[00:01] <bryce> ok good to know
[00:02] <bryce> I've filed https://bugs.edge.launchpad.net/ubuntu/+source/acpi-support/+bug/372480 to track the vbetool/kms issue in ubuntu
[00:02] <ubottu> Ubuntu bug 372480 in acpi-support "vbetool should not be called on -intel when KMS is in use" [High,New]
[00:05] <bryce> "Known Issues" section added to https://wiki.ubuntu.com/X/KernelModeSetting with this info
[00:05] <bryce> s/info/bug report/
[00:06] <jbarnes> cool
[00:08] <Sarvatt> well it sounded like debian's acpi-support uses vbetool on bootup instead of just those calls in suspend/resume like ubuntu's, i dont have any problems with KMS here but that guy's i915 is using MSI so i'm guessing its 965+ and i'm on 945GME
[01:10] <bryce> jbarnes: according to slangasek -
[01:10] <bryce> In karmic, acpi-support no longer ships its own suspend/resume handling
[01:10] <bryce> at all, so this must already be fixed.
[01:19] <jbarnes> cool
[01:31] <mnemo> jbarnes: do you know if it's -psb that's being submitted by intel for xorg mainline?
[01:31] <jbarnes> mnemo: I'm not sure... I've asked the submitter to clarify
[03:06] <bryce> http://www.h-online.com/open/Ubuntu-9-04-and-Intel-graphics--/features/113196
[03:16] <LLStarks> bryce. uxa has finally stabilized.
[03:17] <LLStarks> 48+ hours on compiz without a crash.
[03:30] <Sarvatt> yeah it stabalized for me somewhere between the 2.7.99.1 snapshot and e5e0fb846bda3e1757b89e50d5244d28457b9fe3 Call down to lower CloseScreen before shutting down DRM allocator
[03:31] <Sarvatt> not many commits between those
[03:41] <bryce> good to hear, although since 2.7.99.x lacks EXA, I'm not ready to upgrade us to that quite yet
[03:42] <bryce> bbl
[04:11] <Sarvatt> i saw a few of the commits between those two lined up for 2.7.1 that might be the ones that fixed it
[04:13] <Sarvatt> doh karmic update just now totally screwed up my fonts :D
[04:19] <LLStarks> sarv
[06:15] <Duke`> mesa-common-dev_7.5.0~git20090505 conflicts with libglew1.5-dev about the file "/usr/include/GL/glxew.h"
[10:24] <Sarvatt> wish i would have known jaunty changed mkdir /var/run/network to a udev rule before i restarted my vps without udev, that was an adventure
[10:24] <Sarvatt> oops, wrong channel, sorry
[16:58] <jbarnes> apw: ping
[16:58] <apw> hello
[16:58] <jbarnes> apw: hey mchbar patches? :)
[16:59] <apw> heh damn i remembered to send them out to our kernel list todaqy
[16:59] <apw> and forgot send them out to you
[16:59] <jbarnes> np... can you send them out to intel-gfx and cc bjorn so all the right bits get picked up?
[16:59] <jbarnes> with your tested-by, signed-off-by etc
[17:00] <apw> yeah will sort them out now
[17:00] <jbarnes> cool thanks
[17:05] <bryce> heya guys
=== kees_ is now known as kees
[18:20] <federico1> tseliot: ping
[18:20] <tseliot> federico1: pong
[18:20] <federico1> tseliot: hey
[18:21] <federico1> tseliot: mclasen tells me you were looking for me?
[18:21] <tseliot> federico1: yes, but I misread his patch which I think is ok now
[18:22] <federico1> ah, ok
[18:22] <tseliot> yes, nothing to worry about
[18:23] <federico1> I screwed up when merging your patch for GetResourcesCurrent(), and missed how the sizes were updated
[18:23] <federico1> ok, cool
[18:23] <federico1> tseliot: by the way, I recall seeing a fix for when "Laptop" appears as "Laptop\n" (with an extra newline) --- do you know something about that?
[18:24] <tseliot> federico1: isn't that a string that we use when we find LVDS, etc. ?
[18:26] <federico1> yeah
[18:27] <federico1> gnome-display-properties shows "Laptop" in the little monitor boxes, and it's also what appears in the per-monitor labels
[18:27] <federico1> and I *think* we have an extra newline sneaking in, because when it actually says "Laptop", the rectangles are taller than normal
[18:27]  * federico1 goes to RTFS
[18:28] <tseliot> it could be. I've never noticed the problem
[18:29] <federico1> ok
[18:46] <federico1> oh,duh
[18:47] <federico1> tseliot: return g_strdup_printf ("%s\n", vendor);
[18:47] <federico1> no wonder :)
[19:42]  * jbarnes waits patiently for apw's patches to show up
[19:42]  * apw is smaching git-send-email with hammers as we speak
[19:42] <tseliot> federico1: hehe, there was a new line then ;)
[19:42] <apw> it is not playing along
[19:42] <jbarnes> ah yeah it can be a pain
[19:42] <jbarnes> but it's just two patches right?  just use git format-patch and send them by hand :)
[19:49] <apw> jbarnes, right in the pipe.  bloody thing
[19:51] <jbarnes> yay
[19:51] <jbarnes> thanks
[21:06] <tormod> michaellarabel: what is PTS live based on?
[21:06] <michaellarabel> tormod: PTS Linux Live is going to be based upon Ubuntu
[21:07] <tormod> is the development open?
[21:07] <michaellarabel> It's going to be developed openly, yes. Right now it's not too far along.
[21:08] <tormod> will it provide latest stuff from xorg on a regular basis?
[21:09] <tormod> I am bit curious because I have been thinking of making an xorg-edgers based live CD
[21:09] <michaellarabel> It will be updated quarterly in conjunction with PTS and yes, I plan to include all of the latest Xorg and kernel bits
[21:10] <tormod> nice. are you alone on this, or is there a team?
[21:11] <tormod> is Phoronix Media = you?
[21:12] <michaellarabel> On the LiveCD? There is me plus two or three others that have voiced interest and were going to start on different things, but so far I haven't seen any work from them.
[21:12] <michaellarabel> Involved with PTS in general is I plus Whuppermann plus about 12+ other random contributors and companies. Yes, I own Phoronix Media.
[21:13] <tormod> thanks for the info
[21:13] <michaellarabel> Yep, there is also #phoronix on FreeNode if you have more questions.
[21:15] <tormod> oh, I should have asked there then :)
[21:16] <bryce_> (I found the answers interesting as well)
[21:23] <hallyn> is the issue with gnome-terminal not updating with compiz on (with nvidia drivers) well understood?
[21:23] <kirkland> bryce_: i directed hallyn here
[21:23] <kirkland> bryce_: any idea?
[21:23] <hallyn> is there any solution (even just in the pipeline) other than turning off compiz
[21:25] <bryce_> hallyn: dunno, I don't really do much with the proprietary -nvidia drivers
[21:25] <bryce_> hallyn: you could look in launchpad to see if others have encountered the same problem and engineered a workaround
[21:26] <kirkland> bryce_: anyone better i can point hallyn to?
[21:26] <kirkland> hallyn: http://people.ubuntu.com/~kirkland/search.html?cof=FORID%3A9&cx=003883529982892832976%3Aly2fmeg302s&q=nvidia+gnome-terminal&sa=Search
[21:26] <bryce_> kirkland: tseliot generally does more with -nvidia, although I think he's not had a lot of time for bug work lately
[21:26] <hallyn> ok, i was hpoing it wasn't actually the nvidia drivers
[21:26] <hallyn> was hoping it was just gnome-terminal not defining some hook or something :)
[21:27] <hallyn> kirkland: ok, thanks, i'll look through those.  meanwhile in the interest of getting work done compiz is finally off :(
[21:27]  * hallyn glances over at evilwm
[21:27] <hallyn> maybe it's time to go back...
[21:28] <bryce_> I'm not the gnome-terminal maintainer but I use it all the time on various open source drivers and have never seen or heard of that issue
[21:28] <hallyn> ok, thanks.  then i'll consider it my fault for being an evil user of proprietary drivers
[21:28] <bryce_> but -nvidia is sort of an X world unto itself (it reimplements half the X server inside itself) so who can say
[21:28] <bryce_> hallyn: you could try -nouveau
[21:28] <hallyn> (but the free drivers make my laptop run waaaay hot so not acceptable)
[21:28] <hallyn> what is that?
[21:28] <bryce_> that is the new open source solution for nvidia hardware
[21:29] <bryce_> it's very new, but we're accepting bug reports against it now
[21:29] <bryce_> new == "probably buggy"
[21:29] <hallyn> will look.  thanks.
[22:40] <bryce_> http://www2.bryceharrington.org:8080/X/symptoms_intel.html
[22:43] <jbarnes> bryce_: nice, don't like all those "freeze" and "crash" symptoms though
[22:43] <jbarnes> i.e. there are way too many of them :/
[22:43] <bryce_> oh if you don't like 'crash', take care not to look in xorg-server ;-)
[22:43] <bryce_> (most crashes get auto-filed to xorg-server; we've got hundreds)
[22:44] <jbarnes> ouch
[22:44] <bryce_> the good news is most of those have detailed full backtraces associated with them
[22:45] <michaellarabel> bryce_: Soon I will have new -intel results from xorg-edgers and latest karmic packages.
[22:45] <bryce_> michaellarabel: thanks, that will give us some insights into if the stuff in xorg-edgers is going to improve or regress us
[22:46] <bryce_> michaellarabel: btw, we also have -intel 2.7.x in the x-updates PPA, which is what I will probably be pulling in next (maybe tomorrow, maybe next week)
[22:46] <mnemo> bryce: i think we should at least split crashes into xorg crashes and app crashes... currently we have some rss-glx app crashes with this tag.. i.e. https://bugs.launchpad.net/ubuntu/+source/fglrx-installer/+bug/344604
[22:46] <ubottu> Ubuntu bug 344604 in fglrx-installer "biof crashed with SIGSEGV in exit()" [Medium,New]
[22:53] <bryce_> mnemo: I know... but it's not easy to split them out
[22:54] <bryce_> or I should say, it's difficult to come up with a way for the scripts to tell the difference
[22:54] <bryce_> mnemo: I guess they could be cleaned up by hand (although be careful to also remove 'crash' from the subject, so some script doesn't re-add the 'crash' tag later)
[22:58] <mnemo> yea its not like we have a lot fo those anyway
[22:59] <mnemo> bryce_: btw.. did you hear about GPU lockup debugging for radeon is coming through the radeon rewrite branch
[22:59] <mnemo> thats pretty cool
[22:59] <mnemo> maybe we can integrate it with apport later for both intel and ati
[22:59] <bryce_> hadn't heard that.  very good to hear
[23:00] <mnemo> http://lists.freedesktop.org/archives/xorg/2009-May/045486.html
[23:00] <mnemo> its on a branch still though --> http://jglisse.livejournal.com/1822.html
[23:00] <mnemo> so it might be long before we see it in ubuntu
[23:00] <bryce_> kewl
[23:04] <Sarvatt> tormod's trying to get that running in the radeon crack ppa - https://launchpad.net/~xorg-edgers/+archive/radeon 
[23:05] <tormod> well that stuff sits in the kernel, so I am not sure. maybe through drm-snapshots
[23:06] <tormod> (the lockup debugging I mean)
[23:10] <michaellarabel> bryce_: 2D is running faster with the latest karmic and xorg-edgers, though OpenGL may have a problem. Waiting for the tests to finish to see what the logs say.
[23:12] <mnemo> michaellarabel: are you comparing to intrepid as well?
[23:12] <michaellarabel> mnemo: I may, once I see what these karmic results look like.
[23:13] <mnemo> once we get out of the woods with the UXA bugs and enable it by default in karmic, it would be very interesting to see results against intrepid
[23:16] <michaellarabel> bryce_ and tormod and anyone else interested in the latest: http://global.phoronix-test-suite.com/?k=profile&u=phoronix-27748-28806-30895 though yeah, several of the tests are failing with the latest karmic/xorg-edgers right now. Though surprisingly 2D did worse in some.
[23:17] <tormod> michaellarabel: it says "Mesa 7.4", did you not upgrade mesa?
[23:18] <michaellarabel> hmm I upgraded everything, but will check for some reason if it got left behind.
[23:18] <bryce_> michaellarabel: interesting... I wonder how much of the improvement is due to the new driver, and how much to the kernel.  I've been suspecting it is mostly the latter...
[23:19] <michaellarabel> tormod: Ah, you're right, for some reason Mesa got left behind. But upgrading that now.
[23:19] <bryce_> michaellarabel: also, I assume both cases are with UXA enabled?
[23:19] <michaellarabel> defaults in both case, so EXA for 9.04.
[23:19] <bryce_> oh wait, I see EXA
[23:21] <tormod> so 9.04 is the better on all test except the Rects Blended?
[23:22] <michaellarabel> tormod: Transformed Blit Bilinear did better on the latest
[23:22] <michaellarabel> albeit very small
[23:42] <michaellarabel> tormod: Yeah, Mesa is working now, though hitting another bug.
[23:42] <tormod> bug like in crasher?
[23:43] <michaellarabel> Once OpenArena is done running at the final screen, it now locks up.
[23:56] <michaellarabel> smokin-guns is at least running, will have those numbers soon