[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

List of udebs for X11-based d-i

(Removing the — typo'd — “tentative” bits from subject this time,
after a quick discussion on -boot@.)

Cyril Brulebois <kibi@debian.org> (19/03/2010):
> Step 1:
> -------
> We push everything needed for X & Gnome. That means we don't push the
> modified udebs maintained by d-i, so that we don't break the existing
> alpha 1 release.
> This means some packages are going to be broken in testing: the d-i
> udebs are going to still rely on dropped/modified udebs (from the
> gnome stack: cairo, pango1.0, gtk+2.0), but that doesn't matter much
> since they are currently unused[1].
>  1. http://lists.debian.org/debian-boot/2010/03/msg00366.html
> That also means that gnome folks will be able to do any transitions
> they want (their planned experimental → unstable move).
> In the meanwhile, it won't be possible to build graphical images from
> testing, but that's not an issue.
> Step 2:
> -------
> We push d-i udebs during the next d-i release, which will make
> graphical images buildable from testing as well.

Both steps are unchanged.

> Please find below the list of source packages for each step:
> ,---[ Step 1 ]---
> | cairo
> | gtk+2.0
> | gtk2-engines
> | libfontenc
> | libx11
> | libxau
> | libxcb
> | libxcursor
> | libxdmcp
> | libxext
> | libxfixes
> | libxfont
> | libxi
> | libxinerama
> | libxkbfile
> | libxrender
> | pango1.0
> | udev
> | vte
> | x11-xkb-utils
> | xft
> | xkeyboard-config
> | xorg-server                          [ WARNING ]
> | xserver-xorg-input-evdev
> | xserver-xorg-video-fbdev
> `---
> WARNING about xorg-server: the udeb isn't buildable on sparc yet. I
> believe we could skip pushing it for that step 1 (nothing depend on
> it), and only push it later either when we have dropped sparc from the
> list of supported architectures, or better when nettle and/or
> xorg-server are tweaked so that it also builds on sparc. Since nothing
> depends on it, not pushing it during step 1 shouldn't be an issue.

I should add I'm probably not going to be available this weekend, but
a fixed xorg-server should appear in a matter of days, so having it
broken on sparc should only last a few days.

> Also, AFAICT from the current buildd status of those packages,
> besides some of them being only “Built” (because they were uploaded
> a couple of hours ago), there seem to be no other issues.

That still holds.

> The following packages should be the ones broken by this push:
> ,---[ Packages in sid still depending on *directfb* udebs ]---
> | cdebconf-gtk-entropy
> | cdebconf-gtk-terminal
> | cdebconf-gtk-udeb

Those are known and expected.

> | libdirectfb-bin-udeb

Quoting Frans: “This udeb will not be broken as libdirectfb-1.2-0-udeb
still exists. So no problem.”

> | libgtk-directfb-2.0-0-udeb

Actually dropped, the version which appearED is just one which has to
be dropped from the archive/decrufted, it's NBS (thanks, Julien).

> | libsdl1.2debian-udeb

It's not used yet anyway, it was aimed at making it possible to
include some games, which I believe is not quite the top
priority. Frans said it might be nice to port it to X.org as well, but
that's not for the current timeframe, at least if one wants a patch
from me (in case somebody opens a bug for such a switch, X-D-Cc: kibi
is welcome, even if I can't make any promise for that one).

I agree with Frans that it could be better to get all packages in
shape at once, but I really have no time to deal with xorg-server
right now, and I'd like to free pkg-gnome from the udeb constraints
ASAP (they did an awesome work in a very limited timeframe, that was
super-nice already!), so a push despite an arch-specific breakage
looks reasonable to me for the intended goal. Frans also said it was
up to the release team anyway, it looks like he had no objections to
this first step.

So I'm a bit more formally asking for an ACK from -boot@, and
welcoming feedback from -release@. If anything needs fixing before
step 1 happens, please let me know.


Attachment: signature.asc
Description: Digital signature

Reply to: