3.2.0 release

View: New views
7 Messages — Rating Filter:   Alert me  

3.2.0 release

by Jaroslav Hajek-2 :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

hi all,

since nobody reported serious problems with rc7, excep, I turned the
rc7 tarballs into the unofficial 3.2.0 release.
http://artax.karlin.mff.cuni.cz/~hajej2am/ulozna/octave/

a couple of observations:
we had 7 RCs, just like with 3.0.4. However, the 3.0.4 RC loop lasted
for 2.5 months (only to produce a release with a serious bug), while
the 3.2.0 RC loop lasted 11 days, from Monday April 25th till today
(i.e. two workday weeks). It seems a number of people were alerted by
the release actually coming, which is good (and shows more people were
interested in 3.2.0 than in 3.0.4) but I really think the RCs should
be reserved only for fine tuning.
Bugs are being fixed quite contiguously, so even now there are new
patches in the savannah archive that didn't make it into 3.2.0, but I
guess a line needs to be drawn somewhere, and I the time I'm willing
to devote to making RCs is also limited. Depending on the amount of
bug fixes/improvements, I'm planning to make 3.2.1 within 1-2 months.

If anyone knows of a reason why Octave 3.2.0 should not be released as
is, speak now or forever hold your peace.

John, when you have time to do so, can you please upload the tarballs
to GNU site?
You can make the announcement to help@..., or I can do it if you wish.

I sincerely thank all contributors for valuable cooperation.

free computing, free society!

--
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz

Re: 3.2.0 release

by Jaroslav Hajek-2 :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

On Fri, Jun 5, 2009 at 9:14 AM, Jaroslav Hajek<highegg@...> wrote:
> hi all,
>
> since nobody reported serious problems with rc7, excep,

with this, I meant "except J.F. Cardoso, whose problems I believe to
be caused by a miscompilation".


--
RNDr. Jaroslav Hajek
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz

Re: 3.2.0 release

by Søren Hauberg :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

fre, 05 06 2009 kl. 09:14 +0200, skrev Jaroslav Hajek:
> since nobody reported serious problems with rc7, excep, I turned the
> rc7 tarballs into the unofficial 3.2.0 release.
> http://artax.karlin.mff.cuni.cz/~hajej2am/ulozna/octave/

Thanks. It seems that 3.2 is an excellent release. It's nice to have it
out there.

Søren


Re: 3.2.0 release. Compile problem on MacOSX

by Jean-Francois Cardoso :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

Jaroslav Hajek-2 wrote:
On Fri, Jun 5, 2009 at 9:14 AM, Jaroslav Hajek<highegg@gmail.com> wrote:
> hi all,
>
> since nobody reported serious problems with rc7, excep,

with this, I meant "except J.F. Cardoso, whose problems I believe to
be caused by a miscompilation".
Jaroslav was right in suspecting too old a gcc.

I am happy to report that the miscompilation on MacOSX
(leading to test failures related to struct arrays) is fixed
by upgrading gcc from 4.2 to 4.4.

I do not know what other people do on the Mac wrt octave/gcc.
My solution was to use the gfortran+gcc binaries
from http://hpc.sourceforge.net/

Cheers, JF

Re: 3.2.0 release. Compile problem on MacOSX

by Ben Abbott :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

On Jun 5, 2009, at 7:40 AM, Jean-Francois Cardoso wrote:

> Jaroslav Hajek-2 wrote:
>>
>> On Fri, Jun 5, 2009 at 9:14 AM, Jaroslav Hajek<highegg@...>  
>> wrote:
>>> hi all,
>>>
>>> since nobody reported serious problems with rc7, excep,
>>
>> with this, I meant "except J.F. Cardoso, whose problems I believe to
>> be caused by a miscompilation".
>
> Jaroslav was right in suspecting too old a gcc.
>
> I am happy to report that the miscompilation on MacOSX
> (leading to test failures related to struct arrays) is fixed
> by upgrading gcc from 4.2 to 4.4.
>
> I do not know what other people do on the Mac wrt octave/gcc.
> My solution was to use the gfortran+gcc binaries
> from http://hpc.sourceforge.net/

For reference, on Mac OSX 10.5.7, I'm using gfortran bundled with gcc  
4.4.

Ben


3.2.0 release

by John W. Eaton-3 :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

On  5-Jun-2009, Jaroslav Hajek wrote:

| John, when you have time to do so, can you please upload the tarballs
| to GNU site?

3.2.0 is now available from ftp.octave.org and ftp.gnu.org.

| You can make the announcement to help@..., or I can do it if you wish.

I'll post an announcement soon.

Thanks to everyone for all the hard work that went into making this
release possible.

jwe

Re: 3.2.0 release

by Petr Gajdos :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

Some parts of this message have been removed. Learn more about Nabble's security policy.

On Friday 05 June 2009 21:52:16 John W. Eaton wrote:

> 3.2.0 is now available from ftp.octave.org and ftp.gnu.org.

>

> Thanks to everyone for all the hard work that went into making this

> release possible.

Hi,

I would like to join: many many thanks for your hard work!

Petr