar@microsoft.com Fri Feb 23 14:07:22 2001 Date: Fri, 23 Feb 2001 06:07:22 -0800 From: Simon Marlow simonmar@microsoft.com Subject: --add-package 
> On Fri, 23 Feb 2001, Simon Marlow wrote:
> 
> > I don't think I agree.  Installation of an external package 
> shouldn't be
> > doing --delete-package first: that should be left to the 
> user (or rpm
> > -e, or whatever).
> 
> rpm is a different story. I would expect 'make install' to be 
> idempotent.
> With your scheme one has to write 'make uninstall install'.

ghc --add-package behaves just like rpm --install, which seems
reasonable to me.  Why should it behave differently?

If you want 'make install' to be idempotent (which I'm not sure is
desirable anyway), then you will indeed have to ghc --delete-package
first, but then I hope you're going to print a big warning and wait for
the user to press any key first :)

Cheers,
	Simon
From andreas.marth@daimlerchrysler.com Fri Feb 23 19:53:59 2001 Date: Fri, 23 Feb 2001 20:53:59 +0100 From: andreas.marth@daimlerchrysler.com andreas.marth@daimlerchrysler.com Subject: Problems when mounting c: to / 
Hallo everybody!

When installing ghc-4.08.2 (onWinNT) as described in the installation=20
instructions under c:, I couldn't get ghc to run.=20
It invoked the C-preprozessor and stoped after cleaning.
When I unmount c:, it works fine. (If I install cygwin and dont mount c=
: to /=20
it works fine too.)=20
But now //c/ is the normal win-root. That's not a problem for me, but h=
ow about=20
ghc?=20
Does it depend in any way on c: mounted to /?
If not why do you want this mount (in the  installation instructions se=
ction=20
2.2.2.1)?

Thanks,
Andreas

PS: Thanks Reuben for the new install shield and a very big  thanks to =
all of=20
you who have their parts in getting the profiling to run under WinNT=
From rrt1001@cam.ac.uk Sat Feb 24 12:59:04 2001 Date: Sat, 24 Feb 2001 12:59:04 +0000 (GMT) From: Reuben Thomas rrt1001@cam.ac.uk Subject: Problems when mounting c: to / 
> When installing ghc-4.08.2 (onWinNT) as described in the installation
> instructions under c:, I couldn't get ghc to run.
> It invoked the C-preprozessor and stoped after cleaning.

I'm puzzled by this. Could you please send some sample output? Preferably
running ghc -v.

> When I unmount c:, it works fine. (If I install cygwin and dont mount c: to /
> it works fine too.)
> But now //c/ is the normal win-root. That's not a problem for me, but how about
> ghc?

Well, if it works, I suppose it's not a problem...

> Does it depend in any way on c: mounted to /?

I'm not sure. I think it used to (because of using /tmp as the temporary
directory). Now it's mostly for convenience.

> If not why do you want this mount (in the  installation instructions section
> 2.2.2.1)?

I think it used to be necessary, and I kept it in the instructions partly
because I find it useful to be able to refer to the drive with Unix-style
paths, with the root directory being /, and partly because it worked, and
the less I change the instructions, the less likely I am to break something.

-- 
http://sc3d.org/rrt/ | wit, n.  educated insolence (Aristotle)
From stolz@I2.Informatik.RWTH-Aachen.DE Sat Feb 24 18:41:48 2001 Date: Sat, 24 Feb 2001 19:41:48 +0100 From: Volker Stolz stolz@I2.Informatik.RWTH-Aachen.DE Subject: Sockets on Windows 
In local.glasgow-haskell-users, you wrote:
>You need to wrap up the use of sockets in a "withSocketsDo", as in:
>I seem to recall the documentation says you always need to use that, so I
>don't know why it works without on Linux :-)

No, that notice only applies to Windows.
-- 
\usepackage[latin1]{inputenc}!
Volker Stolz * stolz@i2.informatik.rwth-aachen.de * PGP + S/MIME
From James Lelyveld Mon Feb 26 11:32:55 2001 Date: Mon, 26 Feb 2001 11:32:55 GMT From: James Lelyveld James Lelyveld Subject: message 
[This message is sent through a WWW-Email gateway.]
[The authenticity of the sender can not be validated.]
[Message sent from the following machine - 62.172.105.102 ]
[after accessing this URL http://www.dcs.gla.ac.uk/scripts/global/send_message?glasgow-haskell-users@dcs.gla.ac.uk+message]
--

Hi,
 
 If you want to instal Perl if you go to www.activestate.com then you can dowload it and it instals with no trouble (and gives you help if you do come accross any problems). As to the #! bit, not sure yet, it's the reason I came accross your query in the first place. If you've come up with an answer could you let me know.
 
 Cheers
 
 James

--
From andreas.marth@daimlerchrysler.com Tue Feb 27 10:11:47 2001 Date: Tue, 27 Feb 2001 11:11:47 +0100 From: andreas.marth@daimlerchrysler.com andreas.marth@daimlerchrysler.com Subject: Problems when mounting c: to / 
--Boundary=_0.0_=0057440008329344
Content-Type: text/plain; charset=ISO-8859-1; name="MEMO 02/27/01 11:08:30"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

Hallo!

I just run "ghc -c -v Main.hs 2>&1| tee withMount.log" while c: was mou=
nted at=20
/ and "ghc -c -v Main.hs 2>&1| tee withoutMount.log" after unmounting i=
t.
I attached both Files.
because cygwin complains about a missing /tmp after mounting c: to / I =
have a=20
c:\tmp directory but ghc uses c:\Temp.
I think you should drop the mount c: at / because it is sure a problem =
you run=20
into when installing ghc.
I used a fresh WinNT (service pack 5 and inernet explorer 5 installed) =

installed cygwinat c:\cygwin, mounted c: to /, set PATH, SHELL, HOME, M=
AKE_MODE=20
and TMPDIR, restarted cygwin, created /tmp, installed ghc-4.08.2 at=20
c:\ghc\ghc-4.08.2, added it to PATH, restarted cygwin and went to my co=
de=20
directory. The rest is described above.

So until the next problem,
Andreas





rrt1001@cam.ac.uk@cl.cam.ac.uk am 24.02.2001 13:57:17
Gesendet von:	reuben.thomas@cl.cam.ac.uk
An:	Andreas Marth/FT/DCAG/DCX@WK-EMEA2
Kopie:	glasgow-haskell-users@haskell.org =20

Thema:	Re: Problems when mounting c: to /

> When installing ghc-4.08.2 (onWinNT) as described in the installation=

> instructions under c:, I couldn't get ghc to run.
> It invoked the C-preprozessor and stoped after cleaning.

I'm puzzled by this. Could you please send some sample output? Preferab=
ly
running ghc -v.

> When I unmount c:, it works fine. (If I install cygwin and dont mount=
 c: to /
> it works fine too.)
> But now //c/ is the normal win-root. That's not a problem for me, but=
 how=20
about
> ghc?

Well, if it works, I suppose it's not a problem...

> Does it depend in any way on c: mounted to /?

I'm not sure. I think it used to (because of using /tmp as the temporar=
y
directory). Now it's mostly for convenience.

> If not why do you want this mount (in the  installation instructions =
section
> 2.2.2.1)?

I think it used to be necessary, and I kept it in the instructions part=
ly
because I find it useful to be able to refer to the drive with Unix-sty=
le
paths, with the root directory being /, and partly because it worked, a=
nd
the less I change the instructions, the less likely I am to break somet=
hing.

--=20
http://sc3d.org/rrt/ =A6 wit, n.  educated insolence (Aristotle)




=

--Boundary=_0.0_=0057440008329344
Content-Type: application/octet-stream; name=withMount.log
Content-Transfer-Encoding: base64
Content-Disposition: attachment; filename=withMount.log

VGhlIEdsb3Jpb3VzIEdsYXNnb3cgSGFza2VsbCBDb21waWxhdGlvbiBTeXN0ZW0sIHZlcnNpb24g
NC4wOC4yDQoNCkVmZmVjdGl2ZSBjb21tYW5kIGxpbmU6IC1jIC12DQoNCkluZWZmZWN0aXZlIEMg
cHJlLXByb2Nlc3NvcjoNCgllY2hvICd7LSMgTElORSAxICJNYWluLmhzIiAtfScgPiBDOi9URU1Q
L2doYzEzMy5jcHAgJiYgY2F0IE1haW4uaHMgPj4gQzovVEVNUC9naGMxMzMuY3BwDQpkZWxldGlu
Zy4uLiBDOi9URU1QL2doYzEzMy5jcHANCg0Kcm0gLWYgQzovVEVNUC9naGMxMzMqDQo=

--Boundary=_0.0_=0057440008329344
Content-Type: application/octet-stream; name=withoutMount.log
Content-Transfer-Encoding: base64
Content-Disposition: attachment; filename=withoutMount.log
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==

--Boundary=_0.0_=0057440008329344--
From rrt1001@cam.ac.uk Tue Feb 27 10:25:11 2001 Date: Tue, 27 Feb 2001 10:25:11 +0000 (GMT) From: Reuben Thomas rrt1001@cam.ac.uk Subject: Problems when mounting c: to / 
> I think you should drop the mount c: at / because it is sure a problem you run
> into when installing ghc.

I think you're right. I really should reinstall Cygwin on my machine in the
default location, rather than at root. I'll change the instructions. Thanks.

-- 
http://sc3d.org/rrt/
"Reality is what refuses to disappear when you stop believing in it" -
Philip K. Dick
From josefs@cs.chalmers.se Wed Feb 28 09:46:58 2001 Date: Wed, 28 Feb 2001 10:46:58 +0100 (MET) From: Josef Svenningsson josefs@cs.chalmers.se Subject: Lightningspeed haskell 
Hi all.

Some days ago someone posted this url:
http://www.bagley.org/~doug/shootout/

which is a page benchmarking a number of different languages and
compilers where ghc is one of them. Some benchmarks lacked a haskell
versions (and some still do) and so I decided to fill in some of the gaps.

One benchmark turned out to give pretty remarkable results. It's the
producer/consumer benchmark. I suggest you all take a look at it. The
haskell version is six (SIX!!!) times faster than the c version. Hey,
what's going on here? I would really like to hear some comments from our
dear implementors.

It should be noted that synchronisation is achieved by using
slightly different kinds of primitives. But still... six times...

I lift my hat of for the ghc-implementors.

	/Josef
From simonpj@microsoft.com Wed Feb 28 10:16:11 2001 Date: Wed, 28 Feb 2001 02:16:11 -0800 From: Simon Peyton-Jones simonpj@microsoft.com Subject: Lightningspeed haskell 
Cool!  Thanks for doing this.

I guess GHC is fast because it's implementing lightweight
threads inside a single OS thread.

Simon

| -----Original Message-----
| From: Josef Svenningsson [mailto:josefs@cs.chalmers.se]
| Sent: 28 February 2001 09:47
| To: glasgow-haskell-users@haskell.org
| Subject: Lightningspeed haskell
| 
| 
| Hi all.
| 
| Some days ago someone posted this url:
| http://www.bagley.org/~doug/shootout/
| 
| which is a page benchmarking a number of different languages and
| compilers where ghc is one of them. Some benchmarks lacked a haskell
| versions (and some still do) and so I decided to fill in some 
| of the gaps.
| 
| One benchmark turned out to give pretty remarkable results. It's the
| producer/consumer benchmark. I suggest you all take a look at it. The
| haskell version is six (SIX!!!) times faster than the c version. Hey,
| what's going on here? I would really like to hear some 
| comments from our
| dear implementors.
| 
| It should be noted that synchronisation is achieved by using
| slightly different kinds of primitives. But still... six times...
| 
| I lift my hat of for the ghc-implementors.
| 
| 	/Josef
| 
| 
| _______________________________________________
| Glasgow-haskell-users mailing list
| Glasgow-haskell-users@haskell.org
| http://www.haskell.org/mailman/listinfo/glasgow-haskell-users
|
From simonmar@microsoft.com Wed Feb 28 12:03:20 2001 Date: Wed, 28 Feb 2001 04:03:20 -0800 From: Simon Marlow simonmar@microsoft.com Subject: Lightningspeed haskell 
> It should be noted that synchronisation is achieved by using
> slightly different kinds of primitives. But still... six times...

And it's about to get faster still, because CVars can now be implemented
with a single MVar instead of two.  The reason is that putMVar now
blocks on a full MVar rather than raising an exception.

But as Simon said, the main reason is surely that GHC is using
lightweight threads compared to C.  BTW, was this on Linux?  I'd be
interested to see the results on systems that have different threading
models, because Linux's threads implementation maps threads onto
processes (albeit lightweight kind of process, but still a process), so
the context switch overhead is going to be much higher than a threads
library which sits in a single process.

Cheers,
	Simon
From josefs@cs.chalmers.se Wed Feb 28 14:04:22 2001 Date: Wed, 28 Feb 2001 15:04:22 +0100 (MET) From: Josef Svenningsson josefs@cs.chalmers.se Subject: Lightningspeed haskell 
On Wed, 28 Feb 2001, Simon Marlow wrote:

> > It should be noted that synchronisation is achieved by using
> > slightly different kinds of primitives. But still... six times...
>
> And it's about to get faster still, because CVars can now be implemented
> with a single MVar instead of two.  The reason is that putMVar now
> blocks on a full MVar rather than raising an exception.
>
Cool. It'll be interesting to see the speedup.

> But as Simon said, the main reason is surely that GHC is using
> lightweight threads compared to C.  BTW, was this on Linux?  I'd be
> interested to see the results on systems that have different threading
> models, because Linux's threads implementation maps threads onto
> processes (albeit lightweight kind of process, but still a process), so
> the context switch overhead is going to be much higher than a threads
> library which sits in a single process.
>
I don't know what OS he's using. The homepage only says what kind of CPU
and amount of memory his computer has.

	/Josef
From jmaessen@mit.edu Wed Feb 28 15:11:01 2001 Date: Wed, 28 Feb 2001 10:11:01 -0500 From: Jan-Willem Maessen jmaessen@mit.edu Subject: Lightningspeed haskell 
Josef Svenningsson <josefs@cs.chalmers.se> wrote:
> One benchmark turned out to give pretty remarkable results. It's the
> producer/consumer benchmark. I suggest you all take a look at it. The
> haskell version is six (SIX!!!) times faster than the c version. Hey,
> what's going on here? I would really like to hear some comments from our
> dear implementors.

And Simon Peyton-Jones <simonpj@microsoft.com> replied:
> I guess GHC is fast because it's implementing lightweight
> threads inside a single OS thread.

Absolutely.  Good high-level thread support trumps anything provided
by the operating system.  I'd second Simon's guess that this was on
Linux or some other system where pthreads map one-to-one to OS
threads.  

Similar dramatic performance disparities have cropped up in the Java
community.  There are Java benchmarks which create thousands of
threads; most implementations slow to a crawl when this happens, as
the operating system collapses under the crushing load.  Those which
don't (IBM's Jalapeno, a research JVM, comes to mind) use this as a
*big* bragging point.

Unless you're mucking with thread priorities, mapping language-level
threads one-to-one to operating-system threads is a terrible idea.

-Jan-Willem Maessen
Eager Haskell project
jmaessen@mit.edu
From rrt1001@cam.ac.uk Wed Feb 28 19:17:58 2001 Date: Wed, 28 Feb 2001 19:17:58 +0000 (GMT) From: Reuben Thomas rrt1001@cam.ac.uk Subject: New InstallShield 
This one corrects a problem with the fix made in the last one that stopped
anything to do with stat() working (e.g. hFileSize, reading directories &c.
&c.).

Sorry about that. The fix is still fragile and temporary; I'm waiting for
the underlying mingwin problems to be fixed, hopefully in time for the GHC
5.0 release.

-- 
http://sc3d.org/rrt/ | humour, n.  unexpected recognition
From Ralf Hinze Thu Feb 1 10:00:53 2001 From: Ralf Hinze (Ralf Hinze) Date: Thu, 1 Feb 2001 11:00:53 +0100 (MET) Subject: Documentation Message-ID: <20010201100053.DB5C64546@mail.cs.uu.nl> Dear all, I just successfully installed GHC 4.08.1 and Happy 1.9 from source under SuSE linux. However, it seems that the doumentation is not generated. Do I have to take special actions (NB configure seems to be satisfied with the DocBook installation)? Thanks, Ralf From simonmar@microsoft.com Thu Feb 1 12:06:52 2001 From: simonmar@microsoft.com (Simon Marlow) Date: Thu, 1 Feb 2001 04:06:52 -0800 Subject: Documentation Message-ID: <9584A4A864BD8548932F2F88EB30D1C60171F416@TVP-MSG-01.europe.corp.microsoft.com> > > Yup, you have to explicitly ask for the documentation. Go to > > ghc/docs/set, and type 'make set'. The full documentation (combined > > Users' Guide and Library reference) will end up in set/ > after a while. > > That does not work. make answers > make: *** No rule to make target `set'. > Any ideas? Do I have to take special actions when > configuring? oops, sorry I meant 'make html'. You can also do 'make dvi', 'make ps', 'make pdf' etc. Cheers, Simon From simonmar@microsoft.com Thu Feb 1 11:14:14 2001 From: simonmar@microsoft.com (Simon Marlow) Date: Thu, 1 Feb 2001 03:14:14 -0800 Subject: Documentation Message-ID: <9584A4A864BD8548932F2F88EB30D1C60171F415@TVP-MSG-01.europe.corp.microsoft.com> > Dear all, > > I just successfully installed GHC 4.08.1 and Happy 1.9 > from source under SuSE linux. However, it seems that > the doumentation is not generated. Do I have to take > special actions (NB configure seems to be satisfied with > the DocBook installation)? > Yup, you have to explicitly ask for the documentation. Go to ghc/docs/set, and type 'make set'. The full documentation (combined Users' Guide and Library reference) will end up in set/ after a while. Cheers, Simon From Ralf Hinze Thu Feb 1 12:02:39 2001 From: Ralf Hinze (Ralf Hinze) Date: Thu, 1 Feb 2001 13:02:39 +0100 (MET) Subject: Documentation Message-ID: <20010201120239.9D3C1453D@mail.cs.uu.nl> > Yup, you have to explicitly ask for the documentation. Go to > ghc/docs/set, and type 'make set'. The full documentation (combined > Users' Guide and Library reference) will end up in set/ after a while. That does not work. make answers make: *** No rule to make target `set'. Any ideas? Do I have to take special actions when configuring? Ralf From rrt1001@cam.ac.uk Fri Feb 2 14:29:25 2001 From: rrt1001@cam.ac.uk (Reuben Thomas) Date: Fri, 2 Feb 2001 14:29:25 +0000 (GMT) Subject: Documentation In-Reply-To: <20010201120239.9D3C1453D@mail.cs.uu.nl> Message-ID: > That does not work. make answers > make: *** No rule to make target `set'. > Any ideas? Do I have to take special actions when > configuring? Simon meant "make html" and "make ps". "make set.html" and "make set.ps" also work, I think. -- http://sc3d.org/rrt/ | free, a. already paid for (Peyton Jones) From ger@Informatik.Uni-Bremen.DE Tue Feb 6 13:04:30 2001 From: ger@Informatik.Uni-Bremen.DE (George Russell) Date: Tue, 06 Feb 2001 14:04:30 +0100 Subject: Semi-applied datatypes in instance declarations Message-ID: <3A7FF65E.6FA75E18@informatik.uni-bremen.de> I apologise if this has been raised before, but the code I am writing now would look rather nicer if "partially applied type constructors" were permitted in instances. For example: class Event e where sync :: e a -> IO a data Event extraData a = blah blah . . . instance (context on extradata) => Event extradata where blah blah . . . Any chance of this? Or are there reasons why this would be wholly ridiculous? From v-julsew@microsoft.com Tue Feb 6 13:57:57 2001 From: v-julsew@microsoft.com (Julian Seward (Intl Vendor)) Date: Tue, 6 Feb 2001 05:57:57 -0800 Subject: ANNOUNCE: The Glasgow Haskell Compiler -- version 4.08.2 Message-ID: <68B95AA1648D1840AB0083CC63E57AD60F22D6@red-msg-06.redmond.corp.microsoft.com> The Glasgow Haskell Compiler -- version 4.08.2 ================================================ We are pleased to announce a new release of the Glasgow Haskell Compiler (GHC), version 4.08.2. The source distribution is freely available via the World-Wide Web and through anon. FTP, under a BSD-style license. See below for download details. Pre-built packages for Linux, FreeBSD, Solaris and Win32 are also available. Haskell is "the" standard lazy functional programming language; the current language version is Haskell 98, agreed in December 1998. GHC is a state-of-the-art optimising compiler for Haskell, generating good code for a variety of platforms. The distribution includes space and time profiling facilities, a large collection of libraries, and support for various language extensions, including concurrency, exceptions, and foreign language interfaces (C, C++, whatever). A wide variety of Haskell related resources (tutorials, libraries, specifications, documentation, compilers, interpreters, references, contact information, links to research groups) are available from the Haskell home page at http://www.haskell.org/ GHC's Web page lives at http://www.haskell.org/ghc/ + What's new in 4.08.2 ======================= No new features, just a few minor bug fixes. + What's new in 4.08 ===================== This should be a stable release. There have been many enhancements since 4.06, and shed-loads of bug-fixes (one shed (imperial) ~ one ton (US)). There are the following changes - New profiling subsystem, based on cost-centre stacks. - Working x86 native code generator: now it works properly, runs about twice as fast as compiling via C, and is on a par for run-time speed (except in FP-intensive programs). - Implicit parameters (i.e. dynamic scoping without the pain). - DEPRECATED pragma for marking obsolescent interfaces. - In the wake of hslibs, a new package system for libraries. -package should now be used instead of -syslib. - Result type signatures work. - Many tiresome long-standing bugs and problems (e.g. the trace problem) have been fixed. - Many error messages have been made more helpful and/or accurate. For full details see the release notes: http://www.haskell.org/ghc/docs/4.08/users_guide/release-4-08.html + Important Info For Win32 users ================================= ALERT: For reasons as yet not understood, as of January 2001 Cygwin's mingw package has a problem that causes GHC-compiled binaries to be built incorrectly. The fix is to ensure that your mingw package is dated 20001111 (you can set this in the Cygwin installer). See the installation guide for more details. + Mailing lists ================ We run mailing lists for GHC users and bug reports; to subscribe, use the web interfaces at http://www.haskell.org/mailman/listinfo/glasgow-haskell-users http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs There are several other haskell and ghc-related mailing lists on www.haskell.org; for the full list, see http://www.haskell.org/mailman/listinfo/ Please send bug reports about GHC to glasgow-haskell-bugs@haskell.org; GHC users hang out on glasgow-haskell-users@haskell.org. Bleeding edge CVS users party on cvs-ghc@haskell.org. + On-line GHC-related resources ================================ Relevant URLs on the World-Wide Web: GHC home page http://www.haskell.org/ghc/ Haskell home page http://www.haskell.org/ comp.lang.functional FAQ http://www.cs.nott.ac.uk/~gmh/faq.html + How to get it ================ The easy way is to go to the WWW page, which should be self-explanatory: http://www.haskell.org/ghc/ Once you have the distribution, please follow the pointers in the README file to find all of the documentation about this release. NB: preserve modification times when un-tarring the files (no `m' option for tar, please)! + System requirements ====================== To compile the sources, you need a machine with 32+MB memory, GNU C (`gcc'), `perl' plus a version of GHC installed (3.02 at least). This release is known to work on the following platforms: * i386-unknown-{linux,freebsd,netbsd,cygwin32,mingw32} * sparc-sun-solaris2 * hppa1.1-hp-hpux{9,10} Ports to the following platforms should be relatively easy (for a wunderhacker), but haven't been tested due to lack of time/hardware: * i386-unknown-solaris2 * alpha-dec-osf{2,3} * mips-sgi-irix{5,6} * {rs6000,powerpc}-ibm-aix The builder's guide included in distribution gives a complete run-down of what ports work; an on-line version can be found at http://www.haskell.org/ghc/docs/4.08/building/building-guide.html From mk167280@students.mimuw.edu.pl Tue Feb 6 14:29:31 2001 From: mk167280@students.mimuw.edu.pl (Marcin 'Qrczak' Kowalczyk) Date: Tue, 6 Feb 2001 15:29:31 +0100 (CET) Subject: Semi-applied datatypes in instance declarations In-Reply-To: <3A7FF65E.6FA75E18@informatik.uni-bremen.de> Message-ID: On Tue, 6 Feb 2001, George Russell wrote: > I apologise if this has been raised before, but the code I am > writing now would look rather nicer if "partially applied > type constructors" were permitted in instances. They are. For example monads. -- Marcin 'Qrczak' Kowalczyk From ger@Informatik.Uni-Bremen.DE Tue Feb 6 14:34:28 2001 From: ger@Informatik.Uni-Bremen.DE (George Russell) Date: Tue, 06 Feb 2001 15:34:28 +0100 Subject: Semi-applied datatypes in instance declarations References: Message-ID: <3A800B74.A4B5F597@informatik.uni-bremen.de> Marcin 'Qrczak' Kowalczyk wrote: > > On Tue, 6 Feb 2001, George Russell wrote: > > > I apologise if this has been raised before, but the code I am > > writing now would look rather nicer if "partially applied > > type constructors" were permitted in instances. > > They are. For example monads. Yes you are right. I'm sorry for wasting your time - I made a mistake in constructing my test case. From Malcolm.Wallace@cs.york.ac.uk Tue Feb 6 14:31:02 2001 From: Malcolm.Wallace@cs.york.ac.uk (Malcolm Wallace) Date: Tue, 6 Feb 2001 14:31:02 +0000 Subject: GHC-4.08.2 Message-ID: The download links on http://www.mirror.ac.uk/sites/www.haskell.org/ghc/ for the new GHC 4.08.2 have somewhat (ahem) strange referents. (The RedHat 6 binary package gives me a picture of Haskell B Curry, the RH6 profiling package gives me a .dvi file of the dynamic semantics of the language, RH7 source RPM gives me an Emacs hugs-mode....) The original haskell.org site seems ok, just mirror.ac.uk is broken. Malcolm From Malcolm.Wallace@cs.york.ac.uk Tue Feb 6 14:33:22 2001 From: Malcolm.Wallace@cs.york.ac.uk (Malcolm Wallace) Date: Tue, 6 Feb 2001 14:33:22 +0000 Subject: GHC-4.08.2 In-Reply-To: Message-ID: > The original haskell.org site seems ok, just mirror.ac.uk is broken. Oops, retract that. The RedHat 6 packages all seem to be 4.08.1, not the advertised 4.08.2. Regards, Malcolm From simonmar@microsoft.com Tue Feb 6 16:21:41 2001 From: simonmar@microsoft.com (Simon Marlow) Date: Tue, 6 Feb 2001 08:21:41 -0800 Subject: GHC on Sourceforge Message-ID: <9584A4A864BD8548932F2F88EB30D1C61157C7@TVP-MSG-01.europe.corp.microsoft.com> Hi folks, A while ago we opened a SourceForge project for GHC, mainly to make use of the bug tracking facilities. It's been working quite happily for some time now, so I think it's time to start using the bug tracking system for all bug reports. To report a bug, go to http://sourceforge.net/projects/ghc/ and follow the "bugs" link, then "submit a bug". Please include all the usual information you'd normally give with a bug report on the mailing list. We also plan to start using the task manager (as a kind of wish list / job adverts for GHC), and perhaps some of the other cool gadgets that SourceForge provides. Cheers, Simon From chak@cse.unsw.edu.au Wed Feb 7 02:57:23 2001 From: chak@cse.unsw.edu.au (Manuel M. T. Chakravarty) Date: Wed, 07 Feb 2001 13:57:23 +1100 Subject: GHC-4.08.2 In-Reply-To: References: Message-ID: <20010207135723U.chak@cse.unsw.edu.au> Malcolm Wallace wrote, > > The original haskell.org site seems ok, just mirror.ac.uk is broken. > > Oops, retract that. The RedHat 6 packages all seem to be 4.08.1, not > the advertised 4.08.2. There shouldn't be any links to RedHat 6 packages. I don't have a RedHat 6.x machine anymore, and so, can't make any packages for it. If you have a RedHat 6.x box with an older version of GHC already installed, just run rpm --rebuild ghc-4.08.2-1.src.rpm This should (with a little waiting ;-) produce binary packages for 4.08.2 running on your flavour of RedHat. Cheers, Manuel From rrt1001@cam.ac.uk Wed Feb 7 10:27:11 2001 From: rrt1001@cam.ac.uk (Reuben Thomas) Date: Wed, 7 Feb 2001 10:27:11 +0000 (GMT) Subject: GHC-4.08.2 In-Reply-To: <20010207135723U.chak@cse.unsw.edu.au> Message-ID: > Malcolm Wallace wrote, > > > > The original haskell.org site seems ok, just mirror.ac.uk is broken. > > > > Oops, retract that. The RedHat 6 packages all seem to be 4.08.1, not > > the advertised 4.08.2. > > There shouldn't be any links to RedHat 6 packages. We're providing the links for 4.08.1 since that's all we have, and a lot of people still seem to use RH6. > packages for it. If you have a RedHat 6.x box with an older > version of GHC already installed, just run > > rpm --rebuild ghc-4.08.2-1.src.rpm This will only work if you install RPM 4, no? I couldn't find a way to make RPM 4 produce v3 RPMs, either. But never mind, I installed RPM 4, installed the source RPM, downgraded again to RPM 3, and am now building RH 6 RPMs, which I'll make available. One problem I had: the spec file doesn't have a dependency on the jade package, but it needs it for some DSSSL files without which the docs won't build. -- http://sc3d.org/rrt/ | RSA, n. safety in numbers From chak@cse.unsw.edu.au Thu Feb 8 06:09:19 2001 From: chak@cse.unsw.edu.au (Manuel M. T. Chakravarty) Date: Thu, 08 Feb 2001 17:09:19 +1100 Subject: GHC-4.08.2 In-Reply-To: References: <20010207135723U.chak@cse.unsw.edu.au> Message-ID: <20010208170919Y.chak@cse.unsw.edu.au> Reuben Thomas wrote, > > packages for it. If you have a RedHat 6.x box with an older > > version of GHC already installed, just run > > > > rpm --rebuild ghc-4.08.2-1.src.rpm > > This will only work if you install RPM 4, no? I couldn't find a way to make > RPM 4 produce v3 RPMs, either. But never mind, I installed RPM 4, installed > the source RPM, downgraded again to RPM 3, and am now building RH 6 RPMs, > which I'll make available. True. I didn't know whether the src.rpm format also changed. > One problem I had: the spec file doesn't have a dependency on the jade > package, but it needs it for some DSSSL files without which the docs won't > build. Right - there should be a build dependency. Cheers, Manuel From qrczak@knm.org.pl Thu Feb 8 07:36:38 2001 From: qrczak@knm.org.pl (Marcin 'Qrczak' Kowalczyk) Date: 8 Feb 2001 07:36:38 GMT Subject: GHC-4.08.2 References: <20010207135723U.chak@cse.unsw.edu.au> <20010208170919Y.chak@cse.unsw.edu.au> Message-ID: Thu, 08 Feb 2001 17:09:19 +1100, Manuel M. T. Chakravarty pisze: > > This will only work if you install RPM 4, no? > > True. I didn't know whether the src.rpm format also changed. AFAIK it did not change. -- __("< Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/ \__/ ^^ SYGNATURA ZASTĘPCZA QRCZAK From chak@cse.unsw.edu.au Sun Feb 11 12:20:07 2001 From: chak@cse.unsw.edu.au (Manuel M. T. Chakravarty) Date: Sun, 11 Feb 2001 23:20:07 +1100 Subject: ANN: C->Haskell 0.8.1 Message-ID: <20010211232007N.chak@cse.unsw.edu.au> I am pleased to announce the availability of version 0.8.1 of the interface generator C->Haskell. It works with the current stable release series 4.08.x of GHC as well as the current development series 4.11. For both versions of GHC, it supports the *same* FFI library that GHC natively only supports in the development version 4.11 and which constitutes the result of the work of the FFI Task Force over the last couple of months. The interface specification of the library is online available at http://www.cse.unsw.edu.au/~chak/haskell/c2hs/docu/c2hs-4.html For more information on C->Haskell and for downloading, see http://www.cse.unsw.edu.au/~chak/haskell/c2hs/ The main feature in this release is the new FFI library. It should allow users of the GHC stable series to use the new FFI library interface, which we hope to keep stable from now on. An update of the C->Haskell tool proper is being worked at. Happy Hacking, Manuel From qrczak@knm.org.pl Sun Feb 11 13:36:31 2001 From: qrczak@knm.org.pl (Marcin 'Qrczak' Kowalczyk) Date: 11 Feb 2001 13:36:31 GMT Subject: ANN: C->Haskell 0.8.1 References: <20010211232007N.chak@cse.unsw.edu.au> Message-ID: Sun, 11 Feb 2001 23:20:07 +1100, Manuel M. T. Chakravarty pisze: > I am pleased to announce the availability of version 0.8.1 > of the interface generator C->Haskell. Wow! A problem: c2hs/gen/CInfo.lhs imports class Storable from module C2HSConfig, which does not export it. Also, the compilation fails (on ghc-4.11): /usr/local/bin/ghc -c -syslib lang -syslib posix -O -recomp -fno-warn-incomplete-patterns -i. -fglasgow-exts C2HSMarsh.hs C2HSMarsh.hs:43: Bad interface file: ./Ptr.hi does not exist and similarly for other files. This is because there exists Ptr.hs in this directory - but it should not be taken from there, because the right version is in ghc's own libraries. I don't know how to solve such problem in general, except by splitting modules into many directories playing with -i options (that's why QForeign had to use different module names for modules it brings which may or may not come with the compiler, except some modules not present in nhc98 which are handled by -i options). -- __("< Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/ \__/ ^^ SYGNATURA ZASTĘPCZA QRCZAK From qrczak@knm.org.pl Sun Feb 11 13:38:18 2001 From: qrczak@knm.org.pl (Marcin 'Qrczak' Kowalczyk) Date: 11 Feb 2001 13:38:18 GMT Subject: ANN: C->Haskell 0.8.1 References: <20010211232007N.chak@cse.unsw.edu.au> Message-ID: 11 Feb 2001 13:36:31 GMT, Marcin 'Qrczak' Kowalczyk pisze: > A problem: [...] These problems apply to cvs' HEAD. -- __("< Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/ \__/ ^^ SYGNATURA ZASTĘPCZA QRCZAK From chak@cse.unsw.edu.au Mon Feb 12 03:07:01 2001 From: chak@cse.unsw.edu.au (Manuel M. T. Chakravarty) Date: Mon, 12 Feb 2001 14:07:01 +1100 Subject: ANN: C->Haskell 0.8.1 In-Reply-To: References: <20010211232007N.chak@cse.unsw.edu.au> Message-ID: <20010212140701N.chak@cse.unsw.edu.au> qrczak@knm.org.pl (Marcin 'Qrczak' Kowalczyk) wrote, > Sun, 11 Feb 2001 23:20:07 +1100, Manuel M. T. Chakravarty pisze: > > > I am pleased to announce the availability of version 0.8.1 > > of the interface generator C->Haskell. > > Wow! > > A problem: c2hs/gen/CInfo.lhs imports class Storable from module > C2HSConfig, which does not export it. > > Also, the compilation fails (on ghc-4.11): > /usr/local/bin/ghc -c -syslib lang -syslib posix -O -recomp -fno-warn-incomplete-patterns -i. -fglasgow-exts C2HSMarsh.hs > > C2HSMarsh.hs:43: > Bad interface file: ./Ptr.hi > does not exist > > and similarly for other files. This is because there exists Ptr.hs > in this directory - but it should not be taken from there, because > the right version is in ghc's own libraries. This is very strange, because I have tested this and it worked for me. In fact, if you look at the Makefile in the c2hs/lib/ directory, then you will see that the `depend' target moves these troublesome files into a subdirectory before compilation (called ghc411hackdir). Is it possible that you used the same build tree to compile with 4.08 first and then re-compile with 4.11? This won't work, because of the hack mentioned above. In fact, I think, this is a bug in ghc 4.11 - I have filled it as bug #131631 in the bug tracker. Cheers, Manuel From info@tips2you.co.uk Mon Feb 12 07:51:46 2001 From: info@tips2you.co.uk (* TIPS 4 U *) Date: Mon, 12 Feb 2001 07:51:46 Subject: * TIPS 4 U * Message-ID: This is the first time we have advertised this service anywhere as most of our clients/participants have been referred to us by word of mouth or are friends of the family who have benefited from our information. We feel it only fair, bearing in mind the enormous profit generating potential here, that we share this information with others and enable everyone to enjoy a better standard of living. This is why we have kept the fees for providing such information as low as Ł10. Our horses generate a steady profit week in week out and month in and month out. We have never shown a month of loss. Many tipsters nowadays have runs of good luck in accordance with seasonal changes. Some do well in winter others in summer, some on the flat and others on the jumps. We have maintained a consistent profit stream, which ensures that we do not have to charge exorbitant prices in order to offset any losses. We are sure that you will find us one of the most competitive tipping services anywhere. The service is launched for members only now. For more information e-mail us at info@tips2you.co.uk ************************************************************************************************* To be removed please state remove in the header and your details will be deleted immediately. ************************************************************************************************** From qrczak@knm.org.pl Mon Feb 12 14:48:09 2001 From: qrczak@knm.org.pl (Marcin 'Qrczak' Kowalczyk) Date: 12 Feb 2001 14:48:09 GMT Subject: ANN: C->Haskell 0.8.1 References: <20010211232007N.chak@cse.unsw.edu.au> <20010212140701N.chak@cse.unsw.edu.au> Message-ID: Mon, 12 Feb 2001 14:07:01 +1100, Manuel M. T. Chakravarty pisze: > This is very strange, because I have tested this and it > worked for me. It could be my fault: I ran ./configure before updating it with autoconf, and perhaps did not clean afterwards. > In fact, if you look at the Makefile in the c2hs/lib/ directory, > then you will see that the `depend' target moves these troublesome > files into a subdirectory before compilation So they will be downloaded each time from the CVS... Perhaps it would be better to store them under different names and make symlinks if needed? -- __("< Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/ \__/ ^^ SYGNATURA ZASTĘPCZA QRCZAK From dongen@cs.ucc.ie Mon Feb 12 21:05:15 2001 From: dongen@cs.ucc.ie (Marc van Dongen) Date: Mon, 12 Feb 2001 21:05:15 +0000 Subject: lreadline Message-ID: <20010212210515.R13263@cs.ucc.ie> Hello all, When linking, ghc-4.08.1 fails due to a ld: fatal: library -lreadline: not found error. Is this a ghc error or should I get that library from somewhere else? Thanks in advance. Marc van Dongen -- Marc van Dongen, CS Dept | phone: +353 21 4903578 University College Cork, NUIC | Fax: +353 21 4903113 College Road, Cork, Ireland | Email: dongen@cs.ucc.ie From chak@cse.unsw.edu.au Tue Feb 13 07:46:58 2001 From: chak@cse.unsw.edu.au (Manuel M. T. Chakravarty) Date: Tue, 13 Feb 2001 18:46:58 +1100 Subject: ANN: C->Haskell 0.8.1 In-Reply-To: References: <20010212140701N.chak@cse.unsw.edu.au> Message-ID: <20010213184658T.chak@cse.unsw.edu.au> qrczak@knm.org.pl (Marcin 'Qrczak' Kowalczyk) wrote, > Mon, 12 Feb 2001 14:07:01 +1100, Manuel M. T. Chakravarty pisze: > > > In fact, if you look at the Makefile in the c2hs/lib/ directory, > > then you will see that the `depend' target moves these troublesome > > files into a subdirectory before compilation > > So they will be downloaded each time from the CVS... Perhaps it > would be better to store them under different names and make symlinks > if needed? They are only moved away in the build tree (where they are only sym links anyway). if you clean the whole tree and start again with configuration and build, you should be fine. Manuel From andreas.marth@daimlerchrysler.com Tue Feb 13 12:01:19 2001 From: andreas.marth@daimlerchrysler.com (andreas.marth@daimlerchrysler.com) Date: Tue, 13 Feb 2001 13:01:19 +0100 Subject: mingw-20001111-1.tar.gz still needed for cygwin WinNT? Message-ID: <0057440006754438000002L482*@MHS> Hi is the mingw-20001111-1.tar.gz still needed? How do I install it wit= h the=20 actual cygwin (just downloaded)? I put it in the ../latest/mingw/ directory but can not chose it in the = installation program. Any help appreciated. Thanks, Andreas= From andreas.marth@daimlerchrysler.com Tue Feb 13 13:08:39 2001 From: andreas.marth@daimlerchrysler.com (andreas.marth@daimlerchrysler.com) Date: Tue, 13 Feb 2001 14:08:39 +0100 Subject: Antwort: Re: mingw-20001111-1.tar.gz still needed for cygwin WinNT? Message-ID: <0057440006766422000002L422*@MHS> It cycles throug 2 alternatives: "20010130-1" and "20001225-1". Even the choices what to downlod is to this mingw packages restricted. = So I see only two alternatives: 1.) Use 20010130-1 or 20001225-1 2.) Use an older cygwin I have not found an old cygwin1.1 at http://sources.redhat.com/cygwin/.= Any comments? Bye, Andreas rrt1001@cam.ac.uk@cl.cam.ac.uk am 13.02.2001 13:21:34 Gesendet von: reuben.thomas@cl.cam.ac.uk An: Andreas Marth/FT/DCAG/DCX@WK-EMEA2 Kopie: =20 Thema: Re: mingw-20001111-1.tar.gz still needed for cygwin WinNT? > Hi is the mingw-20001111-1.tar.gz still needed? Yes, as far as I know. > How do I install it with the > actual cygwin (just downloaded)? You should be able to select it in the installer even without downloadi= ng it: if you click on the package date, it should cycle through several c= hoices. --=20 http://sc3d.org/rrt/ "Reality is what refuses to disappear when you stop believing in it" - Philip K. Dick = From christian.lescher@icn.siemens.de Tue Feb 13 16:03:03 2001 From: christian.lescher@icn.siemens.de (Lescher Christian) Date: Tue, 13 Feb 2001 17:03:03 +0100 Subject: mingw-20001111-1.tar.gz still needed for cygwin WinNT? Message-ID: <353063C297A9D3118F760008C791E269011D0755@MCHH263E> I had the same problem, and I solved it by changing the setup.ini file after downloading & putting the old mingw archive in latest/mingw: @ mingw version: 20001111-1 install: latest/mingw/mingw-20001111-1.tar.gz 130113 source: latest/mingw/mingw-20001111-1-src.tar.gz 121498 Cheers, Christian From christian.lescher@icn.siemens.de Tue Feb 13 16:11:14 2001 From: christian.lescher@icn.siemens.de (Lescher Christian) Date: Tue, 13 Feb 2001 17:11:14 +0100 Subject: Message: "DEFAULT_TMPDIR has not been set to anything useful!" Message-ID: <353063C297A9D3118F760008C791E269011D0756@MCHH263E> After installing the newest version of GHC on WinNT/Cygwin, I get the message "DEFAULT_TMPDIR has not been set to anything useful!" when executing GHC (however, compilation is ok.) Setting DEFAULT_TMPDIR to "c:/tmp" didn't change anything. What may be wrong with my settings? Christian From andreas.marth@daimlerchrysler.com Tue Feb 13 20:06:41 2001 From: andreas.marth@daimlerchrysler.com (andreas.marth@daimlerchrysler.com) Date: Tue, 13 Feb 2001 21:06:41 +0100 Subject: questions about the installation instructions for WinNT and ghc-4.08x Message-ID: <0057440006816774000002L442*@MHS> I have some question relating the intsallation of ghc-4.08x on a WinNT/= cygwin=20 system. In the installation instruction is under 2.2.2.1 paragraph "Here's how = to=20 install Cygwin" written: After installation, start up a Cygwin shell and issue the following=20 command:=20 mount -f c: / assuming you installed Cygwin at C:\cygwin; otherwise change the driv= e=20 and directory as appropriate. (So it definitely says / is at c: not at c:\cygwin. But later, when set= ting the=20 environment variables it seems like it is assumed that / is at c:\cygwin.) Is it right to mount / at c: or should it be c:\cygwin? In "2.2.2.2 Environment variables" is a table that says I should add C:= \usr\bin=20 to PATH. That directory does not exist. the directory C:\cygwin\usr\bin exists (= assuming=20 you installed Cygwin at C:\cygwin) but is empty. The C:\cygwin\bin directory seems to be the one meant. Should I add C:\cygwin\bin to PATH? The same table says that SHELL should be C:/usr/bin/bash. (Observe the = '/'=20 signs instead of '\'.) But bash is in C:\cygwin\bin (assuming you installed Cygwin at C:\cygwi= n). Should I set SHELL to C:\cygwin\bin\bash (does it matter if I write=20 C:\cygwin\bin\bash or C:/cygwin/bin/bash?)? Okay that are my questions at the moment. Any help would be appreciated= . PS:Is there still something else to observe than putting ghc into scope= of PATH? = From rrt1001@cam.ac.uk Wed Feb 14 11:07:56 2001 From: rrt1001@cam.ac.uk (Reuben Thomas) Date: Wed, 14 Feb 2001 11:07:56 +0000 (GMT) Subject: questions about the installation instructions for WinNT and ghc-4.08x In-Reply-To: <0057440006816774000002L442*@MHS> Message-ID: > In the installation instruction is under 2.2.2.1 paragraph "Here's how to > install Cygwin" written: > After installation, start up a Cygwin shell and issue the following > command: > mount -f c: / > assuming you installed Cygwin at C:\cygwin; otherwise change the drive > and directory as appropriate. > (So it definitely says / is at c: not at c:\cygwin. But later, when setting the > environment variables it seems like it is assumed > that / is at c:\cygwin.) There are two things going on here: 1. You're mounting C: at / so you can refer to the whole drive with UNIX-style paths. 2. Cygwin automatically mounts parts of its directory tree under /usr. > In "2.2.2.2 Environment variables" is a table that says I should add C:\usr\bin > to PATH. > That directory does not exist. the directory C:\cygwin\usr\bin exists (assuming > you installed Cygwin at C:\cygwin) but is empty. > The C:\cygwin\bin directory seems to be the one meant. > Should I add C:\cygwin\bin to PATH? No. Add /usr/bin. Cygwin mounts \cygwin\usr\bin at /usr/bin. > The same table says that SHELL should be C:/usr/bin/bash. (Observe the '/' > signs instead of '\'.) > But bash is in C:\cygwin\bin (assuming you installed Cygwin at C:\cygwin). > Should I set SHELL to C:\cygwin\bin\bash (does it matter if I write > C:\cygwin\bin\bash or C:/cygwin/bin/bash?)? As above. I think these instructions have worked for other people; my own system is in a slightly different configuration (which for various reasons we used to recommend). -- http://sc3d.org/rrt/ | humour, n. unexpected recognition From qrczak@knm.org.pl Wed Feb 14 23:14:24 2001 From: qrczak@knm.org.pl (Marcin 'Qrczak' Kowalczyk) Date: 14 Feb 2001 23:14:24 GMT Subject: ANN: C->Haskell 0.8.1 References: <20010211232007N.chak@cse.unsw.edu.au> <20010212140701N.chak@cse.unsw.edu.au> Message-ID: 12 Feb 2001 14:48:09 GMT, Marcin 'Qrczak' Kowalczyk pisze: > It could be my fault: I ran ./configure before updating it with > autoconf, and perhaps did not clean afterwards. Ok, it was my fault. But the error in CInfo is still there. And C2HS does not export most of cast*Ptr functions, freeHaskellFunPtr and newStablePtr, which leads to errors in compilig gtk+hs. And GtkCList (from the CVS version of gtk+hs) does not import C2HS but uses newStablePtr etc. And ghc version check in gtk+hs fails for 4.11. -- __("< Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/ \__/ ^^ SYGNATURA ZASTĘPCZA QRCZAK From trebla@vex.net Thu Feb 15 04:35:36 2001 From: trebla@vex.net (Albert Lai) Date: 14 Feb 2001 23:35:36 -0500 Subject: Confused about enabling parallel ghc Message-ID: <4u8zn8va07.fsf@vex.net> I installed ghc 4.08.2 binaries for sparc solaris (and those for intel linux, too), then tried compiling a sample program (the fibonacci one in the user guide) with ghc -parallel. The compiler complained: Could not find interface file for `Prelude' in the directories ./*.hi /u/trebla/lib/ghc-4.08.2/imports/lang/*.mp_hi /u/trebla/lib/ghc-4.08.2/imports/concurrent/*.mp_hi /u/trebla/lib/ghc-4.08.2/imports/lang/*.mp_hi /u/trebla/lib/ghc-4.08.2/imports/concurrent/*.mp_hi /u/trebla/lib/ghc-4.08.2/imports/std/*.mp_hi (and similarly for Parallel.) Does it mean I must build ghc myself from source? The user guide promises a lot of binary bundles such as par. Ever since 4.0x, the ghc download page ceases to mention them. Where are they? From chak@cse.unsw.edu.au Thu Feb 15 07:44:56 2001 From: chak@cse.unsw.edu.au (Manuel M. T. Chakravarty) Date: Thu, 15 Feb 2001 18:44:56 +1100 Subject: ANN: C->Haskell 0.8.1 In-Reply-To: References: <20010212140701N.chak@cse.unsw.edu.au> Message-ID: <20010215184456L.chak@cse.unsw.edu.au> qrczak@knm.org.pl (Marcin 'Qrczak' Kowalczyk) wrote, > 12 Feb 2001 14:48:09 GMT, Marcin 'Qrczak' Kowalczyk pisze: > > > It could be my fault: I ran ./configure before updating it with > > autoconf, and perhaps did not clean afterwards. > > Ok, it was my fault. But the error in CInfo is still > there. That's strange. Could you try this with either the release of 0.8.2 that I put up yesterday or with the CVS version again? I don't have any trouble with CInfo with both 4.08 and 4.11. > And > C2HS does not export most of cast*Ptr functions, freeHaskellFunPtr > and newStablePtr, which leads to errors in compilig gtk+hs. And > GtkCList (from the CVS version of gtk+hs) does not import C2HS but > uses newStablePtr etc. And ghc version check in gtk+hs fails for 4.11. As noted on the Web page, 0.8.1 didn't work with gtk+hs. You need c2hs 0.8.2 and the new version 0.10.2 of gtk+hs, which I released yesterday. (Alternatively, the versions of both packages in CVS should work, too.) Cheers, Manuel From qrczak@knm.org.pl Thu Feb 15 08:20:00 2001 From: qrczak@knm.org.pl (Marcin 'Qrczak' Kowalczyk) Date: 15 Feb 2001 08:20:00 GMT Subject: ANN: C->Haskell 0.8.1 References: <20010212140701N.chak@cse.unsw.edu.au> <20010215184456L.chak@cse.unsw.edu.au> Message-ID: Thu, 15 Feb 2001 18:44:56 +1100, Manuel M. T. Chakravarty pisze: > > But the error in CInfo is still there. > > That's strange. Could you try this with either the release of > 0.8.2 that I put up yesterday or with the CVS version again? Hmm, it works now. There are two files called C2HSConfig.hs (in c2hs/lib and c2hs/toplevel). I don't know why the wrong one was being imported. > As noted on the Web page, 0.8.1 didn't work with gtk+hs. > You need c2hs 0.8.2 and the new version 0.10.2 of gtk+hs, > which I released yesterday. (Alternatively, the versions of > both packages in CVS should work, too.) Both are current versions from CVS, with version numbers as you say. Does C2HSDeprecated export newStablePtr and freeHaskellFunPtr? Currently it does not, but GtkCList assumes it does. Does C2HS export castPtrToFunPtr? Currently it does not, but GMarsh assumes it does. -- __("< Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/ \__/ ^^ SYGNATURA ZASTĘPCZA QRCZAK From simonmar@microsoft.com Thu Feb 15 11:00:55 2001 From: simonmar@microsoft.com (Simon Marlow) Date: Thu, 15 Feb 2001 03:00:55 -0800 Subject: Confused about enabling parallel ghc Message-ID: <9584A4A864BD8548932F2F88EB30D1C60171F479@TVP-MSG-01.europe.corp.microsoft.com> > I installed ghc 4.08.2 binaries for sparc solaris (and those for intel > linux, too), then tried compiling a sample program (the fibonacci one > in the user guide) with ghc -parallel. The compiler complained: > > Could not find interface file for `Prelude' > in the directories ./*.hi > /u/trebla/lib/ghc-4.08.2/imports/lang/*.mp_hi > > /u/trebla/lib/ghc-4.08.2/imports/concurrent/*.mp_hi > /u/trebla/lib/ghc-4.08.2/imports/lang/*.mp_hi > > /u/trebla/lib/ghc-4.08.2/imports/concurrent/*.mp_hi > /u/trebla/lib/ghc-4.08.2/imports/std/*.mp_hi > > (and similarly for Parallel.) > > Does it mean I must build ghc myself from source? > > The user guide promises a lot of binary bundles such as par. Ever > since 4.0x, the ghc download page ceases to mention them. Where > are they? This just means that Parallel support isn't included in the binary bundle you have. In fact, parallel support is developed and distributed separately. Contact the GPH (Glasgow Parallel Haskell) folks for more information: http://www.cee.hw.ac.uk/~dsg/gph/ Cheers, Simon From qrczak@knm.org.pl Thu Feb 15 15:35:45 2001 From: qrczak@knm.org.pl (Marcin 'Qrczak' Kowalczyk) Date: 15 Feb 2001 15:35:45 GMT Subject: ANN: C->Haskell 0.8.1 References: <20010212140701N.chak@cse.unsw.edu.au> <20010215184456L.chak@cse.unsw.edu.au> Message-ID: 15 Feb 2001 08:20:00 GMT, Marcin 'Qrczak' Kowalczyk pisze: > Does C2HSDeprecated export newStablePtr and freeHaskellFunPtr? > Currently it does not, but GtkCList assumes it does. > > Does C2HS export castPtrToFunPtr? Currently it does not, but GMarsh > assumes it does. Now I see: these functions in module C2HS are present in the released c2hs-0.8.2, but not in the CVS. I guess you have not committed changes. gtk+hs' examples which compile with the present interface don't link on ghc-4.08.2, because of the ghc's bug in handling newtypes in foreign exports which references rts_mkPtr. This is because you made Addr a type synonym for Ptr (). It can be "fixed" without making Addr incompatible with Ptr (which I guess is needed because c2hs generates Addr and code uses Ptr) by something like this: module PtrHack where import qualified Addr newtype Addr a = Ptr Addr.Addr module C2HSSomething where import qualified PtrHack type Ptr = PtrHack.Addr type Addr = Ptr () This ensures that the "real" name of the Ptr type is Addr. I'll try this hack for QForeign to see if it can reduce the amount of #ifdefs for broken compilers. It applies to newtypes in arguments and results of functions in foreign export and foreign export dynamic in ghc-4.08*. It should be applied to CInt etc. too, to let them work there. I can provide my own CTypes for ghc-4.08*, but at least I will get rid of many of those stupid #ifdefs. Unfortunately it does not help for Ptr in the result of foreign export dynamic (ghc-4.08) nor in the argument of foreign import dynamic (ghc-4.08*), where newtypes don't work. This means that gtk+hs does not compile on 4.08 because of Ptr () (spelled as Addr) in the result of foreign export dynamic. Here is which ghc versions are broken in which ways: | newtypes work in foreign... | | | | export | export | import | import | | | stat.& dyn.| dynamic | stat.& dyn.| dynamic | label | | (function) | (pointer) | (function) | (pointer) | (pointer) | ------------+------------+-----------+------------+-----------+-----------+ ghc-4.08 | hacked | no | yes | no | no | ghc-4.08.1 | hacked | yes | yes | no | yes | ghc-4.08.2 | hacked | yes | yes | no | yes | ghc-4.11 | yes | yes | yes | yes | yes | "Hacked" means that they work as long as the type name after expanding type synonyms is recognized by the rts (and there is no way to #include something in stubs I think). -- __("< Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/ \__/ ^^ SYGNATURA ZASTĘPCZA QRCZAK From simonmar@microsoft.com Thu Feb 15 17:31:36 2001 From: simonmar@microsoft.com (Simon Marlow) Date: Thu, 15 Feb 2001 09:31:36 -0800 Subject: Haskell threads & pipes & UNIX processes Message-ID: <9584A4A864BD8548932F2F88EB30D1C61157E2@TVP-MSG-01.europe.corp.microsoft.com> [ moved to glasgow-haskell-users@haskell.org... ] > I need to compress the output of my Hakell program. To avoid > the creation > of huge files, I want to compress before writing by means of gzip or > bzip2. However, this seems to be quite involved. > > I decided to run the compressor with > > runProcess :: FilePath -- Command > -> [String] -- Arguments > -> Maybe [(String, String)] -- > Environment(Nothing -> Inherited) > -> Maybe FilePath -- Working > directory (Nothing -> inherited) > -> Maybe Handle -- stdin (Nothing > -> inherited) > -> Maybe Handle -- stdout (Nothing > -> inherited) > -> Maybe Handle -- stderr (Nothing > -> inherited) > -> IO () > > First I wanted to use Posix.createPipe to connect to the > compressor but > Posix.createPipe returns a Fd. So this does not fit. Aha, you want the (unadvertised) function PosixIO.fdToHandle :: Fd -> Handle > I ended up with using Posix.createNamedPipe. This creates a named pipe > that can be openend with IO.openFile (which again yields the > Handle needed > for runProcess). > > At this point, it turned out that the pipe has to be opened > for reading > before it can be opened for writing (ghc-4.08.1). This seems > to be a bug. > (In a shell, the order does not matter; the processes are suspended > as expected.) We've had some problems with named pipes - check the archives of glasgow-haskell-bugs. It seems certain operating systems disagree on the semantics of a non-blocking open of a named pipe. I don't recommend using named pipes if you want portable code. > Then, as expected, the problem occured that writing to the pipe and > compressing cannot be sequenced. If the producer is started first, the > producer blocks because there is no consumer. If the consumer > is started > first, the consumer blocks because there is no input. > > So I played around with threads. If both consumer and producer are > executed as threads, the program terminates immediately without any > output. There seems to be no need to start executing any of > the threads. In GHC, the main program terminates as soon as the main thread terminates. It doesn't wait for any child threads to terminate - if you want this behaviour, you can program it using MVars. We'll be happy to look at any of the other problems you mentioned: please some code demonstrating the problem to glasgow-haskell-bugs@haskell.org. Cheers, Simon From chak@cse.unsw.edu.au Fri Feb 16 07:42:08 2001 From: chak@cse.unsw.edu.au (Manuel M. T. Chakravarty) Date: Fri, 16 Feb 2001 18:42:08 +1100 Subject: ANN: C->Haskell 0.8.1 In-Reply-To: References: <20010215184456L.chak@cse.unsw.edu.au> Message-ID: <20010216184208D.chak@cse.unsw.edu.au> qrczak@knm.org.pl (Marcin 'Qrczak' Kowalczyk) wrote, > 15 Feb 2001 08:20:00 GMT, Marcin 'Qrczak' Kowalczyk pisze: > > > Does C2HSDeprecated export newStablePtr and freeHaskellFunPtr? > > Currently it does not, but GtkCList assumes it does. > > > > Does C2HS export castPtrToFunPtr? Currently it does not, but GMarsh > > assumes it does. > > Now I see: these functions in module C2HS are present in the released > c2hs-0.8.2, but not in the CVS. I guess you have not committed changes. Oops, you are right, I did only partially commit my last changes. I am sorry for that. Now, everything should be checked in. > gtk+hs' examples which compile with the present interface don't link on > ghc-4.08.2, because of the ghc's bug in handling newtypes in foreign > exports which references rts_mkPtr. This is because you made Addr a > type synonym for Ptr (). > > It can be "fixed" without making Addr incompatible with Ptr (which > I guess is needed because c2hs generates Addr and code uses Ptr) > by something like this: > module PtrHack where > import qualified Addr > newtype Addr a = Ptr Addr.Addr > module C2HSSomething where > import qualified PtrHack > type Ptr = PtrHack.Addr > type Addr = Ptr () > This ensures that the "real" name of the Ptr type is Addr. > > I'll try this hack for QForeign to see if it can reduce the amount > of #ifdefs for broken compilers. It applies to newtypes in arguments > and results of functions in foreign export and foreign export dynamic > in ghc-4.08*. > > It should be applied to CInt etc. too, to let them work there. I can > provide my own CTypes for ghc-4.08*, but at least I will get rid of > many of those stupid #ifdefs. > > Unfortunately it does not help for Ptr in the result of foreign export > dynamic (ghc-4.08) nor in the argument of foreign import dynamic > (ghc-4.08*), where newtypes don't work. This means that gtk+hs does > not compile on 4.08 because of Ptr () (spelled as Addr) in the result > of foreign export dynamic. I thought that I had fixed all this for Gtk+HS. (In fact, all Gtk+HS examples are running fine with GHC 4.08 on my machine.) Have a look at the file gtk+hs/gtk/ghcRtsAux.c. It defines rts_mkPtr in a somewhat nasty way, but it works :-) It's a bit like your hack, but on the C level. Maybe you forgot to run autoconf and ./configure after your last cvs update for Gtk+HS? > Here is which ghc versions are broken in which ways: > > | newtypes work in foreign... | > | | > | export | export | import | import | | > | stat.& dyn.| dynamic | stat.& dyn.| dynamic | label | > | (function) | (pointer) | (function) | (pointer) | (pointer) | > ------------+------------+-----------+------------+-----------+-----------+ > ghc-4.08 | hacked | no | yes | no | no | > ghc-4.08.1 | hacked | yes | yes | no | yes | > ghc-4.08.2 | hacked | yes | yes | no | yes | > ghc-4.11 | yes | yes | yes | yes | yes | > > "Hacked" means that they work as long as the type name after expanding > type synonyms is recognized by the rts (and there is no way to #include > something in stubs I think). Yes, I agree. I also tried to get something #include'ed in stubs, but failed. That would have been the easiest solution. Anyway, thanks for checking that stuff. Cheers, Manuel PS: With the current Gtk+HS source in CVS, all Gtk+HS examples as well as the iHaskell library and its three examples should now all work again. I tested it all on my machine. From qrczak@knm.org.pl Fri Feb 16 09:21:51 2001 From: qrczak@knm.org.pl (Marcin 'Qrczak' Kowalczyk) Date: 16 Feb 2001 09:21:51 GMT Subject: ANN: C->Haskell 0.8.1 References: <20010215184456L.chak@cse.unsw.edu.au> <20010216184208D.chak@cse.unsw.edu.au> Message-ID: Fri, 16 Feb 2001 18:42:08 +1100, Manuel M. T. Chakravarty pisze: > Now, everything should be checked in. Seems OK, thanks. ghc411hack_dir does not work for me again: ghc -M in build/ghc4/chs/lib can't find NewStablePtr.hs (because there is only NewStablePtr.hs.in) and the whole 'make depend' there fails, without removing files conflicing with ghc' libraries. The failure of 'make depend' is ignored and finally 'make' there fails to compile C2HS.hs. > I thought that I had fixed all this for Gtk+HS. (In fact, > all Gtk+HS examples are running fine with GHC 4.08 on my > machine.) Have a look at the file gtk+hs/gtk/ghcRtsAux.c. > It defines rts_mkPtr in a somewhat nasty way, but it works :-) I've seen the hack, but on another box freshly installed ghc-4.08.2, c2hs from tarball and gtk+hs from tarball did not work (linker can't find rts_mkPtr). I must see if ghcRtsAux.c is compied at all there. > PS: With the current Gtk+HS source in CVS, all Gtk+HS > examples as well as the iHaskell library and its three > examples should now all work again. I tested it all on > my machine. ghc version check fails on 4.11 (it should be lexicographic comparison of version number components, not conjunction of independent comparisons). Checking for buggy readXXXOffAddr (for ghc-4.03..4.06) is now unnecessary as gtk+hs requires ghc-4.08.1 anyway. -- __("< Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/ \__/ ^^ SYGNATURA ZASTĘPCZA QRCZAK From qrczak@knm.org.pl Fri Feb 16 09:23:28 2001 From: qrczak@knm.org.pl (Marcin 'Qrczak' Kowalczyk) Date: 16 Feb 2001 09:23:28 GMT Subject: ANN: C->Haskell 0.8.1 References: <20010215184456L.chak@cse.unsw.edu.au> <20010216184208D.chak@cse.unsw.edu.au> Message-ID: 16 Feb 2001 09:21:51 GMT, Marcin 'Qrczak' Kowalczyk pisze: > ghc411hack_dir does not work for me again: Ah, I see: haven't run autoconf in c2hs subdirectory. Now it will hopefully work. -- __("< Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/ \__/ ^^ SYGNATURA ZASTĘPCZA QRCZAK From qrczak@knm.org.pl Fri Feb 16 09:40:02 2001 From: qrczak@knm.org.pl (Marcin 'Qrczak' Kowalczyk) Date: 16 Feb 2001 09:40:02 GMT Subject: ANN: C->Haskell 0.8.1 References: <20010215184456L.chak@cse.unsw.edu.au> <20010216184208D.chak@cse.unsw.edu.au> Message-ID: Fri, 16 Feb 2001 18:42:08 +1100, Manuel M. T. Chakravarty pisze: > PS: With the current Gtk+HS source in CVS, all Gtk+HS > examples as well as the iHaskell library and its three > examples should now all work again. I tested it all on > my machine. gtk+hs under ghc-4.11 needs -package lang in HCFLAGS in configure.in. -- __("< Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/ \__/ ^^ SYGNATURA ZASTĘPCZA QRCZAK From qrczak@knm.org.pl Fri Feb 16 19:12:14 2001 From: qrczak@knm.org.pl (Marcin 'Qrczak' Kowalczyk) Date: 16 Feb 2001 19:12:14 GMT Subject: ANN: C->Haskell 0.8.1 References: <20010215184456L.chak@cse.unsw.edu.au> <20010216184208D.chak@cse.unsw.edu.au> Message-ID: Fri, 16 Feb 2001 18:42:08 +1100, Manuel M. T. Chakravarty pisze: > PS: With the current Gtk+HS source in CVS, all Gtk+HS > examples as well as the iHaskell library and its three > examples should now all work again. They compile but they don't run correctly on ghc-4.11. It's because GMarsh.writeCharOffAddr does not simulate Addr.writeCharOffAddr when Char is wide, because instance Storable Char treats Char as wide. You need to cast to CChar, Word8, or Int8 (be careful with sign extension). -- __("< Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/ \__/ ^^ SYGNATURA ZASTĘPCZA QRCZAK From chak@cse.unsw.edu.au Sun Feb 18 02:25:24 2001 From: chak@cse.unsw.edu.au (Manuel M. T. Chakravarty) Date: Sun, 18 Feb 2001 13:25:24 +1100 Subject: ANN: C->Haskell 0.8.1 In-Reply-To: References: <20010216184208D.chak@cse.unsw.edu.au> Message-ID: <20010218132524J.chak@cse.unsw.edu.au> qrczak@knm.org.pl (Marcin 'Qrczak' Kowalczyk) wrote, > > ghc411hack_dir does not work for me again: > > Ah, I see: haven't run autoconf in c2hs subdirectory. Now it will > hopefully work. Automating the invocation fo autoconf is on the TODO list, but I didn't get around to it yet. Manuel From chak@cse.unsw.edu.au Mon Feb 19 04:53:41 2001 From: chak@cse.unsw.edu.au (Manuel M. T. Chakravarty) Date: Mon, 19 Feb 2001 15:53:41 +1100 Subject: ANN: C->Haskell 0.8.1 In-Reply-To: References: <20010216184208D.chak@cse.unsw.edu.au> Message-ID: <20010219155341J.chak@cse.unsw.edu.au> qrczak@knm.org.pl (Marcin 'Qrczak' Kowalczyk) wrote, > Fri, 16 Feb 2001 18:42:08 +1100, Manuel M. T. Chakravarty pisze: > > > I thought that I had fixed all this for Gtk+HS. (In fact, > > all Gtk+HS examples are running fine with GHC 4.08 on my > > machine.) Have a look at the file gtk+hs/gtk/ghcRtsAux.c. > > It defines rts_mkPtr in a somewhat nasty way, but it works :-) > > I've seen the hack, but on another box freshly installed ghc-4.08.2, > c2hs from tarball and gtk+hs from tarball did not work (linker can't > find rts_mkPtr). I must see if ghcRtsAux.c is compied at all there. Strange - meanwhile, I have put release 0.10.4 out. Among other things, it finally has an `install' target. > > PS: With the current Gtk+HS source in CVS, all Gtk+HS > > examples as well as the iHaskell library and its three > > examples should now all work again. I tested it all on > > my machine. > > ghc version check fails on 4.11 (it should be lexicographic comparison > of version number components, not conjunction of independent > comparisons). > > Checking for buggy readXXXOffAddr (for ghc-4.03..4.06) is now > unnecessary as gtk+hs requires ghc-4.08.1 anyway. Ok, I fixed these. > > PS: With the current Gtk+HS source in CVS, all Gtk+HS > > examples as well as the iHaskell library and its three > > examples should now all work again. > > They compile but they don't run correctly on ghc-4.11. It's because > GMarsh.writeCharOffAddr does not simulate Addr.writeCharOffAddr when > Char is wide, because instance Storable Char treats Char as wide. > You need to cast to CChar, Word8, or Int8 (be careful with sign > extension). `castCCharToChar' and `castCharToCChar' should be sufficient here, right? I have added them in the 0.10.4 release. Thanks, Manuel From rrt1001@cam.ac.uk Mon Feb 19 15:35:24 2001 From: rrt1001@cam.ac.uk (Reuben Thomas) Date: Mon, 19 Feb 2001 15:35:24 +0000 (GMT) Subject: [Bug #133086] complete failure (fwd) Message-ID: [This sort of thing is probably better on the list, because it doesn't really isolate a bug] > Date: Mon, 19 Feb 2001 07:25:29 -0800 > From: noreply@sourceforge.net > To: noreply@sourceforge.net, noreply@sourceforge.net, cvs-ghc@haskell.org > Subject: [Bug #133086] complete failure > Bug #133086, was updated on 2001-Feb-19 07:25 > Here is a current snapshot of the bug. > Details: I have tried to install ghc-4.08.2 under Windows NT > and (as usal with the last few releases) failed. Sorry to hear you've had difficulties with recent releases; for most people things seem to have improved. > I get an immediate warning that DEFAULT_TMPDIR is not set to anything > useful. Check you've got the most recent InstallShield (same date as the current one); there's a bug in previous versions. > Although files will compile they run without producing > any output (even Hello World). This is the mingwin problem. > By the way, > I note that mingw from Cygwin has no version dated > 20001111 but the only version before this January > is listed as 20001225 (yes really Christmas day). There is a 20001111, but since it's now the last version but one it's not available directly from the installer; you have to get it by hand. Sorry about that. It looks as though the latest version (from this year) has cured the problem, so you could try using that instead. As I write I'm building GHC 4.08.2 with the new mingwin package installed; if that works, I'll ship a new InstallShield shortly. From qrczak@knm.org.pl Mon Feb 19 19:02:29 2001 From: qrczak@knm.org.pl (Marcin 'Qrczak' Kowalczyk) Date: 19 Feb 2001 19:02:29 GMT Subject: ANN: C->Haskell 0.8.1 References: <20010216184208D.chak@cse.unsw.edu.au> <20010219155341J.chak@cse.unsw.edu.au> Message-ID: Mon, 19 Feb 2001 15:53:41 +1100, Manuel M. T. Chakravarty pisze: > Strange - meanwhile, I have put release 0.10.4 out. > Among other things, it finally has an `install' target. Thanks, everything compiles and mostly works. Examples still behave strangely: rngtest goes crazy when changing the number of digits, all rngtest's widgets are displayed with some black border unless the mouse cursor is over them, the scale in ih/examples/Counter does not work, threadtest does not work etc. > `castCCharToChar' and `castCharToCChar' should be sufficient > here, right? Yes. Same in C2HSDeprecated. -- __("< Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/ \__/ ^^ SYGNATURA ZASTĘPCZA QRCZAK From janna@cs.chalmers.se Tue Feb 20 10:31:11 2001 From: janna@cs.chalmers.se (Janna Khegai) Date: Tue, 20 Feb 2001 11:31:11 +0100 Subject: Exposing Haskell functions with String arguments in dll Message-ID: <001201c09b28$44d9dea0$5c00140a@alman> This is a multi-part message in MIME format. ------=_NextPart_000_000F_01C09B30.A07791A0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Hi, I want to expose the following Haskell function in dll to call it from = Java application:=20 =20 trans: String -> IO String=20 I read that I shoud use tools like GreenCard, HDirect or some = marshalling libraries, but not sure, which way is better. If anyone has experience in this, I will be very grateful for an = example, which shows the steps of bilding such dll. Thanks in advance, Janna Khegai. ------=_NextPart_000_000F_01C09B30.A07791A0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
Hi,
 
I want to expose the=20 following Haskell function in dll to call it from Java=20 application: 
 
trans: String -> IO = String 
 
I read that I shoud use tools like = GreenCard,=20 HDirect or some marshalling libraries, but not sure, which way is=20 better.
If anyone has experience in this, = I will be=20 very grateful for an example, which shows the steps of bilding such=20 dll.
 
Thanks in advance, Janna=20 Khegai.
------=_NextPart_000_000F_01C09B30.A07791A0-- From bgregor@BUPHY.bu.edu Tue Feb 20 13:25:46 2001 From: bgregor@BUPHY.bu.edu (Brian Gregor) Date: Tue, 20 Feb 2001 08:25:46 -0500 Subject: language shootout Message-ID: There's a page with microbenchmarks for a large array of languages, including Haskell: http://www.bagley.org/~doug/shootout/ and please note his disclaimer :-) : http://www.lib.uchicago.edu/keith/crisis/disclaimer.html Anyway, many of the categories are lacking a Haskell program. I've sent in 2. I thought that this could also serve as a nice little collection of examples for style and the use of some of the libraries in HSlibs for those of us (particularly me :) ) who are still learning the language. Any takers? -brian From byron.hale@einfo.com Wed Feb 21 09:35:09 2001 From: byron.hale@einfo.com (Byron Hale) Date: Wed, 21 Feb 2001 01:35:09 -0800 Subject: Installation of mingw-20001111-1.tar.gz Message-ID: <4.3.2.7.2.20010221012736.040b2e68@einfo.com> Hi, I'm trying to install mingw-20001111-1.tar.gz for ghc4.08.2. The instructions on Win2K. The Web instructions seem a bit cavalier, but perhaps it is just my ignorance. ..."directory called mingw, place that in a directory called latest, then run the Cygwin installer "... Are these intended to be sub directories of Cygwin's "bin" directory, or what? Is the Cygwin "installer" "bin/install.exe?" If so, install with what options? Best Regards, Byron Hale byron.hale@einfo.com From rrt1001@cam.ac.uk Wed Feb 21 10:07:59 2001 From: rrt1001@cam.ac.uk (Reuben Thomas) Date: Wed, 21 Feb 2001 10:07:59 +0000 (GMT) Subject: Installation of mingw-20001111-1.tar.gz In-Reply-To: <4.3.2.7.2.20010221012736.040b2e68@einfo.com> Message-ID: > I'm trying to install mingw-20001111-1.tar.gz for ghc4.08.2. The > instructions on Win2K. The Web instructions seem a bit cavalier, but > perhaps it is just my ignorance. Neither, really, it's just that we can't document our entire world (documenting our system is hard enough). We are forced just to skate over the externals. If you don't know about one or more of the pieces we rely on, you should read their documentation (if it exists). However, within an hour or so I should have uploaded a fixed InstallShield that works around what does indeed turn out to be a bug in mingwin, so you won't need to bother with all this junk. -- http://sc3d.org/rrt/ | Careful Cyclists Approaching From Right From rrt1001@cam.ac.uk Wed Feb 21 10:10:45 2001 From: rrt1001@cam.ac.uk (Reuben Thomas) Date: Wed, 21 Feb 2001 10:10:45 +0000 (GMT) Subject: Installation of mingw-20001111-1.tar.gz In-Reply-To: <4.3.2.7.2.20010221012736.040b2e68@einfo.com> Message-ID: > Are these intended to be sub directories of Cygwin's "bin" directory, or what? > Is the Cygwin "installer" "bin/install.exe?" If so, install with what options? ...and just to clear these up: the directories I mention are not supposed to be subdirectories of anywhere in particular; it doesn't matter where you put them. The Cygwin installer is the program you run to install Cygwin (setup.exe, as downloaded from http://cygwin.com/ or one of its aliases). -- http://sc3d.org/rrt/ | certain, a. insufficiently analysed From rrt1001@cam.ac.uk Wed Feb 21 15:59:45 2001 From: rrt1001@cam.ac.uk (Reuben Thomas) Date: Wed, 21 Feb 2001 15:59:45 +0000 (GMT) Subject: New GHC InstallShield: mingwin problem fixed Message-ID: I have just uploaded a new GHC InstallShield. It fixes the recent problem with needing a particular version of the mingw package. Unfortunately the situation here is still unstable (roll on Cygwin 1.2), so it may break again, but it seems to work for now. -- http://sc3d.org/rrt/ | certain, a. insufficiently analysed From byron.hale@einfo.com Wed Feb 21 21:30:18 2001 From: byron.hale@einfo.com (Byron Hale) Date: Wed, 21 Feb 2001 13:30:18 -0800 Subject: Installation of mingw-20001111-1.tar.gz In-Reply-To: References: <4.3.2.7.2.20010221012736.040b2e68@einfo.com> Message-ID: <4.3.2.7.2.20010221132911.023bcd88@einfo.com> Thanks, Reuben. You're very prompt. Byron byron.hale@einfo.com At 10:07 AM 2/21/2001 +0000, you wrote: > > I'm trying to install mingw-20001111-1.tar.gz for ghc4.08.2. The > > instructions on Win2K. The Web instructions seem a bit cavalier, but > > perhaps it is just my ignorance. > >Neither, really, it's just that we can't document our entire world >(documenting our system is hard enough). We are forced just to skate over >the externals. If you don't know about one or more of the pieces we rely on, >you should read their documentation (if it exists). > >However, within an hour or so I should have uploaded a fixed InstallShield >that works around what does indeed turn out to be a bug in mingwin, so you >won't need to bother with all this junk. > >-- >http://sc3d.org/rrt/ | Careful Cyclists Approaching From Right > > > >_______________________________________________ >Glasgow-haskell-users mailing list >Glasgow-haskell-users@haskell.org >http://www.haskell.org/mailman/listinfo/glasgow-haskell-users From qrczak@knm.org.pl Fri Feb 23 00:23:24 2001 From: qrczak@knm.org.pl (Marcin 'Qrczak' Kowalczyk) Date: 23 Feb 2001 00:23:24 GMT Subject: --add-package Message-ID: In practice installation script of an external package does --del-package before --add-package, because if it was installed before, --add-package would cause an error. Wouldn't be reasonable then to let --add-package overwrite any existing package of that name? There is no real safety, because the package is explicitly deleted anyway. -- __("< Marcin Kowalczyk * qrczak@knm.org.pl http://qrczak.ids.net.pl/ \__/ ^^ SYGNATURA ZASTĘPCZA QRCZAK From simonmar@microsoft.com Fri Feb 23 10:50:27 2001 From: simonmar@microsoft.com (Simon Marlow) Date: Fri, 23 Feb 2001 10:50:27 -0000 Subject: --add-package Message-ID: <9584A4A864BD8548932F2F88EB30D1C60171F4AD@TVP-MSG-01.europe.corp.microsoft.com> > In practice installation script of an external package does > --del-package before --add-package, because if it was=20 > installed before, > --add-package would cause an error. >=20 > Wouldn't be reasonable then to let --add-package overwrite=20 > any existing > package of that name? There is no real safety, because the package > is explicitly deleted anyway. I don't think I agree. Installation of an external package shouldn't be doing --delete-package first: that should be left to the user (or rpm -e, or whatever). If you want to install a new package that replaces an existing one, you have to remove the existing one first. Cheers, Simon From Dominic.J.Steinitz@BritishAirways.com Fri Feb 23 11:09:34 2001 From: Dominic.J.Steinitz@BritishAirways.com (Steinitz, Dominic J) Date: 23 Feb 2001 11:09:34 Z Subject: Sockets on Windows Message-ID: <"0504B3A9644EE033*/c=GB/admd=ATTMAIL/prmd=BA/o=British Airways PLC/ou=CORPLN1/s=Steinitz/g=Dominic/i=J/"@MHS> --PART.BOUNDARY.bawhub1.4eaf.3a9645c6.0001 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: Quoted-Printable Content-Disposition: inline I am was trying out sockets. These worked fine under linux. Under windows I= get the following error. Does anyone know what it means and how I can fix = it? I can telnet from within the cygwin window so I assume tcp/ip is runnin= g. Thanks, Dominic. administrator@DEFAULT /cygdrive/d/dom/home/socketTest $ ./server1 8000 Fail: does not exist Action: getProtocolByName Reason: no such protocol entry ---------------------------------------------------------------------------= ---------------------- 21st century air travel http://www.britishairways.com --PART.BOUNDARY.bawhub1.4eaf.3a9645c6.0001 Content-Type: application/octet-stream; name="ATTKHVFA" Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="ATTKHVFA" Content-Description: ATTKHVFA FTBP-Modification-Date: 23 Feb 2001 11:09:00 Z FTBP-Object-Size: 1390 -------------------------------------------------- -- Simple Server -- Chris Reade, Nov 2000 -- Based on example by Hermann Oliveira Rodrigues -------------------------------------------------- module Main (main) where import System import IO import Time import Socket main :: IO () main = do prog <- getProgName args <- getArgs if (length args /= 1) then do putStrLn ("Use: " ++ prog ++ " ") exitWith (ExitFailure (-1)) else return () let port = read (args !! 0) :: Int in server (PortNumber (mkPortNumber port)) -- The server function creates a socket to listen on the port and -- loops to service requests continuously server :: PortID -> IO () server port = do socket <- listenOn port -- Fique em `loop' atendendo requisicoes. let loop = do (sh, host, portid) <- accept socket procRequest sh loop in loop -- The procRequest function deals with an individual request procRequest :: Handle -> IO () procRequest hd = do clock <- getClockTime calendar <- toCalendarTime clock hPutStrLn hd (calendarTimeToString calendar) hFlush hd hClose hd --PART.BOUNDARY.bawhub1.4eaf.3a9645c6.0001-- From rrt1001@cam.ac.uk Fri Feb 23 11:41:56 2001 From: rrt1001@cam.ac.uk (Reuben Thomas) Date: Fri, 23 Feb 2001 11:41:56 +0000 (GMT) Subject: Sockets on Windows In-Reply-To: <"0504B3A9644EE033*/c=GB/admd=ATTMAIL/prmd=BA/o=British Airways PLC/ou=CORPLN1/s=Steinitz/g=Dominic/i=J/"@MHS> Message-ID: > administrator@DEFAULT /cygdrive/d/dom/home/socketTest > $ ./server1 8000 > > Fail: does not exist > Action: getProtocolByName > Reason: no such protocol entry Just from looking at the manpage for getprotobyname, I surmise that the problem is that you don't have an /etc/protocols file (or equivalent) under Windows. I'm not sure what's supposed to happen (or for that matter, what's supposed to work). Look in hslibs/net/BSD.lhs; I couldn't work out what was going on in a minute; perhaps someone else knows. The other thing I couldn't work out was where getProtocolByName was being called in the first place, but presumably it's to find the protocol number of tcp. -- http://sc3d.org/rrt/ | Slow Pedestrian Crossing From mwr22@cam.ac.uk Fri Feb 23 12:19:41 2001 From: mwr22@cam.ac.uk (Matthew Richards) Date: Fri, 23 Feb 2001 12:19:41 -0000 Subject: Sockets on Windows In-Reply-To: <"0504B3A9644EE033*/c=GB/admd=ATTMAIL/prmd=BA/o=British Airways PLC/ou=CORPLN1/s=Steinitz/g=Dominic/i=J/"@MHS> Message-ID: You need to wrap up the use of sockets in a "withSocketsDo", as in: > server port = withSocketsDo $ > do socket <- listenOn port > -- etc I seem to recall the documentation says you always need to use that, so I don't know why it works without on Linux :-) Hope this helps, Matthew -----Original Message----- From: glasgow-haskell-users-admin@haskell.org [mailto:glasgow-haskell-users-admin@haskell.org]On Behalf Of Steinitz, Dominic J Sent: Friday, February 23, 2001 11:10 AM To: glasgow-haskell-users Subject: Sockets on Windows I am was trying out sockets. These worked fine under linux. Under windows I get the following error. Does anyone know what it means and how I can fix it? I can telnet from within the cygwin window so I assume tcp/ip is running. Thanks, Dominic. administrator@DEFAULT /cygdrive/d/dom/home/socketTest $ ./server1 8000 Fail: does not exist Action: getProtocolByName Reason: no such protocol entry ---------------------------------------------------------------------------- --------------------- 21st century air travel http://www.britishairways.com From mk167280@students.mimuw.edu.pl Fri Feb 23 12:37:32 2001 From: mk167280@students.mimuw.edu.pl (Marcin 'Qrczak' Kowalczyk) Date: Fri, 23 Feb 2001 13:37:32 +0100 (CET) Subject: --add-package In-Reply-To: <9584A4A864BD8548932F2F88EB30D1C60171F4AD@TVP-MSG-01.europe.corp.microsoft.com> Message-ID: On Fri, 23 Feb 2001, Simon Marlow wrote: > I don't think I agree. Installation of an external package shouldn't be > doing --delete-package first: that should be left to the user (or rpm > -e, or whatever). rpm is a different story. I would expect 'make install' to be idempotent. With your scheme one has to write 'make uninstall install'. -- Marcin 'Qrczak' Kowalczyk From simonmar@microsoft.com Fri Feb 23 14:07:22 2001 From: simonmar@microsoft.com (Simon Marlow) Date: Fri, 23 Feb 2001 06:07:22 -0800 Subject: --add-package Message-ID: <9584A4A864BD8548932F2F88EB30D1C61157ED@TVP-MSG-01.europe.corp.microsoft.com> > On Fri, 23 Feb 2001, Simon Marlow wrote: > > > I don't think I agree. Installation of an external package > shouldn't be > > doing --delete-package first: that should be left to the > user (or rpm > > -e, or whatever). > > rpm is a different story. I would expect 'make install' to be > idempotent. > With your scheme one has to write 'make uninstall install'. ghc --add-package behaves just like rpm --install, which seems reasonable to me. Why should it behave differently? If you want 'make install' to be idempotent (which I'm not sure is desirable anyway), then you will indeed have to ghc --delete-package first, but then I hope you're going to print a big warning and wait for the user to press any key first :) Cheers, Simon From andreas.marth@daimlerchrysler.com Fri Feb 23 19:53:59 2001 From: andreas.marth@daimlerchrysler.com (andreas.marth@daimlerchrysler.com) Date: Fri, 23 Feb 2001 20:53:59 +0100 Subject: Problems when mounting c: to / Message-ID: <0057440008066266000002L462*@MHS> Hallo everybody! When installing ghc-4.08.2 (onWinNT) as described in the installation=20 instructions under c:, I couldn't get ghc to run.=20 It invoked the C-preprozessor and stoped after cleaning. When I unmount c:, it works fine. (If I install cygwin and dont mount c= : to /=20 it works fine too.)=20 But now //c/ is the normal win-root. That's not a problem for me, but h= ow about=20 ghc?=20 Does it depend in any way on c: mounted to /? If not why do you want this mount (in the installation instructions se= ction=20 2.2.2.1)? Thanks, Andreas PS: Thanks Reuben for the new install shield and a very big thanks to = all of=20 you who have their parts in getting the profiling to run under WinNT= From rrt1001@cam.ac.uk Sat Feb 24 12:59:04 2001 From: rrt1001@cam.ac.uk (Reuben Thomas) Date: Sat, 24 Feb 2001 12:59:04 +0000 (GMT) Subject: Problems when mounting c: to / In-Reply-To: <0057440008066266000002L462*@MHS> Message-ID: > When installing ghc-4.08.2 (onWinNT) as described in the installation > instructions under c:, I couldn't get ghc to run. > It invoked the C-preprozessor and stoped after cleaning. I'm puzzled by this. Could you please send some sample output? Preferably running ghc -v. > When I unmount c:, it works fine. (If I install cygwin and dont mount c: to / > it works fine too.) > But now //c/ is the normal win-root. That's not a problem for me, but how about > ghc? Well, if it works, I suppose it's not a problem... > Does it depend in any way on c: mounted to /? I'm not sure. I think it used to (because of using /tmp as the temporary directory). Now it's mostly for convenience. > If not why do you want this mount (in the installation instructions section > 2.2.2.1)? I think it used to be necessary, and I kept it in the instructions partly because I find it useful to be able to refer to the drive with Unix-style paths, with the root directory being /, and partly because it worked, and the less I change the instructions, the less likely I am to break something. -- http://sc3d.org/rrt/ | wit, n. educated insolence (Aristotle) From stolz@I2.Informatik.RWTH-Aachen.DE Sat Feb 24 18:41:48 2001 From: stolz@I2.Informatik.RWTH-Aachen.DE (Volker Stolz) Date: Sat, 24 Feb 2001 19:41:48 +0100 Subject: Sockets on Windows In-Reply-To: References: <"0504B3A9644EE033* Message-ID: <20010224194148.A17985@agamemnon.informatik.rwth-aachen.de> In local.glasgow-haskell-users, you wrote: >You need to wrap up the use of sockets in a "withSocketsDo", as in: >I seem to recall the documentation says you always need to use that, so I >don't know why it works without on Linux :-) No, that notice only applies to Windows. -- \usepackage[latin1]{inputenc}! Volker Stolz * stolz@i2.informatik.rwth-aachen.de * PGP + S/MIME From James Lelyveld Mon Feb 26 11:32:55 2001 From: James Lelyveld (James Lelyveld ) Date: Mon, 26 Feb 2001 11:32:55 GMT Subject: message Message-ID: <200102261132.LAA13329@www.dcs.gla.ac.uk> [This message is sent through a WWW-Email gateway.] [The authenticity of the sender can not be validated.] [Message sent from the following machine - 62.172.105.102 ] [after accessing this URL http://www.dcs.gla.ac.uk/scripts/global/send_message?glasgow-haskell-users@dcs.gla.ac.uk+message] -- Hi, If you want to instal Perl if you go to www.activestate.com then you can dowload it and it instals with no trouble (and gives you help if you do come accross any problems). As to the #! bit, not sure yet, it's the reason I came accross your query in the first place. If you've come up with an answer could you let me know. Cheers James -- From andreas.marth@daimlerchrysler.com Tue Feb 27 10:11:47 2001 From: andreas.marth@daimlerchrysler.com (andreas.marth@daimlerchrysler.com) Date: Tue, 27 Feb 2001 11:11:47 +0100 Subject: Problems when mounting c: to / Message-ID: <0057440008329344000002L442*@MHS> --Boundary=_0.0_=0057440008329344 Content-Type: text/plain; charset=ISO-8859-1; name="MEMO 02/27/01 11:08:30" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Hallo! I just run "ghc -c -v Main.hs 2>&1| tee withMount.log" while c: was mou= nted at=20 / and "ghc -c -v Main.hs 2>&1| tee withoutMount.log" after unmounting i= t. I attached both Files. because cygwin complains about a missing /tmp after mounting c: to / I = have a=20 c:\tmp directory but ghc uses c:\Temp. I think you should drop the mount c: at / because it is sure a problem = you run=20 into when installing ghc. I used a fresh WinNT (service pack 5 and inernet explorer 5 installed) = installed cygwinat c:\cygwin, mounted c: to /, set PATH, SHELL, HOME, M= AKE_MODE=20 and TMPDIR, restarted cygwin, created /tmp, installed ghc-4.08.2 at=20 c:\ghc\ghc-4.08.2, added it to PATH, restarted cygwin and went to my co= de=20 directory. The rest is described above. So until the next problem, Andreas rrt1001@cam.ac.uk@cl.cam.ac.uk am 24.02.2001 13:57:17 Gesendet von: reuben.thomas@cl.cam.ac.uk An: Andreas Marth/FT/DCAG/DCX@WK-EMEA2 Kopie: glasgow-haskell-users@haskell.org =20 Thema: Re: Problems when mounting c: to / > When installing ghc-4.08.2 (onWinNT) as described in the installation= > instructions under c:, I couldn't get ghc to run. > It invoked the C-preprozessor and stoped after cleaning. I'm puzzled by this. Could you please send some sample output? Preferab= ly running ghc -v. > When I unmount c:, it works fine. (If I install cygwin and dont mount= c: to / > it works fine too.) > But now //c/ is the normal win-root. That's not a problem for me, but= how=20 about > ghc? Well, if it works, I suppose it's not a problem... > Does it depend in any way on c: mounted to /? I'm not sure. I think it used to (because of using /tmp as the temporar= y directory). Now it's mostly for convenience. > If not why do you want this mount (in the installation instructions = section > 2.2.2.1)? I think it used to be necessary, and I kept it in the instructions part= ly because I find it useful to be able to refer to the drive with Unix-sty= le paths, with the root directory being /, and partly because it worked, a= nd the less I change the instructions, the less likely I am to break somet= hing. --=20 http://sc3d.org/rrt/ =A6 wit, n. educated insolence (Aristotle) = --Boundary=_0.0_=0057440008329344 Content-Type: application/octet-stream; name=withMount.log Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename=withMount.log VGhlIEdsb3Jpb3VzIEdsYXNnb3cgSGFza2VsbCBDb21waWxhdGlvbiBTeXN0ZW0sIHZlcnNpb24g NC4wOC4yDQoNCkVmZmVjdGl2ZSBjb21tYW5kIGxpbmU6IC1jIC12DQoNCkluZWZmZWN0aXZlIEMg cHJlLXByb2Nlc3NvcjoNCgllY2hvICd7LSMgTElORSAxICJNYWluLmhzIiAtfScgPiBDOi9URU1Q L2doYzEzMy5jcHAgJiYgY2F0IE1haW4uaHMgPj4gQzovVEVNUC9naGMxMzMuY3BwDQpkZWxldGlu Zy4uLiBDOi9URU1QL2doYzEzMy5jcHANCg0Kcm0gLWYgQzovVEVNUC9naGMxMzMqDQo= --Boundary=_0.0_=0057440008329344 Content-Type: application/octet-stream; name=withoutMount.log Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename=withoutMount.log VGhlIEdsb3Jpb3VzIEdsYXNnb3cgSGFza2VsbCBDb21waWxhdGlvbiBTeXN0ZW0sIHZlcnNpb24g NC4wOC4yDQoNCkVmZmVjdGl2ZSBjb21tYW5kIGxpbmU6IC1jIC12DQoNCkluZWZmZWN0aXZlIEMg cHJlLXByb2Nlc3NvcjoNCgllY2hvICd7LSMgTElORSAxICJNYWluLmhzIiAtfScgPiBDOi9URU1Q L2doYzE2Mi5jcHAgJiYgY2F0IE1haW4uaHMgPj4gQzovVEVNUC9naGMxNjIuY3BwDQowLjA1dXNl ciAwLjA5c3lzdGVtIDA6MDAuMTllbGFwc2VkIDczJUNQVSAoMGF2Z3RleHQrMGF2Z2RhdGEgMG1h eHJlc2lkZW50KWsNCjBpbnB1dHMrMG91dHB1dHMgKDBtYWpvciswbWlub3IpcGFnZWZhdWx0cyAw c3dhcHMNCmdoYzpjb21waWxlOk91dHB1dCBmaWxlIE1haW4ubyBkb2Vzbid0IGV4aXN0DQpnaGM6 Y29tcGlsZTpJbnRlcmZhY2UgZmlsZSBNYWluLmhpIGRvZXNuJ3QgZXhpc3QNCmdoYzpyZWNvbXBp bGU6SW5wdXQgZmlsZSBNYWluLmhzIG5ld2VyIHRoYW4gTWFpbi5vDQoNCkhhc2tlbGwgY29tcGls ZXI6DQoJQzovZ2hjL2doYy00LjA4LjIvbGliL2hzYyBDOi9URU1QL2doYzE2Mi5jcHAgIC1maWdu b3JlLWludGVyZmFjZS1wcmFnbWFzIC1mb21pdC1pbnRlcmZhY2UtcHJhZ21hcyAtZnNpbXBsaWZ5 IFsgLWZtYXgtc2ltcGxpZmllci1pdGVyYXRpb25zNCBdICAgLWZ3YXJuLW92ZXJsYXBwaW5nLXBh dHRlcm5zIC1md2Fybi1taXNzaW5nLW1ldGhvZHMgLWZ3YXJuLW1pc3NpbmctZmllbGRzIC1md2Fy bi1kZXByZWNhdGlvbnMgLWZ3YXJuLWR1cGxpY2F0ZS1leHBvcnRzIC1maGktdmVyc2lvbj00MDgg LXN0YXRpYyAiLWhpbWFwPS4lLmhpOkM6L2doYy9naGMtNC4wOC4yL2xpYi9pbXBvcnRzL3N0ZCUu aGkiICItaGltYXAtc2VwPToiICAgIC12IC1oaWZpbGU9QzovVEVNUC9naGMxNjIuaGkgLW9sYW5n PWFzbSAtb2ZpbGU9QzovVEVNUC9naGMxNjIucyAtRj1DOi9URU1QL2doYzE2Ml9zdGIuYyAtRkg9 QzovVEVNUC9naGMxNjJfc3RiLmggK1JUUyAtSDYwMDAwMDAgLUsxMDAwMDAwDQpHbGFzZ293IEhh c2tlbGwgQ29tcGlsZXIsIHZlcnNpb24gNC4wOCwgZm9yIEhhc2tlbGwgOTgsIGNvbXBpbGVkIGJ5 IEdIQyB2ZXJzaW9uIDQuMDgNCjAuMDN1c2VyIDAuMDRzeXN0ZW0gMDowMS42NWVsYXBzZWQgNCVD UFUgKDBhdmd0ZXh0KzBhdmdkYXRhIDBtYXhyZXNpZGVudClrDQowaW5wdXRzKzBvdXRwdXRzICgw bWFqb3IrMG1pbm9yKXBhZ2VmYXVsdHMgMHN3YXBzDQoNClBpbiBvbiBIYXNrZWxsIGNvbnNpc3Rl bmN5IGluZm86DQoJZWNobyAnDQoJLnRleHQNCmhzYy5NYWluLmhzLjQwLjAuLjonID4+IEM6L1RF TVAvZ2hjMTYyLnMNCjAuMDZ1c2VyIDAuMDdzeXN0ZW0gMDowMC4yMGVsYXBzZWQgNjUlQ1BVICgw YXZndGV4dCswYXZnZGF0YSAwbWF4cmVzaWRlbnQpaw0KMGlucHV0cyswb3V0cHV0cyAoMG1ham9y KzBtaW5vcilwYWdlZmF1bHRzIDBzd2Fwcw0KKioqIE5ldyBoaSBmaWxlIGZvbGxvd3MuLi4NCl9f aW50ZXJmYWNlICJNYWluIiBNYWluICB3aGVyZQ0KX19leHBvcnQgTWFpbiBtYWluOw0KaW1wb3J0 IFByZWxGbG9hdCAxICE7DQppbXBvcnQgUHJlbHVkZSAxOw0KbWFpbiA6OiBQcmVsSU9CYXNlLklP IFByZWxCYXNlLlowVCA7DQoNCg0KZ2hjOiBtb2R1bGUgdmVyc2lvbiBjaGFuZ2VkIHRvIDE7IHJl YXNvbjogbm8gb2xkIC5oaSBmaWxlDQoNClJlcGxhY2UgLmhpIGZpbGUsIGlmIGNoYW5nZWQ6DQoJ Y21wIC1zIE1haW4uaGkgQzovVEVNUC9naGMxNjIuaGktbmV3IHx8ICggcm0gLWYgTWFpbi5oaSAm JiBjcCBDOi9URU1QL2doYzE2Mi5oaS1uZXcgTWFpbi5oaSApDQpDb21tYW5kIGV4aXRlZCB3aXRo IG5vbi16ZXJvIHN0YXR1cyAyDQowLjAzdXNlciAwLjA5c3lzdGVtIDA6MDAuMTdlbGFwc2VkIDcw JUNQVSAoMGF2Z3RleHQrMGF2Z2RhdGEgMG1heHJlc2lkZW50KWsNCjBpbnB1dHMrMG91dHB1dHMg KDBtYWpvciswbWlub3IpcGFnZWZhdWx0cyAwc3dhcHMNCg0KVW5peCBhc3NlbWJsZXI6DQoJZ2Nj IC1vIE1haW4ubyAtYyAgLUkuIC1JQzovZ2hjL2doYy00LjA4LjIvbGliL2luY2x1ZGVzIEM6L1RF TVAvZ2hjMTYyLnMNCjAuMDh1c2VyIDAuMThzeXN0ZW0gMDowMC4zNWVsYXBzZWQgNzQlQ1BVICgw YXZndGV4dCswYXZnZGF0YSAwbWF4cmVzaWRlbnQpaw0KMGlucHV0cyswb3V0cHV0cyAoMG1ham9y KzBtaW5vcilwYWdlZmF1bHRzIDBzd2Fwcw0KDQpybSAtZiBDOi9URU1QL2doYzE2MioNCg== --Boundary=_0.0_=0057440008329344-- From rrt1001@cam.ac.uk Tue Feb 27 10:25:11 2001 From: rrt1001@cam.ac.uk (Reuben Thomas) Date: Tue, 27 Feb 2001 10:25:11 +0000 (GMT) Subject: Problems when mounting c: to / In-Reply-To: <0057440008329344000002L442*@MHS> Message-ID: > I think you should drop the mount c: at / because it is sure a problem you run > into when installing ghc. I think you're right. I really should reinstall Cygwin on my machine in the default location, rather than at root. I'll change the instructions. Thanks. -- http://sc3d.org/rrt/ "Reality is what refuses to disappear when you stop believing in it" - Philip K. Dick From josefs@cs.chalmers.se Wed Feb 28 09:46:58 2001 From: josefs@cs.chalmers.se (Josef Svenningsson) Date: Wed, 28 Feb 2001 10:46:58 +0100 (MET) Subject: Lightningspeed haskell Message-ID: Hi all. Some days ago someone posted this url: http://www.bagley.org/~doug/shootout/ which is a page benchmarking a number of different languages and compilers where ghc is one of them. Some benchmarks lacked a haskell versions (and some still do) and so I decided to fill in some of the gaps. One benchmark turned out to give pretty remarkable results. It's the producer/consumer benchmark. I suggest you all take a look at it. The haskell version is six (SIX!!!) times faster than the c version. Hey, what's going on here? I would really like to hear some comments from our dear implementors. It should be noted that synchronisation is achieved by using slightly different kinds of primitives. But still... six times... I lift my hat of for the ghc-implementors. /Josef From simonpj@microsoft.com Wed Feb 28 10:16:11 2001 From: simonpj@microsoft.com (Simon Peyton-Jones) Date: Wed, 28 Feb 2001 02:16:11 -0800 Subject: Lightningspeed haskell Message-ID: <37DA476A2BC9F64C95379BF66BA2690260DB9C@red-msg-09.redmond.corp.microsoft.com> Cool! Thanks for doing this. I guess GHC is fast because it's implementing lightweight threads inside a single OS thread. Simon | -----Original Message----- | From: Josef Svenningsson [mailto:josefs@cs.chalmers.se] | Sent: 28 February 2001 09:47 | To: glasgow-haskell-users@haskell.org | Subject: Lightningspeed haskell | | | Hi all. | | Some days ago someone posted this url: | http://www.bagley.org/~doug/shootout/ | | which is a page benchmarking a number of different languages and | compilers where ghc is one of them. Some benchmarks lacked a haskell | versions (and some still do) and so I decided to fill in some | of the gaps. | | One benchmark turned out to give pretty remarkable results. It's the | producer/consumer benchmark. I suggest you all take a look at it. The | haskell version is six (SIX!!!) times faster than the c version. Hey, | what's going on here? I would really like to hear some | comments from our | dear implementors. | | It should be noted that synchronisation is achieved by using | slightly different kinds of primitives. But still... six times... | | I lift my hat of for the ghc-implementors. | | /Josef | | | _______________________________________________ | Glasgow-haskell-users mailing list | Glasgow-haskell-users@haskell.org | http://www.haskell.org/mailman/listinfo/glasgow-haskell-users | From simonmar@microsoft.com Wed Feb 28 12:03:20 2001 From: simonmar@microsoft.com (Simon Marlow) Date: Wed, 28 Feb 2001 04:03:20 -0800 Subject: Lightningspeed haskell Message-ID: <9584A4A864BD8548932F2F88EB30D1C61157F4@TVP-MSG-01.europe.corp.microsoft.com> > It should be noted that synchronisation is achieved by using > slightly different kinds of primitives. But still... six times... And it's about to get faster still, because CVars can now be implemented with a single MVar instead of two. The reason is that putMVar now blocks on a full MVar rather than raising an exception. But as Simon said, the main reason is surely that GHC is using lightweight threads compared to C. BTW, was this on Linux? I'd be interested to see the results on systems that have different threading models, because Linux's threads implementation maps threads onto processes (albeit lightweight kind of process, but still a process), so the context switch overhead is going to be much higher than a threads library which sits in a single process. Cheers, Simon From josefs@cs.chalmers.se Wed Feb 28 14:04:22 2001 From: josefs@cs.chalmers.se (Josef Svenningsson) Date: Wed, 28 Feb 2001 15:04:22 +0100 (MET) Subject: Lightningspeed haskell In-Reply-To: <9584A4A864BD8548932F2F88EB30D1C61157F4@TVP-MSG-01.europe.corp.microsoft.com> Message-ID: On Wed, 28 Feb 2001, Simon Marlow wrote: > > It should be noted that synchronisation is achieved by using > > slightly different kinds of primitives. But still... six times... > > And it's about to get faster still, because CVars can now be implemented > with a single MVar instead of two. The reason is that putMVar now > blocks on a full MVar rather than raising an exception. > Cool. It'll be interesting to see the speedup. > But as Simon said, the main reason is surely that GHC is using > lightweight threads compared to C. BTW, was this on Linux? I'd be > interested to see the results on systems that have different threading > models, because Linux's threads implementation maps threads onto > processes (albeit lightweight kind of process, but still a process), so > the context switch overhead is going to be much higher than a threads > library which sits in a single process. > I don't know what OS he's using. The homepage only says what kind of CPU and amount of memory his computer has. /Josef From jmaessen@mit.edu Wed Feb 28 15:11:01 2001 From: jmaessen@mit.edu (Jan-Willem Maessen) Date: Wed, 28 Feb 2001 10:11:01 -0500 Subject: Lightningspeed haskell Message-ID: <200102281511.KAA00619@lauzeta.mit.edu> Josef Svenningsson wrote: > One benchmark turned out to give pretty remarkable results. It's the > producer/consumer benchmark. I suggest you all take a look at it. The > haskell version is six (SIX!!!) times faster than the c version. Hey, > what's going on here? I would really like to hear some comments from our > dear implementors. And Simon Peyton-Jones replied: > I guess GHC is fast because it's implementing lightweight > threads inside a single OS thread. Absolutely. Good high-level thread support trumps anything provided by the operating system. I'd second Simon's guess that this was on Linux or some other system where pthreads map one-to-one to OS threads. Similar dramatic performance disparities have cropped up in the Java community. There are Java benchmarks which create thousands of threads; most implementations slow to a crawl when this happens, as the operating system collapses under the crushing load. Those which don't (IBM's Jalapeno, a research JVM, comes to mind) use this as a *big* bragging point. Unless you're mucking with thread priorities, mapping language-level threads one-to-one to operating-system threads is a terrible idea. -Jan-Willem Maessen Eager Haskell project jmaessen@mit.edu From rrt1001@cam.ac.uk Wed Feb 28 19:17:58 2001 From: rrt1001@cam.ac.uk (Reuben Thomas) Date: Wed, 28 Feb 2001 19:17:58 +0000 (GMT) Subject: New InstallShield Message-ID: This one corrects a problem with the fix made in the last one that stopped anything to do with stat() working (e.g. hFileSize, reading directories &c. &c.). Sorry about that. The fix is still fragile and temporary; I'm waiting for the underlying mingwin problems to be fixed, hopefully in time for the GHC 5.0 release. -- http://sc3d.org/rrt/ | humour, n. unexpected recognition