Re: Things holding gcc-defaults out of etch
On Thu, Aug 04, 2005 at 05:43:11AM -0400, Nathanael Nerode wrote:
> So there's something weird keeping gcc-defaults out of etch.
> gcc-defaults is waiting for gcc-3.4.
> britney thinks that updating gcc-3.4 in testing will break jigdo
> on arm. Presumably this is because jigdo in testing depends on
> "libstdc++6-0" on arm, which isn't going to exist any more.
> jigdo in unstable is waiting (on ARM!) for new pango1.0 to get into testing.
> pango1.0 is "in freeze", whatever that means, and is only 4/10 days old.
> jigdo also needs the successful sparc build uploaded from the buildd,
> and is only 4/10 days old.
> So:
> * pango1.0 must go into etch first (rush it?)
> * jigdo sparc build uploaded
> * jigdo goes into etch (rush it?)
> * Then gcc-3.4 goes in
> * gcc-defaults can go in as soon as gcc-3.4 does (after 2 days)
>
> Now, I noticed this because perl is waiting for gcc-defaults (sigh).
> Perl also failed to build on m68k and arm, of course.
The perl build failures are the real issue here. I have no expectation that
both build failures are going to be resolved in the next six days without
any need for a new upload.
I've unfrozen pango1.0, at any rate; pango and jigdo can both enter testing
once they've aged and all builds are available. I don't think that the
gcc-defaults change is urgent enough that it warrants shorting the testing
period for these two packages. The compiler update is a change that it
would be good to have make it into testing soon, though, so I hope there
won't be any need for further pango or jigdo uploads in the next week.
> Unfortunately, libgnome2-wnck-perl is waiting for perl, and libwnck won't
> be able to go in until libgnome2-wnck-perl can.
> Unless libgnome2-wnck-perl is just removed from testing for now,
> which might be the right thing to do; its only rdepends is perlpanel,
> which has no rdepends.
Yeah, I had just talked to the libgnome2-wnck-perl maintainer on IRC, and he
requested exactly this.
--
Steve Langasek
postmodern programmer
Reply to: