This is the mail archive of the
gdb@sourceware.org
mailing list for the GDB project.
Re: Moving GDB sources to subversion?
- From: Daniel Jacobowitz <drow at false dot org>
- To: Mark Kettenis <mark dot kettenis at xs4all dot nl>
- Cc: brobecker at adacore dot com, gdb at sources dot redhat dot com
- Date: Sun, 6 Nov 2005 19:27:08 -0500
- Subject: Re: Moving GDB sources to subversion?
- References: <20051028222253.GX1155@adacore.com> <200510282256.j9SMu2pQ002862@elgar.sibelius.xs4all.nl> <20051028231430.GA9909@nevyn.them.org> <200510282324.j9SNOHql024377@elgar.sibelius.xs4all.nl>
On Sat, Oct 29, 2005 at 01:24:17AM +0200, Mark Kettenis wrote:
> > Date: Fri, 28 Oct 2005 19:14:30 -0400
> > From: Daniel Jacobowitz <drow@false.org>
> >
> > On Sat, Oct 29, 2005 at 12:56:02AM +0200, Mark Kettenis wrote:
> > >
> > > Why should we?
> >
> > I'd rather not rehash the months of discussion from the GCC list about
> > this :-)
>
> The we're on the same wavelength. Let's put our effort into improving
> GDB; not in changing our infrastructure, let alone discussions about
> our infrastructure.
On the contrary, I think I could do GDB development more effectively if
it were not managed by CVS. Subversion is a lot more useful for things
like branching, and finding bug fixes to merge them back to an old
release branch your customers are still using. Which I waste a lot of
time on.
The only reason I'm not being more vocal about my preference to switch
is that the src repository poses certain unique challenges that GCC
didn't. Specifically, while cvs modules are insufficient for our
needs, we've gotten used to their quirks and can cope with them.
Subversion would have a whole different set of problems if you didn't
want to check out the whole src tree.
I don't think it's feasible for us to switch until someone has invested
some time thinking about that problem, and coming up with a better
solution than anything I've come up with so far.
--
Daniel Jacobowitz
CodeSourcery, LLC