On 09/12/10 20:25, Matthias Kilian wrote:
> Even if this patch is trivial, someone using FreeBSD should check
> that I didn't miss something obvious before it's pushed into the
> main repository.
>
Looks fine to me. Thanks!
:g
___
Cvs-ghc mailing
On 08/10/10 14:10, Johan Tibell wrote:
> Both Bryan and I have blogged about it a couple of times. There's also
> a paper in this year's Haskell Symposium.
Oh, sorry, I should have looked it up before asking questions :) Thank
you for the links.
:g
_
On 08/10/10 13:29, Johan Tibell wrote:
> The new I/O manager has now been integrated into GHC and base. Please
> be on the lookout for any I/O related anomalies and report them to me
> (preferably via a bug assigned to my username "tibbe").
>
Yes, I am curious how FreeBSD 8.x will like it :) By t
Hello,
On 06/02/10 09:52, Simon Marlow wrote:
> I did add the #ifdef as you suggested:
Argh, sorry, I had a different (but unspoken) interpretation:
#if !defined(freebsd_HOST_OS)
# define _POSIX_SOURCE 1
# define _POSIX_C_SOURCE 199506L
# define _XOPEN_SOURCE 500
# define _ISOC99_SOURCE
#else
#
On 06/01/10 17:06, Ian Lynagh wrote:
> I checkpointed the 6.12 GHC repo, and build 2 looks to have gone through
> fine.
>
Yes, because I set the memory limit to 4 GB locally :)
Cheers,
:g
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.has
On Tue, Jun 1, 2010 at 11:11 AM, Builder wrote:
> pgj-freebsd-amd64-stable, build 2
>
> Build succeeded
For your information, I have some bad news (what I expected).
The following error causes many unexpected failures when testing:
/libexec/ld-elf.so.1: Shared object
"libHShaskell98-1.0.1.1-g
Hey,
On Tue, Jun 1, 2010 at 10:47 AM, Simon Marlow wrote:
> We could just back off the recent changes to PosixSource.h. I did try to
> fix things
Actually it fixed things on FreeBSD. Please do not "just back off"...
but do what I have suggested in the beginning: "feel free to ifdef it
around,
On Mon, May 31, 2010 at 5:30 AM, Builder wrote:
> pgj-freebsd-amd64-stable, build 1
>
> Build failed
> Details:
> http://darcs.haskell.org/ghcBuilder/builders/pgj-freebsd-amd64-stable/1.html
>
> darcs checkout | Failure: 1
Perhaps even the 1 GB memory limit is not enough for darcs when
checking
Hey Simon, hey Ben,
On Tue, Jun 1, 2010 at 10:57 AM, Ben Lippmeier wrote:
> I can spend some time hacking on it, but I'll need an account on a FreeBSD
> machine to test on. Gabor: your email addr says FreeBSD.org... ;-)
I would be glad to help you with fixing this issue from the FreeBSD
side.
Hej!
On Sat, May 29, 2010 at 5:33 PM, Matthias Kilian wrote:
> On Sat, May 29, 2010 at 11:41:36AM +0100, Ian Lynagh wrote:
>> Hmm, build 48 used those flags when compiling, but still failed.
>> Any idea what's wrong?
>>
>> http://darcs.haskell.org/ghcBuilder/builders/pgj2/48.html>
No idea :) Th
Hello Ian,
Would you please add the "--with-gmp-includes" and
"--with-gmp-libraries" flags for the "configure" script? The problem
has been fixed, having failures continuously might shadow other
problems.
Thank you,
:g
On Thu, May 27, 2010 at 4:39 AM, Builder wrote:
> pgj2, build 47
>
> Build
Hello,
I wrote a small patch to implement the missing bits for FreeBSD required
in the RTS to do proper threading. (FreeBSD 7.1 or later required.) I
am not sure whether I covered all the missing parts, however the patch
already reduced the number of unexpectedly failing test cases when
validati
Hello,
I found some problems with rts/PosixSource.h that make the validate
script fail on FreeBSD (they emit GCC warnings when compiling the RTS).
The reason is that constants are defined with wrong values and in a
wrong way. For more information see a recent thread on the
freebsd-hackers mailin
Hello,
I got a validation error on FreeBSD/i386 for Linker.c. I attach a
tiny patch to fix it, hopefully other systems will like it too :)
Cheers,
:g
ghc-rts-Linker.c.diff
Description: Binary data
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://w
On 05/17/10 21:52, Ian Lynagh wrote:
> Solaris doesn't have whereis.
Good to know :)
> Does this work for you?
>
The output is:
yes
yes
no
So I presume yes :) Thanks for the correction!
Cheers,
:g
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
Hello there,
The "validate" script for GHC-head does not run on FreeBSD correctly,
because it uses make(1) which is not GNU one. To detect and use
gmake(1) when it is available I created a small patch for "validate"
(see attached).
Cheers,
:g
--- validate.orig 2010-05-05 17:09:43.000
On Mon, Apr 19, 2010 at 5:55 PM, Simon Marlow wrote:
> How many cores do you have?
Four. It is an "Intel(R) Core(TM)2 Quad CPU Q8400 @ 2.66GHz".
:g
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
On Mon, Apr 19, 2010 at 2:26 PM, Simon Marlow wrote:
> It's supposed to run in very little memory, if it doesn't that's a bug. Is
> it displaying this behaviour for all the ways?
No, only 'threaded2' does this, 'normal' and 'threaded1' pass.
Cheers,
:g
_
On 04/16/10 16:09, Ian Lynagh wrote:
> Thu Apr 15 07:39:19 PDT 2010 Ian Lynagh
> * Convert boot and boot-pkgs to perl
> This stops us having to worry about sh/sed/... portability.
>
Thank you very much!
:g
___
Cvs-ghc mailing list
Cvs-ghc@hask
Hello,
Based on my experiences with the nightly builds and tests, it seems that
2 + 4 GB (real + swap) memory is not enough sometimes :) There is a
process (throwto003, test 173) during running `make -C
testsuite/tests/ghc-regress BINDIST=YES` (phase 11 in the builder) which
eats up a lot of memo
On Wed, Apr 14, 2010 at 10:10 PM, Ian Lynagh wrote:
> Hmm, we already need perl for darcs-all. Perhaps the best solution would
> be to rewrite boot* in perl. That would also mean that we could use
> perl's regexps, without having to worry about differences in sed's
> behaviour on different platfor
Hello,
On Wed, Apr 14, 2010 at 11:28 AM, Roman Leshchinskiy
wrote:
> While that is true, we should certainly expect that patches which obviously
> affect cross-platform compatibility are validated on all supported platforms.
Is there any progress on this (I mean has anybody tried the patch on
On Mon, Apr 12, 2010 at 12:43 PM, Matthias Kilian
wrote:
> just put a symlink [..] into your buildbots bin directory. Yes, it's still a
> workaround, but at least
> you don't break your system.
Oops, I should have thought about that :D Thanks for the hints! I
mainly picked this approach becaus
Hej Ben, Lars, och Kili,
First of all, sorry for not mentioning that FreeBSD has a base system
+ ports model. As Lars has pointed out, using bash on a FreeBSD
system (whether may I say on any other BSD system?) has a fixed cost
of installing it as a build-time dependency. Building GHC 6.10.4
wor
Hello,
While working with the latest GHC sources on FreeBSD, I found that the
"boot" script calls bash without any reason. In that script, bash is
used for running "boot-pkgs" which has a shebang line referring to sh
only (i.e. it is not a bash script). In my opinion, bash should be
omitted for
25 matches
Mail list logo
