=== ogra [n=ogra@ubuntu/member/ogra]  has joined #ubuntu-x
[08:15] <fabbione> morning guys
[08:21] <fabbione> LOCK xorg-server
[08:21] <fabbione> once this is done we can do drivers and so on
=== ogra [n=ogra@ubuntu/member/ogra]  has joined #ubuntu-x
=== Hobbsee [n=Hobbsee@ubuntu/member/hobbsee]  has joined #ubuntu-x
[09:01] <fabbione> dpkg-checkbuilddeps: Unmet build dependencies: mesa-swx11-source (>> 6.4.0)
[09:01] <fabbione> dpkg-buildpackage: Build dependencies/conflicts unsatisfied; aborting.
[09:01] <fabbione> dpkg-buildpackage: (Use -d flag to override.)
[09:01] <fabbione> W T F
[09:17] <fabbione> this merge is almost insanbe
[09:29] <fabbione> Mithrandir: ping?
[09:30] <Mithrandir> uh, what package is that?
[09:30] <Mithrandir> xorg-server?
[09:30] <fabbione> Mithrandir: yeah that's not the problem really
[09:31] <Mithrandir> what's the problem then?  Having to get mesa merged first?
[09:31] <fabbione> Mithrandir: i need you to merge xkeyboard-config: and Depends on xorg-server_1.0.2-9 or higher
[09:32] <Mithrandir> why?
[09:32] <fabbione> that mesa thing is just a package rename. It doesn't really bother me
[09:32] <Mithrandir> xkeyboard-config should be fine as-is.
[09:32] <fabbione> xkeyboard-config -> our /etc/X11/xkb
[09:32] <fabbione> debian /usr/share/X11/xkb
[09:32] <fabbione> i can change the server to point to our path
[09:32] <fabbione> but it's another diversion
[09:32] <Mithrandir> please just do that, I'm not going to move the conffiles there just yet.  It'll take a bit of time to work that bit out.
[09:33] <fabbione> ok
=== fabbione [i=fabbione@gordian.fabbione.net]  has joined #ubuntu-x
[12:15] <fabbione> new server is on the way up
[12:15] <fabbione> i think i did a mistake when merging one of the first packages with a lot of C/F/R
[12:15] <fabbione> the versions were too tight against our versions
[12:16] <fabbione> an upload to -security or -updates would break
[12:22] <fabbione> server is up
[12:23] <fabbione> all drivers are basically unlocked now
[12:23] <fabbione> there is one important bit we want to look at
[12:23] <fabbione> Debian is using some xserver-video-all provide/depends stuff
[12:23] <fabbione> our is called s/video/driver
[12:23] <fabbione> the server as it is understands both
[12:23] <fabbione> but we should be able to drop drivers to be in sync with Debian
[12:41] <Mithrandir> xorg depends on -driver, though.
[12:43] <fabbione> Mithrandir: yes, but it provides -driver-all
[12:43] <Mithrandir> what provides driver-all?
[12:43] <fabbione> and the server Depends: on driver-all | driver | video-all | video
[12:43] <fabbione> xorg does
[12:43] <Mithrandir> oh, it probably does, yes.
[12:43] <fabbione> it's a meta pacakge more than a provides
[12:44] <fabbione> so start changing xorg and then all drivers to provide/have -video- instead of -drivers- and i will be able to drop -drivers- from -server
[12:44] <Mithrandir> sure, sounds easy enough.
[12:44] <fabbione> yeps
[12:45] <fabbione> it just needs to be done by who will do single drivers
[12:45] <fabbione> i also strongly suggest that who does the ati driver to talk to airled and/or benh
[12:45] <fabbione> since we were using a special -stable branch with extra bug fixes
[12:45] <Mithrandir> then do -ati last?
[12:46] <fabbione> i might..
[12:52] <fabbione> or actually.. we could get it directly from 7.1
[12:52] <fabbione> or the last one released standalone
[12:52] <fabbione> that would be even better
=== rodarvus [n=rodarvus@ubuntu/member/rodarvus]  has joined #ubuntu-x
[01:03] <fabbione> hey Rodrigo
[01:05] <rodarvus> hey fabbione
[01:05] <rodarvus> you were supposed to be sleeping by now
[01:05] <rodarvus> or at least suffering the effects of huge beer last night :D
[01:07] <fabbione> rodarvus: it's 1 pm here :)
[01:07] <fabbione> i had only 3 beers
[01:07] <fabbione> but it was very hard to wake up today
[01:07] <fabbione> rodarvus: a small /msg to update you
[01:07] <fabbione> the server is up
[01:08] <fabbione> so now you can do all driver
[01:08] <fabbione> just be careful of what i pasted to you
=== rodarvus reads
[01:11] <fabbione> -> food
[01:11] <fabbione> bbl
[01:41] <rodarvus> fabbione, Mithrandir: xorg already has the machinery to understand xserver-xorg-video-*, then?
[01:41] <rodarvus> so, basically
[01:41] <rodarvus> add Replaces: xserver-xorg-driver-<driver>
[01:41] <rodarvus> Conflicts: xserver-xorg-driver-<driver>
[01:41] <rodarvus> (i.e., no need for Provides:, then)
[01:41] <rodarvus> I'll be back in 20 minutes
[01:42] <rodarvus> If I have the nod by then, I'll start with the drivers right away
[01:43] <fabbione> rodarvus: no xorg doesn't.
[01:43] <fabbione> xorg needs that change but it needs to be coordinated
[01:44] <fabbione> xorg source generates a binary called xserver-xorg-driver-all that Depends: on all driver
[01:44] <fabbione> that package needs to be changed to xserver-xorg-video-all
[01:44] <fabbione> each driver Provides: xserver-xorg-driver
[01:44] <fabbione> these need to be changed to xserver-xorg-video
[01:44] <fabbione> once this is all done i can change the server to -video- only
[01:48] <Mithrandir> just do the change, we don't need to transition this now.
[01:48] <Mithrandir> as long as it'll all be fixed in a day or so.
[02:19] <rodarvus> ok, /me proceeds with drivers
[02:19] <rodarvus> it should be doable to have them all ready by the end of day today
=== rodarvus_ [n=rodarvus@200.146.21.149.adsl.gvt.net.br]  has joined #ubuntu-x
=== rodarvus_ is now known as rodarvus
[02:41] <Mithrandir> rodarvus: how is the driver merge going?  Should I help out?
[02:42] <rodarvus> its supposed to be pretty easy, but I've just started downloading stuff
[02:49] <rodarvus> Mithrandir: up to you, really - maybe I can "raise the flag" if I find myself in trouble in the next two-three hours
[02:55] <Mithrandir> ok, feel free to
[02:58] <Mithrandir> you're doing -video, right?  I can do the few -input ones we need
[03:00] <rodarvus> sure, great
[03:04] <Mithrandir> fabbione: I'm grabbing -evdev
[03:04] <fabbione> Mithrandir: go ahead
[03:04] <fabbione> it's all your
[03:05] <Mithrandir> lucky me. :-P
[03:06] <fabbione> clearly
[03:12] <rodarvus> NEW: xserver-xorg-video-apm_1.0.1.5-2ubuntu1.dsc
[03:12] <rodarvus>  OK: xserver-xorg-video-apm_1.0.1.5.orig.tar.gz
[03:12] <rodarvus>  OK: xserver-xorg-video-apm_1.0.1.5-2ubuntu1.diff.gz
[03:13] <rodarvus> do I need to poke someone to manually accept NEW packages, or does it happen automatically?
[03:13] <rodarvus> I suppose not, as they are for main
[03:13] <fabbione> actually
[03:14] <fabbione> we will also need to get rid of their -driver- counterparts while you switch
[03:14] <fabbione> rodarvus: NEW is not an issue
[03:14] <fabbione> please tell ftp-masters that all the *-video-* are the old *-drivers-*
[03:14] <fabbione> so that they get accepted and sent to the proper pool
[03:14] <fabbione> s/pool/component
[03:16] <fabbione> keep track of all of them
[03:16] <fabbione> once we get there, we will need to remove the old *-drivers-*
[03:16] <rodarvus> fabbione: do I need to do that *right away*, or only after uploading all *-video-* ?
[03:16] <fabbione> up to you
[03:17] <fabbione> actually..
[03:17] <Mithrandir> rodarvus: tell Scott and Colin about it at least, since they're the ones doing most of NEW atm.
[03:18] <fabbione> hmm
[03:18] <fabbione> i don't think we can force people to update to the new driver until we remove the old one from archive
[03:18] <fabbione> because we don't have versioned Provide or something
[03:19] <Mithrandir> it'll happen once xorg depends on the new one.
[03:19] <fabbione> or we need to change the Depends: -drivers- into -video-
[03:19] <fabbione> exactly
[03:19] <fabbione> we can do that at last
[03:19] <rodarvus> *nods*
[03:19] <fabbione> so it won't be painful
[03:19] <Mithrandir> hmm, the server ABI is the same still, right?
[03:19] <Mithrandir> so we won't get into hilarity there?
[03:20] <fabbione> 1:0.99.2+cvs.20051025-1 <-old
[03:20] <fabbione> 1:0.99.2-1 <-old
[03:20] <fabbione> make the last one new
[03:20] <fabbione> but it's the same abi.. i am pretty sure
[03:20] <fabbione> we carry the same patches
[03:20] <Mithrandir> let's hope so.
[03:20] <fabbione> on the same source
[03:20] <fabbione> i am running it
[03:20] <fabbione> if it was different it would have exploded
[03:21] <Mithrandir> not necessarily.
[03:21] <fabbione> well it's even the same .orig.tar.gz
[03:21] <fabbione> md5sum checked :)
[03:21] <Mithrandir> true.
[03:21] <fabbione> and same patches
[03:21] <Mithrandir> we'll see if it all blows up or not, then.
[03:21] <fabbione> so i don't see how it could break the abi
[03:21] <fabbione> exactly
[03:21] <Mithrandir> it'll for 7.1, iirc.
[03:22] <fabbione> yes mostlikely
[03:23] <fabbione> Mithrandir: do you have any idea at what speed 7.1 is entering experimental?
[03:23] <Mithrandir> not really.
[03:24] <Mithrandir> I've been looking for gravity for the last few days, but he has eluded me.
[04:10] <rodarvus> -apm, -ark, -chips, -cirrus, -cyrix, -dummy uploaded as NEW
[04:10] <rodarvus> LOCK on -fbdev, -glint, -i128, -740, -810
[04:11] <rodarvus> (all updated on x-pkgs)
[04:25] <rodarvus> -fbdev, -glint, -i128, -i740 uploaded as NEW
[04:25] <rodarvus> working on -i810 now
[04:25] <fabbione> hmmm
[04:26] <fabbione> i am a bit puzzled
[04:26] <fabbione> most of these drivers don't have patches
[04:26] <fabbione> what are you merging exactly?
[04:27] <rodarvus> Added Conflicts:, Replaces: xserver-xorg-driver-glint
[04:27] <rodarvus> for most of them is basically this
[04:27] <fabbione> oh right
=== fabbione HEAD -> WALL
[04:28] <rodarvus> possibly sync would be better to describe the changelog entries
[04:28] <fabbione> yeah
[04:28] <fabbione> we will be able to sync sometimes after we will do a new LTS release
[04:29] <fabbione> because i assume we don't want dapper customers to have to go trough edgy[0..N]  to get to a new LTS version
[04:29] <rodarvus> at least I hope they won't ;)
[04:29] <fabbione> so we will need to support dapper -> new LTS version direct update.. i guess
[04:29] <rodarvus> for sure
[04:29] <fabbione> if that's the case we need to keep this stuff around for a long time
[04:30] <rodarvus> given the timeframe of LTS support, "forever"
[04:30] <fabbione> almost
[04:31] <rodarvus> fabbione: do you remember package changes from 5 years ago? :)
[04:31] <fabbione> rodarvus: no, that's why a proper changelog is important
[04:31] <rodarvus> I agree
[04:31] <fabbione> we should probably write the changes somewhere we can look at in 5 years
[04:31] <fabbione> or who for us can look them up
[04:32] <fabbione> not that i care if somebody will replace me
[04:32] <fabbione> it will suck to be them..
[04:33] <ogra> launchpad ....
[04:33] <fabbione> xserver-xorg-driver-apm (rodarvus, NEW xserver-xorg-video-apm)
[04:33] <fabbione> xserver-xorg-driver-ark (rodarvus, NEW xserver-xorg-driver-ark)
[04:33] <rodarvus> I'm writing all changes to the changelog - you mean write them somewhere else (such as wiki), so they know exactly all kinds of quirks they'll have to support for "lts 2"?
[04:33] <fabbione> spot the error ;)
[04:33] <rodarvus> haha
[04:34] <rodarvus> what error? :)
[04:34] <rodarvus> argh
[04:34] <ogra> :)
[04:34] <fabbione> rodarvus: i think we should just take notes that the changes we kept in dapper -> edgy are the same that we will need to preserve between dapper and LTS 2
=== rodarvus bangs head against wall repeatedly
[04:34] <ogra> shit happens
[04:34] <fabbione> ahahha
[04:34] <fabbione> rodarvus: you fixed the only good one :OP
[04:35] <rodarvus> x-pkgs fixed
[04:36] <rodarvus> fabbione: yeah, thats why I blowed my head against the wall :)
[04:36] <fabbione> rodarvus: there are 2 drivers you want to skip for now... -ati and sunffb iirc
[04:36] <fabbione> ati for sure.. sunffb i need to check
[04:36] <rodarvus> nods
[04:36] <rodarvus> but note that sunffb is not present in debian, I think
[04:36] <fabbione> yes it is
[04:37] <fabbione> for the ati one i need to speak with benh and airlied on #xorg-devel
[04:37] <fabbione> because our source was made up a -stable bug fix only cvs branch
[04:37] <fabbione> and i need to make sure debian has our fixe
[04:37] <fabbione> +s
[04:37] <fabbione> for the sunffb the one we have works
[04:37] <fabbione> the one from xorg doesn't
[04:38] <fabbione> so we might want to preserve that source
[04:38] <rodarvus> sure
=== ogra [n=ogra@ubuntu/member/ogra]  has joined #ubuntu-x
[06:57] <fabbione> rodarvus: how are you doing with the merges?
[06:58] <rodarvus> fabbione: halfway done
[06:58] <rodarvus> its very likely that I'll have all video synced today
[06:58] <fabbione> rodarvus: ok, just remember that the 13 of July is deadline for merges
[06:58] <fabbione> and we need to finish by that time
[06:59] <rodarvus> *nods*
[06:59] <rodarvus> actually, I was thinking about this earlier today
[06:59] <rodarvus> the deadline for merges basically means its going to be very hard to get X7.1 into Edgy
[07:00] <fabbione> let's worry about one thing at a time. We need to merge ASAP
[07:01] <fabbione> 7.1 might come as UVF exception
[07:04] <rodarvus> fabbione: merges/syncs are my current worry - as I said, it was just a thought :)
[07:04] <fabbione> rodarvus: yeps... let's worry about one thing at a time for now
[07:04] <fabbione> jumping from 7.0 to 7.1 can probably be done in half day
[07:28] <rodarvus> (just for reference)
[07:28] <rodarvus> -Package: xserver-xorg-driver-newport
[07:28] <rodarvus> -Architecture: any
[07:28] <rodarvus> +Package: xserver-xorg-video-newport
[07:28] <rodarvus> +Architecture: alpha amd64 arm hppa hurd-i386 i386 ia64 kfreebsd-i386 m68k mips mipsel netbsd-i386 powerpc
[07:30] <fabbione> yeah that's ok
[07:31] <fabbione> tho it builds also on sparc...
[07:31] <fabbione> or there are more?
[07:31] <fabbione> anyway it's fine as it is
[07:31] <fabbione> we can sort out the arch mess (if any) later on as bug fixing
[07:32] <rodarvus> changelog is here:
[07:32] <rodarvus>    * Don't build on sparc. Partial port of
[07:32] <rodarvus>      sparc/103_sparc_dont_build_useless_drivers.diff.
[07:41] <rodarvus> xserver-xorg-driver-nv is newer than xserver-xorg-video-nv
[07:41] <rodarvus> I'll merge their debian/ directory into our package
[08:09] <rodarvus> fabbione: what is your feeling about our packaging having non-dfsg-compliant code?
[08:10] <rodarvus> this is currently the case for (at least) -mga and -rendition
[08:10] <rodarvus> debian strips the non-free code of them, for reference
=== rodarvus plays safe
[08:31] <rodarvus> we can add this code on a later release if so agreed
[09:40] <fabbione> do not strip it
[09:40] <fabbione> leave it there
[10:10] <rodarvus> oh
[10:11] <rodarvus> I uploaded it as NEW (stripping the source code)
[10:11] <rodarvus> I'll ask Kamion & Keybuk to reject upload of -rendition and -mga, and reupload them with the binary blobs
[10:13] <rodarvus> just for reference, -tga is also not built on sparc anymore
[11:01] <rodarvus> family time, I'll be back later to merge/sync the rest of the drivers