sablevm-developer

[Sablevm-developer] Priority list

[Prof. E. M. G. <eti...@uq...>]

Hi again,

I,d like to start drafting a list of tasks to be done in SableVM,
along with priorities.

Please contribute any suggestion for improving the following list:

In descending priority:

- Get SableVM to run with latest Jikes.

- Update to latest GNU Classpath CVS snapshot.

- Merge all SableVM-specific changes into official GNU Classpath CVS
  (along with proposal to use m4 preprocessor for managing VM-specific
  library code).

- Improve abstraction layer for specifying inlinability of bytecode
  implementations.

- Include a regression suite (like Mauve), and automate daily checks
  for failures in CVS snapshot.


Anything else?  Different priorities?  Your opinion is important.

Have fun!

Etienne


--=20
Etienne M. Gagnon, Ph.D.             http://www.info.uqam.ca/~egagnon/
SableVM:                                       http://www.sablevm.org/
SableCC:                                       http://www.sablecc.org/

Re: [Sablevm-developer] Priority list

[Chris P. <chr...@ma...>]

Small task, but then again small tasks are easily accomplished.  Merge 
the functionality of all the build scripts floating around (mine, 
Bruno's, Etienne's, and Grzegorz's) and the two in the CVS into one 
script.  I'd vote for Bruno's script as the basis since it's really easy 
to read and understand (and therefore maintain), and I'd vote for Bruno 
to do the job (unless he no longer wants to).

Ciao-ciao,
Chris

Prof. Etienne M. Gagnon wrote:

>Hi again,
>
>I,d like to start drafting a list of tasks to be done in SableVM,
>along with priorities.
>
>Please contribute any suggestion for improving the following list:
>
>In descending priority:
>
>- Get SableVM to run with latest Jikes.
>
>- Update to latest GNU Classpath CVS snapshot.
>
>- Merge all SableVM-specific changes into official GNU Classpath CVS
>  (along with proposal to use m4 preprocessor for managing VM-specific
>  library code).
>
>- Improve abstraction layer for specifying inlinability of bytecode
>  implementations.
>
>- Include a regression suite (like Mauve), and automate daily checks
>  for failures in CVS snapshot.
>
>
>Anything else?  Different priorities?  Your opinion is important.
>
>Have fun!
>
>Etienne
>
>
>  
>

Re: [Sablevm-developer] Priority list

[David <db...@cs...>]

On Mon, Feb 24, 2003 at 01:22:57AM -0500, Chris Pickett wrote:
> Small task, but then again small tasks are easily accomplished.  Merge=20
> the functionality of all the build scripts floating around (mine,=20
> Bruno's, Etienne's, and Grzegorz's) and the two in the CVS into one=20
> script.  I'd vote for Bruno's script as the basis since it's really eas=
y=20
> to read and understand (and therefore maintain), and I'd vote for Bruno=
=20
> to do the job (unless he no longer wants to).

I don't think the scripts necessarily have to be merged.
We need a simple to use script in the distribution root directory
to simply install sablevm for end-users.  Goal: Easy and error-free
installation.

All the developper scripts should go in some other directory (say
contrib/).

David

>=20
> Ciao-ciao,
> Chris
>=20
> Prof. Etienne M. Gagnon wrote:
>=20
> >Hi again,
> >
> >I,d like to start drafting a list of tasks to be done in SableVM,
> >along with priorities.
> >
> >Please contribute any suggestion for improving the following list:
> >
> >In descending priority:
> >
> >- Get SableVM to run with latest Jikes.
> >
> >- Update to latest GNU Classpath CVS snapshot.
> >
> >- Merge all SableVM-specific changes into official GNU Classpath CVS
> > (along with proposal to use m4 preprocessor for managing VM-specific
> > library code).
> >
> >- Improve abstraction layer for specifying inlinability of bytecode
> > implementations.
> >
> >- Include a regression suite (like Mauve), and automate daily checks
> > for failures in CVS snapshot.
> >
> >
> >Anything else?  Different priorities?  Your opinion is important.
> >
> >Have fun!
> >
> >Etienne
> >
> >
> >=20
> >
>=20
>=20
>=20
> -------------------------------------------------------
> This SF.net email is sponsored by: SlickEdit Inc. Develop an edge.
> The most comprehensive and flexible code editor you can use.
> Code faster. C/C++, C#, Java, HTML, XML, many more. FREE 30-Day Trial.
> www.slickedit.com/sourceforge
> _______________________________________________
> Sablevm-developer mailing list
> Sab...@li...
> https://lists.sourceforge.net/lists/listinfo/sablevm-developer

--=20

---

David B=E9langer
Graduate Student
School of Computer Science
McGill University
Office: MC226

Web page:   http://www.cs.mcgill.ca/~dbelan2/
Public key: http://www.cs.mcgill.ca/~dbelan2/public_key.txt

Re: [Sablevm-developer] Priority list

[Grzegorz B. P. <ga...@de...>]

W li=B6cie z pon, 24-02-2003, godz. 06:19, Prof. Etienne M. Gagnon pisze:=20
> Hi again,
>=20
> I,d like to start drafting a list of tasks to be done in SableVM,
> along with priorities.
>=20
> Please contribute any suggestion for improving the following list:
>=20
> In descending priority:
>=20
> - Get SableVM to run with latest Jikes.
>=20
> - Update to latest GNU Classpath CVS snapshot.
(*)

> - Merge all SableVM-specific changes into official GNU Classpath CVS
>   (along with proposal to use m4 preprocessor for managing VM-specific
>   library code).

- Get the GTK-peers working (so far kissme seems to be best here)
It could be probably even at (*) position after merge w/ latest
classpath.

> - Improve abstraction layer for specifying inlinability of bytecode
>   implementations.
Hmm... did you have to make some arch-specific changes there?
=46rom what I understood - as long as we keep using GCC - we shouldn't
see much problems here. Or maybe you think of speeding things up on
some arches where GCC produces better inlinable code?

> - Include a regression suite (like Mauve), and automate daily checks
>   for failures in CVS snapshot.
> Anything else?  Different priorities?  Your opinion is important.

I also would like to see some one-for-all build script.=20

I am considering to either contribute to Bruno's script (but that
would require not only some improvements but also changes as I don't
agree on some of the ways it does things) or to extend my script
(which, keep in mind, has been written for pure fun between 0am and 3am
in the late night/morning) to be seriously usable.

The main lack I see in mine script are the missing options for
compiling sablevm in different ways.

The very serious question is whether my approach of "swiss-army-knife"
is right (Chris says that Bruno's script is easier to read, so there's
probably sth. wrong with mine).

Cheers,

				Grzegorz B. Prokopski
--=20
Grzegorz B. Prokopski <ga...@de...>
Debian http://www.debian.org/