Stay off the darcs repo!

Hi,

I'm currently moving the darcs repo to haskell.org, please don't
commit to it until I say so!

Thanks

Neil

darcs update

Hi,

We are moving to a new darcs repo on haskell.org, so that non-York
people can get ssh keys and access and have a chance to play in our
repo. We now have a new repo:

darcs.haskell.org/yhc

Unfortunately I haven't tagged the repo yet, so --partial doesn't
work. I am currently the only person with commit access, we're working
on getting access for everyone else.

Can no one commit to the old repo for the next day or so, until
everyone has access to the new repo. Don't start pulling from there
until I've got checkpointing set up, or it will take a while :)

Thanks

Neil

Re: examining yhc bytecode

On Saturday 30 September 2006 13:44, Robert Dockins wrote:
> On Saturday 30 September 2006 12:08, Kartik Vaddadi wrote:
> > Robert Dockins wrote:
> > > Can you give some more details about the problem?  Can you provide a
> > > small hbc file that causes the problem for you?
> >
> > OS: Windows XP SP2
> > architecture: x64 (in x86 compatibility mode)
> > I'm using the pre-built version of yhc; I didn't build it myself. The
> > minor version of the hbc file is 0x0A.
>
> Ah.  Well, this is the problem.  The bytecode library expects exactly
> version 1.9 of the bytecode set.  As far as I was aware, that was the
> latest version.
>
> Has the bytecode set been version bumped recently? It is to do with x64?
> I've been busy with other things recently and haven't been following Yhc
> development as closely as I'd like.

Tom, Neil?  Any word on this?

--

-- 
Rob Dockins

Talk softly and drive a Sherman tank.
Laugh hard, it's a long way to the bank.
       -- TMBG

Re: examining yhc bytecode

Hi

> > Has the bytecode set been version bumped recently? It is to do with x64?
> > I've been busy with other things recently and haven't been following Yhc
> > development as closely as I'd like.
>
> Tom, Neil?  Any word on this?

No idea. Will give Tom a gentle prod tomorrow.

Once we have the bytecode library in the Yhc darcs repo, we might want
to add some basic tests to the Bytecode library to be executed with
buildbot. Ditto for the Yhc.Core library.

Thanks

Neil

darcs patch: Implement IdSupply for real (and 5 more)

Tue Oct  3 13:25:52 EDT 2006  Samuel Bronson <naesten <at> gmail.com>
  * Implement IdSupply for real

Tue Oct  3 14:15:25 EDT 2006  Samuel Bronson <naesten <at> gmail.com>
  * Begin turning ImportState into a record type...

Tue Oct  3 16:44:31 EDT 2006  Samuel Bronson <naesten <at> gmail.com>
  * Refactor ImportState and IExtract some

Tue Oct  3 17:07:19 EDT 2006  Samuel Bronson <naesten <at> gmail.com>
  * Refactor IExtract.importData to use the pseudo-monad

Tue Oct  3 18:52:05 EDT 2006  Samuel Bronson <naesten <at> gmail.com>
  * More refactoring in IExtract

Tue Oct  3 19:53:00 EDT 2006  Samuel Bronson <naesten <at> gmail.com>
  * Refactor IExtract.importField

_______________________________________________
Yhc mailing list
Yhc <at> haskell.org
http://www.haskell.org/mailman/listinfo/yhc

Re: darcs patch: Implement IdSupply for real (and 5 more)

Oh, by the way, this is for review mostly. I could push it if I
wanted. I already pushed my 45 patches to make Id a newtype instead of
a type synonym.

On 10/3/06, Samuel Bronson <naesten <at> gmail.com> wrote:
> Tue Oct  3 13:25:52 EDT 2006  Samuel Bronson <naesten <at> gmail.com>
>   * Implement IdSupply for real
>
> Tue Oct  3 14:15:25 EDT 2006  Samuel Bronson <naesten <at> gmail.com>
>   * Begin turning ImportState into a record type...
>
> Tue Oct  3 16:44:31 EDT 2006  Samuel Bronson <naesten <at> gmail.com>
>   * Refactor ImportState and IExtract some
>
> Tue Oct  3 17:07:19 EDT 2006  Samuel Bronson <naesten <at> gmail.com>
>   * Refactor IExtract.importData to use the pseudo-monad
>
> Tue Oct  3 18:52:05 EDT 2006  Samuel Bronson <naesten <at> gmail.com>
>   * More refactoring in IExtract
>
> Tue Oct  3 19:53:00 EDT 2006  Samuel Bronson <naesten <at> gmail.com>
>   * Refactor IExtract.importField

Re: examining yhc bytecode

On Tue, 2006-10-03 at 22:52 +0100, Neil Mitchell wrote:
> Hi
> 
> > > Has the bytecode set been version bumped recently? It is to do with x64?
> > > I've been busy with other things recently and haven't been following Yhc
> > > development as closely as I'd like.
> >
> > Tom, Neil?  Any word on this?

It's very likely, from time to time it becomes necessary to add extra
information to the bytecode. I'm not sure what it was for though ...
though nothing to do with x64 certainly ...

Thanks

Tom

Move to darcs.haskell.org complete

Hi,

The official repo for Yhc is now at:

http://darcs.haskell.org/yhc

Please do not push patches to the York one, they will be thrown away.
We will be abandoning the york one at some point soon.

We have a new darcs web:

http://darcs.haskell.org/darcsweb/darcsweb.cgi?r=yhc;a=summary

If you do not have commit access, then everything should be as before.
If you do (which now includes a few non-York people, and was the
entire reason for the move!) "scons push" will do a push.

Thanks

Neil

Version Numbers

Hi Everyone,

We need to come up with a versioning scheme that can serve Yhc
forevermore. This is partly needed because the Debian packagers need a
version, but we probably need it anyway. Do yhc --version and you'll
see what i mean...

So what should we have as the version numbering scheme?

I propose:

<major>.<minor>.<date>.<number>

major/minor being the normal release style version numbers. We might
stick with 0.0 for a long time to come, but thats ok. I'd personally
be happy with just one major/minor number, and increment it every time
we feel like it - the major/minor distinction will have little meaning
otherwise.

date is just the date, i.e. 20060905 for the 5th of September 2006, so
its in a nice sorted way.

number starts at 0 each day, and goes up. This is required so each
buildbot created version is a separate atomic version. I would very
much like our build process to be tagging the darcs repo, collecting
the builds that fall out and sending an email. By doing this every
build becomes unique.

Thoughts? Complaints? Ideas?

-package cpphs ? (build problem)

    When I try the build process with scons, I get a failure when it
tries to link the object files together to produce executables.  The
command...

  ghc src/compiler98/FixSyntax.o ...(etc)... -o inst/bin/yhc  

...complains about...

src/compiler98/Front.o(.text+0x155d): In function `s3My_1_alt':
: undefined reference to `LanguageziPreprocessorziUnlit_unlit_closure'
src/compiler98/Front.o(.text+0x1a56): In function `s3pb_info':
: undefined reference to `LanguageziPreprocessorziCpphsziOptions_CpphsMacro_closure'
src/compiler98/Front.o(.text+0x1ab1): In function `s3OV_1_alt':

...etc. If I try the command by hand, and add "-package cpphs", I can
get a yhc executable.  I'm wondering if there is something strange about
my setup (Linux, ghc-6.4.2, scons v0.96.1.D001) which would cause this
problem, or is there something I can change in the configuration files
to get things to build properly?

Thanks,

Greg Buchholz