Re: The Linux binutils 2.14.90.0.1 is released

Re: The Linux binutils 2.14.90.0.1 is released

[Jack Howarth]

Ben,
   I think the rational of the HJ releases is provide
a way to rapidly release as many critical fixes as
possible for use with linux. Those releases are
focused on being bug-free on the linux platform
(hence their hosting on ftp.kernel.org).
                       Jack

Re: The Linux binutils 2.14.90.0.1 is released

[H. J. Lu]

On Fri, May 09, 2003 at 09:04:46AM -0400, Jack Howarth wrote:
> Ben,
>    I think the rational of the HJ releases is provide
> a way to rapidly release as many critical fixes as
> possible for use with linux. Those releases are
> focused on being bug-free on the linux platform
> (hence their hosting on ftp.kernel.org).

Thanks, Jack. That is well said. Also some Linux projects may need
certain new features in a released version of binutils. We can live
with regressions and breakages on non-linux platforms. But I don't
think the FSF binutils should/will make that kind of commitment.

H.J.

Re: The Linux binutils 2.14.90.0.1 is released

[Ben Elliston]

Jack Howarth <howarth@bromo.msbb.uc.edu> writes:

>    I think the rational of the HJ releases is provide a way to
> rapidly release as many critical fixes as possible for use with
> linux.

Hmmm .. this sounds like something with a much greater sense of
urgency than I would expect for binutils.  I can understand that it
makes sense when a new Linux kernel port and binutils are coming up at
the same time, but for established platforms, I can't see the need.

IHMO, there are better things to do with one's time than maintaining a
binutils fork.

Cheers, Ben

Re: The Linux binutils 2.14.90.0.1 is released

[Thiemo Seufer]

Ben Elliston wrote:
> Jack Howarth <howarth@bromo.msbb.uc.edu> writes:
> 
> >    I think the rational of the HJ releases is provide a way to
> > rapidly release as many critical fixes as possible for use with
> > linux.
> 
> Hmmm .. this sounds like something with a much greater sense of
> urgency than I would expect for binutils.  I can understand that it
> makes sense when a new Linux kernel port and binutils are coming up at
> the same time,

New Features or just different usage are also a point to consider,
for the kernel as well as for userland.

> but for established platforms, I can't see the need.

The Linux development tree is probably never a "established platform".
And IMO it's very valuable for binutils to get a wider test audience
sooner than the release cycle suggests.


Thiemo

Re: The Linux binutils 2.14.90.0.1 is released

[Ben Elliston]

Hi HJ,

"H. J. Lu" <hjl@lucon.org> writes:

> This is the beta release of binutils 2.14.90.0.1 for Linux, which is
> based on binutils 2003 0505 in CVS on sourecs.redhat.com plus
> various changes. It is purely for Linux.

As I understand it, you started up your "HJ special" binutils around
the time that binutils development has slowed/stalled.  It is now
actively maintained.  I'm left wondering why it is necessary for you
to continue to maintain your own binutils releases.  Can your changes
be folded into mainline binutils?

Ben