[15:46] <Sarvatt> phew, so many sync requests :)
[15:46] <Sarvatt> BlackZ: thanks a ton for helping with the universe ones
[15:47] <BlackZ> Sarvatt: :)
[15:48] <Sarvatt> BlackZ: I really think we should wait for xterm 267 that fixes all the linking problems btw, it just got released
[15:49] <BlackZ> Sarvatt: is it in Debian?
[15:51] <Sarvatt> not yet, it *just* got released
[15:52] <BlackZ> Sarvatt: ok, then I will wait for the Debian maintainer to update it
[15:54] <BlackZ> Sarvatt: by the way, can you give a look at bug #677961 ?
[15:54] <ubot4> Launchpad bug 677961 in ppa-purge (Ubuntu) (and 1 other project) "Fails to deal with compressed indexes (affects: 8) (heat: 38)" [Undecided,New] https://launchpad.net/bugs/677961
[15:55] <Sarvatt> oh *fun*
[15:56] <BlackZ> Sarvatt: if you have a fix for it, please let me know
[15:57] <BlackZ> Sarvatt: otherwhise maybe I can give it a look tomorrow evening
[16:24] <Sarvatt> double argh, W: Failed to fetch http://extras.ubuntu.com/ubuntu/dists/natty/main/binary-i386/Packages.lzma  404  Not Found
[16:24] <Sarvatt> lzma too?
[16:48] <Sarvatt> tjaalton: so who volunteered to maintain this klingon delta in libX11 again so I can ping them? :)
[16:51] <jcristau> was from lifeless iirc
[16:53] <tseliot> Sarvatt, RAOF: what's the status of bug 650539 in Maverick?
[16:53] <ubot4> Launchpad bug 650539 in xorg (Ubuntu Maverick) (and 6 other projects) "SRU: Launching a Qt app crashes X when using Xinerama (affects: 88) (dups: 10) (heat: 518)" [High,Triaged] https://launchpad.net/bugs/650539
[16:53] <Sarvatt> needs upload desperately
[16:55] <tseliot> Sarvatt: upload or approval by the SRU team?
[16:56] <tseliot> or both?
[16:56] <Sarvatt> not sure if anyone prepared a package there but I know that fix is needed really badly
[16:58] <tseliot> Sarvatt: ok, I'll deal with it then
[17:10] <tseliot> Sarvatt: err... we don't have a branch for Maverick, do we?
[17:11] <tseliot> only one branch, right?
[17:11] <Sarvatt> nope
[17:11] <Sarvatt> yeah
[17:11] <tseliot> I'm asking as it seems that pitti uploaded 1:7.5+6ubuntu3b1 in Natty
[17:12] <jcristau> that doesn't sound like a server version
[17:12] <Sarvatt> xorg-server
[17:12] <tseliot> while we still have 1:7.5+6ubuntu3 UNRELEASED in git (???)
[17:12] <Sarvatt> thats the meta package
[17:12] <Sarvatt> patch is for xorg-server
[17:13] <Sarvatt> http://cgit.freedesktop.org/xorg/xserver/commit/?h=server-1.9-branch&id=3effb61e207478d92ebbcf5dfc75535cdd2dda12
[17:13] <Sarvatt> we'll get the fix soon in natty
[17:13] <tseliot> right, I pulled the wrong code and my eyes are a little tired...
[17:14] <tseliot> sorry for the noise
[17:15] <tseliot> hehe I was even surprised that git clone was so fast
[17:15] <tseliot> thinking that it was xorg-server...
[17:25] <Sarvatt> tseliot: no need to do it in git really, we've got a newer one in natty already and no maverick branch
[17:26] <tseliot> Sarvatt: yes, I figured it out after seeing the git log
[17:26] <tseliot> I'll just change the package
[17:44] <tseliot> Sarvatt: I assigned the task for Maverick to myself. I hope RAOF doesn't mind
[17:45] <Sarvatt> i'm sure he doesn't, thanks a ton tseliot!
[17:46] <tseliot> good :)
[19:22] <tjaalton> Sarvatt: yeah it was lifeless
[20:27] <Sarvatt> yay x11-apps compiles again http://sarvatt.com/downloads/patches/0001-x11-apps-Fix-linking-with-no-add-needed.patch
[20:42] <Sarvatt> bryceh: mind sponsoring http://sarvatt.com/downloads/merges/x11-apps/ if you get a chance?
[20:43] <Sarvatt> ah wait I should include the debian changelog updates in .changes
[20:43] <bryceh> Sarvatt, sure, I'll ask a small favor in return
[20:44] <Sarvatt> there we go http://sarvatt.com/downloads/merges/x11-apps/
[20:44] <Sarvatt> anything, what's up?
[20:45] <bryceh> Sarvatt, I was taking a look at 312756 (it generates ample bug mail to me).  I don't know that there's anything we can actually do about it other than wait on upstream.  But would you mind taking a quick look and tell me your opinion?
[20:46] <bryceh> it's the hybrid graphics hot swap issue
[20:46] <bryceh> currently the bug is having people submit their DSDT's but is that just busy work or is there actually something that can tangibly be done with it?
[20:47]  * bryceh goes a-sponsorin'
[20:47] <Sarvatt> there's nothing we can actually do, but I think having people upload their DSDT's to that bug in a central place is very useful
[20:47] <bryceh> how would the DSDT's be used?
[20:49] <Sarvatt> you can tell how the multiple GPU's are implemented on that platform, i've dug into dsdt's from there many times looking up laptops to recommend to people because some are hopelessly crappy but some you can potentially switch via that acpicall module and there's no way to know without looking at the dsdt
[20:50] <bryceh> )(x11-apps_7.6+2ubuntu1_source.changes uploaded
[20:51] <Sarvatt> is there any way to make wishlist bugs specifically not spam email maybe? :)
[20:51] <Sarvatt> thanks man!
[20:51] <bryceh> procmail ;-)
[20:52] <bryceh> I think what I'm going to do is set up a procmail rule to put ^Subject: .*MASTER: bugmails into launchpad.masters
[20:52] <bryceh> and I prepended MASTER: onto this one
[20:56] <bryceh> Sarvatt, well the only thing we need to think about when asking users "post blah to this bug" is that launchpad will eventually start timing out and throwing oopses when a bug exceeds a certain number of attachments
[20:57] <bryceh> [ubuntu/natty] x11-apps 7.6+2ubuntu1 (Accepted)
[20:57] <Sarvatt> darn, launchpad is that broken?
[20:58] <Sarvatt> I used that bug extensively when shopping for a new laptop for instance :D
[21:09] <bryceh> yeah we had a similar bug report where people posted data about unsupported joystick buttons, that eventually hit the limit
[21:09] <bryceh> I encountered it due to my arsenal scripts choking on it when processing through bugs
[21:10] <bryceh> again it was another situation where there was nothing we could do, it was a kernel / protocol limitation
[21:10] <bryceh> iirc
[21:10] <Sarvatt> I still reference that evtouch lshal bug to tell people quirks for their touchpads :D
[21:10] <bryceh> can you load it?
[21:10] <Sarvatt> I could last week when I did last, hmm
[21:11] <bryceh> well possibly launchpad fixed the problem, but it would oops regularly for me when I looked into the oops a few months ago
[21:11] <bryceh> oh you know what, maybe the did fix it
[21:11] <bryceh> I was bugging people left and right about that issue when I was on launchpad
[21:11] <Sarvatt> https://bugs.launchpad.net/ubuntu/+source/xf86-input-evtouch/+bug/317094?comments=all works fine
[21:11] <ubot4> Launchpad bug 317094 in xf86-input-evtouch (Ubuntu) "meta bug to collect lshal touchscreen info (affects: 44) (dups: 2) (heat: 316)" [Undecided,Won't fix]
[21:11] <Sarvatt> maybe they were playing with timeouts?
[21:12] <bryceh> no one had a clue at the time, but I recall lifeless kept working on it
[21:12] <Sarvatt> https://edge.launchpad.net/ubuntu/+archive/primary/+copy-packages?field.name_filter=linux&field.status_filter=published&field.series_filter=natty  used to timeout oops 99% of the time for me, but its been ok lately
[21:14] <Sarvatt> speaking of which, 5th kernel copy in a row that worked fine, amazing!
[21:14] <Sarvatt> for 6 months it would oops every time but actually do the copy
[21:15] <Sarvatt> OOPS-1787ED1419
[21:15] <ubot4> https://lp-oops.canonical.com/oops.py/?oopsid=1787ED1419
[21:16] <Sarvatt> didn't like searching for linux packages in sid :)
[21:29] <bryceh> interesting; they're changing the wayland license from MIT to LGPLv2
[22:03] <ilmari> bryceh: URL?
[22:05] <Sarvatt> http://lists.freedesktop.org/archives/wayland-devel/2010-November/000249.html
[22:06] <pcjc2> Hi guys, how are things in Intel-gfx land on Ubuntu?
[22:07] <pcjc2> I've been testing for a while on maverick + Xorg-edgers + Natty kernel + drm-intel-next DRM backport, and all is really good
[22:09] <Sarvatt> the usual, many regressions creeping in and getting fixed up shortly after on the kernel side lately :) maverick's 2.12 still has problems with the copy-fb patch, intel GPU's are still slow as dirt, nothing too exciting
[22:12] <Sarvatt> natty wont get "fun" until around christmas when mesa 7.10 goes in
[22:13] <Sarvatt> although compiz 0.9.2.1 that's in here is already lots of "fun"
[22:13] <pcjc2> Using the drivers on a daily basis and tracking kernel changes, I can pretty much pinpoint the regressions, ping Chris Wilson, and he fixes them in about 2 hours or less ;)
[22:13] <pcjc2> "fun" as in bad?
[22:13] <Sarvatt> yeah he's awesome like that, I agree :)
[22:13] <pcjc2> xorg-edges mesa tracks master, right...? That seems pretty good to me
[22:13] <Sarvatt> yeah
[22:14] <pcjc2> shader compiler is pretty rubbish, but I can't see that being my bottleneck yet
[22:14] <Sarvatt> well you probably dont use things like kwin where the fun shows up :)
[22:14] <pcjc2> I have too many damned pixels on screen to get decent framerate, even with glClear
[22:14] <pcjc2> I have occasionally turned compiz on, but tend to switch it back off to do performance testing
[22:15] <pcjc2> What surprises me is that the desktop, compositing, etc.. don't "seem" slow, yet one measly NOOP render running a glClear once per frame can't beat 60fps full-screen (with sync to VBLANK disabled)
[22:16] <Sarvatt> been focused on sandybridge here and its fun finding out all my hardware has errata making it useless on the graphics side
[22:16] <pcjc2> Still trying to find the magic bullet to cure that, in my search have just ordered some more 800Mhz ram (had a 667Mhz chip in there)
[22:18] <Sarvatt> sheesh, mesa is a blast to work with sometimes eh?
[22:18] <pcjc2> yep
[22:18] <pcjc2> Canonical has some Sandybridge pre-production machines to play with?
[22:22] <Sarvatt> yeah, it's surprisingly decent too (just not on linux) :)
[22:23] <pcjc2> from the sounds of the development work, the architecture has changed a lot
[22:29] <Sarvatt> pcjc2: what GPU are you using where things are fine on edgers?
[22:29] <pcjc2> GM45
[22:30] <pcjc2> But note the drm-intel-next kernel backport
[22:30] <Sarvatt> interested because I've seen some reports of people having to disable page flip support still and we dropped the disabling patch in natty (it's been disabled in edgers since 2.12)
[22:30] <pcjc2> woot.. magic switch
[22:30] <pcjc2> (One I'd found before)
[22:30] <pcjc2> sudo intel_reg_write 0x21D0 0x1000207
[22:30] <pcjc2> 59.6 redraws per second -> 82.8 redraws per second
[22:30] <pcjc2> with one BIT of one ECO register
[22:32] <pcjc2> oh, no.. wait - it has gone back again.. WHY must things be so variable
[22:33] <pcjc2> I know that bit has bumped performance for me before.. it disables some clock-gating on the chip's rendercache. It was a documented 965 ECO, but still seems to do "something" in GM45.
[23:04] <nigelb> bryceh: hi, need your take on a bug
[23:05] <nigelb> bryceh: bug 325581, the patch needs to go into kernel or into xorg server?
[23:05] <ubot4> Launchpad bug 325581 in xserver-xorg-input-evdev (Ubuntu) (and 1 other project) "kensington pocket mouse model #72237 USB 0d62:1000 not working under 8.10 (affects: 1) (heat: 17)" [High,Confirmed] https://launchpad.net/bugs/325581