[00:22] <tjaalton> wgrant: toss me the url once you're done
[00:46] <tjaalton> x11perf -a10text has gone from ~55k on intrepid to ~91k with the new xserver, and ~145k with the new xserver & kernel 2.6.28
[00:47] <tjaalton> wgrant: I'm wondering if the problem would go away with the new xserver
[01:10] <tjaalton> [ 1433.531017] irq 16: nobody cared (try booting with the "irqpoll" option)
[01:11] <tjaalton> that makes my machine really slow, since i915 was using that irq
[01:18] <wgrant> It's perfectly fast for me.
[01:18] <tjaalton> probably hw related
[01:19] <tjaalton> something crashes and as a result that irq gets disabled
[01:20] <wgrant> I used to have that.
[01:20] <wgrant> But with .28 I don't.
[01:20] <wgrant> (the 'nobody cared')
[01:20] <tjaalton> yeah :)
[01:44] <tjaalton> wgrant: reported the libxrandr bug upstream yet?
[09:42] <crevette> good morning
[17:05] <tjaalton> 7away
[17:05] <tjaalton> uh
[17:05] <tjaalton> wgrant: what gfx chip do you have?
[17:06] <tjaalton> wgrant: bug 307306
[17:06] <ubottu> Launchpad bug 307306 in libxrandr "upgrade to 2:1.2.99.2-0ubuntu1 makes session utterly slow" [Undecided,New] https://launchpad.net/bugs/307306
[17:07] <tjaalton> can't wait any longer, I'll report it upstream
[17:11] <tjaalton> wgrant: piiinggg? :)
[17:11] <tjaalton> (I hope you have i945)
[17:14] <tjaalton> there is also bug 307150 which has i945
[17:14] <ubottu> Launchpad bug 307150 in xorg "[jaunty] with kernel 2.6.28 my x hangs after 30 sec" [Undecided,New] https://launchpad.net/bugs/307150
[17:23] <albert23> tjaalton: see http://bugs.freedesktop.org/show_bug.cgi?id=19037
[17:23] <ubottu> Freedesktop bug 19037 in App/xrandr "Some libxrandr 1.2.99.2 requests cause hugely repeated EDID requests" [Major,New]
[17:23] <tjaalton> albert23: oh, cool
[17:23] <tjaalton> I didn't know it was reported
[17:23] <tjaalton> I'll mark mine as a dupe then
[17:25] <albert23> tjaalton: FWIW, I have no problems with the libxrandr update with server 1.6beta2
[17:26] <tjaalton> I didn't have either
[17:26] <tjaalton> with 1.5/1.6beta
[17:46] <tjaalton> albert23: btw, you are using the 'us' layout right?
[17:46] <tjaalton> with the new server
[17:47] <albert23> tjaalton: yes, us/intl
[17:48] <tjaalton> albert23: ok, so you are not seeing fdo bug 19048
[17:48] <ubottu> Launchpad bug 19048 in debian-installer "Installer gives unhelpful error messages when it can't read from the CD" [Low,Confirmed] https://launchpad.net/bugs/19048
[17:48] <tjaalton> freedesktop bug 19048
[17:48] <ubottu> Freedesktop bug 19048 in Server/general "xserver 1.5.99.3 fails to set the correct layout" [Normal,New] http://bugzilla.freedesktop.org/show_bug.cgi?id=19048
[17:48] <tjaalton> although you could test it :)
[17:52] <albert23> tjaalton: no, my keyboard is mostly ok.
[17:52] <albert23> But I do have miising keycodes in xsession-errors
[18:01] <albert23> tjaalton: switching to Nederlands seems to work fine.
[18:02] <tjaalton> don't do that from the gui, that works
[18:02] <tjaalton> you need to set  it in /etc/default/console-setup, restart hal & gdm and see what's used in the gdm login screen
[18:03] <tjaalton> and check the Xorg.0.log that the driver is using the layout
[18:08] <albert23> Tjaalton: after reboot GDM is still US. X-log and setxkbmap say I have Nederlands
[18:08] <tjaalton> albert23: touché
[18:08] <tjaalton> so it's not just me :)
[18:08] <tjaalton> hmm, I could git-bisect it
[18:22] <albert23> tjaalton: isn't gdm doing this? i.e. /etc/gdm/Xsession tries to set the keyboard map if ~/.Xkbmap exists. But I don't have ~/.Xkbmap.
[18:22] <tjaalton> me neither
[18:23] <tjaalton> besides it's the login screen
[18:23] <tjaalton> so it's using the system defaults
[19:26] <wgrant> tjaalton: Sorry, quickly-flattening battery and dodgy hotel wireless == not often checking IRC.
[19:29] <tjaalton> wgrant: heh, no worries
[19:30] <tjaalton> I'll upload the xserver to my ppa soonish, so jesse can start working on backporting intel modesetting :)
[19:31] <wgrant> Ooh shiny.
[19:31] <tjaalton> yes, apparently the backporting was accepted
[19:32] <tjaalton> keybuk just asked me the status of the intel driver
[19:33] <wgrant> Oh, I thought it was decided to keep the kernel in a PPA.
[19:33] <wgrant> ANd not backport KMS to .28...
[19:33] <wgrant> But I wasn't in the session.
[19:33] <tjaalton> this is probably newer information
[19:34] <wgrant> Excellent.
[19:34] <tjaalton> yeah, I'm happy, since KMS support is so cool
[19:34] <tjaalton> I've switched my gdm to vt1 already :Å
[19:34] <wgrant> Yep.
[19:35] <wgrant> Plymouth!
[19:35] <tjaalton> um, :P
[19:35] <wgrant> Aha.
[19:35] <wgrant> I have to mount my filesystems and do other booting stuff manually now, so I don't see the point.
[19:36] <tjaalton> well, you don't have gdm running yet, right?
[19:37] <wgrant> Right.
[19:38] <tjaalton> so it shouldn't cause any pain if that was to happen :)
[19:38] <wgrant> Right, but is there any benefit?
[19:38] <tjaalton> less flicker
[19:38] <tjaalton> oh, and we get to follow fedora
[19:38] <wgrant> I see.
[19:43] <tjaalton> well, maybe the benefit of doing so isn't that great when we still use usplash
[19:48] <tjaalton> since it'll be flickery for all users
[20:13] <tjaalton> hmm, I'll package intel-2.6-branch on my ppa as well
[20:21] <wgrant> tjaalton: That has KMS and DRI2?
[20:22] <tjaalton> wgrant: right
[20:22] <tjaalton> might make jesses work a bit easier
[20:23] <wgrant> When are we getting server 1.6 branch?
[20:23] <tjaalton> when the layout bug is fixed
[20:24] <tjaalton> the ppa will have it though
[20:28] <tjaalton> there
[21:00] <tjaalton> hum, intel wants a newer libdrm too, of course
[21:00] <tjaalton> well, lunch
[22:13] <tormod> tjaalton, arel you gonna push your mesa changes to git.d.o ?
[22:14] <tjaalton> tormod: sorry, done
[22:14] <tormod> oh thanks!
[22:15] <tjaalton> tormod: I've uploaded the new xorg-server to my ppa
[22:15] <tjaalton> and pushed the git changes now
[22:15] <tormod> cool, but the keymapping on 1.6 is still broken right?
[22:15] <tjaalton> the ppa branch is a local one, but essentially only the changelog has changed
[22:15] <tjaalton> right
[22:15] <tjaalton> otherwise it would be on the main archive
[22:15] <tjaalton> :)
[22:16] <tormod> your intel mentioned above, that's 2.5.1 right?
[22:17] <tjaalton> tormod: no, the 2.6-branch
[22:17] <tormod> I will make an intel-testing PPA and backport drm and intel for Hardy and intrepid, but not tonight, going skiing early tomorrow.
[22:17] <tjaalton> wanted to push that to my ppa, but it'd need a newer libdrm
[22:17] <tjaalton> heh, have fun
[22:17] <tormod> yeah 2.4.2
[22:18] <tormod> are you thinking of 2.6 in Jaunty soon?
[22:18] <tjaalton> well I'd like to have a libdrm release first
[22:19] <tormod> I have 2.6 (and 2.4.2ish) in xorg-edgers ppa
[22:19] <tjaalton> which is basically required because otherwise intel would be upset
[22:19] <tjaalton> ok, cool
[22:20] <tormod> oh btw that's 2.4.2 + modesetting-gem, but apparently that's too buggy
[22:20] <tormod> I will revert to 2.4.2 + the modesetting patch from Jesse on dri-devel
[22:20] <tormod> (if I get time before he commits it to master)
[22:21] <tjaalton> well eric commented that already, and there are some issues but not necessarily in functionality
[22:22] <tormod> also I instruct people to compile drm modules to match libdrm, but I think that's wrong or not needed at this point. libdrm/linux-core is not up to date for some reason.
[22:24] <tjaalton> they've changed the release process, so linux-core status is a bit unclear to me
[22:25] <tormod> shouldn't they either ditch or update it?
[22:25] <tjaalton> probably :)