[06:21] <infinity> doko: Can you look at http://librarian.launchpad.net/1840556/buildlog_ubuntu-dapper-hppa.ept_1.91ubuntu2_FAILEDTOBUILD.txt.gz
[06:21] <infinity> doko: And yes, that's reproducible.  ept's configure script on hppa does that on every build attempt.
[06:28] <infinity> doko: Argh.  And when I say "that's reproducible", I mean "It happened to me twice in a row".. Of course, on a third try now, it worked. :/
[11:31] <doko> jbailey, infinity: glibc ping
[11:33] <jbailey> doko: What's up?
[11:37] <doko> looking at glibc which patches we miss from unstable, and the differences in patch names
[11:38] <jbailey> Yes, Debian hasn't synced to new patch names.
[11:38] <jbailey> But unstable is also now using quilt.
[11:38] <doko> ;-P
[11:38] <jbailey> They have the plans to, I don't have the spare time to push it right now.
[11:38] <jbailey> Is there a particular bug you're chasing, or just general maintenance.
[11:39] <doko> anyway, could you help in looking, which patches we do miss?
[11:39] <doko> no, general, because I'm tired chasing bugs that are fixed in unstable :-(
[11:42] <jbailey> Has there been more than just the hppa one?
[11:43] <doko> Jeff, I don't know, and I really do not _want_ to know.
[11:43] <jbailey> Right.  The thing is that you would've had this bug with hppa anyway.
[11:43] <jbailey> I used the patch that Carlos O'Donnell said he wanted on Ubuntu.
[11:44] <doko> ok, so how can we ease the comparision?
[11:46] <jbailey> I just need to finish pushing the rename into Debian.
[11:47] <doko> that would be cool
[11:47] <jbailey> But since they're not using dpatch anymore, even an md5sum listing won't help.
[11:48] <doko> could we change to quilt as well?
[11:49] <jbailey> For dapper?  I'd rather not.
[11:50] <doko> why? it would ease the comparision, even during the maintainance period
[11:51] <jbailey> Just because it's more changes to core infrastructure.
[11:51] <jbailey> The change to quilt isn't harmful in anyway, but it's more things changing.
[11:52] <doko> but it lowers the maintainance effort
[11:52] <jbailey> Does it?  Once we release, we're unlikely to ever compare against unstable.
[11:52] <doko> which other things are changing?
[11:52] <jbailey> The locales separation, moving timezones out, etc.
[11:53] <doko> right, but I think, that is addressed by the new naming of patches?
[11:53] <jbailey> I don't have a strong objection to doing quilt.  I'm just not generally in favour of changing things at this point that aren't going to clearly give benefit.
[11:54] <doko> ok, so we come back to the comparision of the current state ...
[11:54] <jbailey> Sure, which I can do.  I spent most of today on the telephone though.
[11:55] <jbailey> Tomorrow seems more likely for that sort of work.
[11:55] <jbailey> Denis isn't around, I can't ask him about the patch name change.
[11:56] <doko> I did not mean, that you have to do it. I'm trying to do it with very low effort
[11:56] <doko> but Denis is doing locales only?
[11:56] <jbailey> Denis is the one who implemented the quilt setup.
[11:56] <jbailey> He and I were the ones talking about the patch rename.
[11:56] <doko> ahh, ok
[11:56] <jbailey> I've pinged aurel32 to ask him.
[11:56] <jbailey> drow had no objection.
[11:57] <jbailey> I haven't seen gotom online in a while.
[11:57] <jbailey> pb doesn't usually get involved at that level.
[11:58] <doko> it would be nice if you could do the comparision between current dapper and unstable
[11:58] <jbailey> I did say that I would do that.
[11:59] <jbailey> Hopefully tomorrow.
[11:59] <jbailey> I don't have major calls scheduled for tomorrow which is nice.
[11:59] <doko> btw, do my amd64 changes make sense?
[12:00] <jbailey> With the gconv stuff?
[12:00] <jbailey> I think so, yeah.
[12:00] <doko> somebody really should convert glibc to some standard cdbs/debhelper stuff ;-P
[12:01] <jbailey> glibc does use debhelper. =)
[12:01] <doko> bah
[12:01] <jbailey> That was one of the big changes that I made when I first started.
[12:01] <jbailey> The glibc package in Ubuntu even uses DH_COMPAT=5 =)
[12:02] <jbailey> And much of the glibc rules file was the prototype for cdbs.
[12:02] <jbailey> But it needed multibuild.
[12:02] <jbailey> The cdbs2 code might be suitable for glibc and gcc.
[12:02] <doko> where is cdbs2?