[00:04] RAOF, got it, thanks. === Amaranthus is now known as Amaranth [07:36] bryce: I'm thinking about defaulting to sna now, ickle recommended it too [07:54] so robert ancell uploaded weston 1.0.3 already.. [07:54] double work [08:35] tjaalton, do we have sna on in edgers? [08:37] bryce: dunno [08:37] but people have tested it with good results [08:40] doesn't look like it's enabled there [08:42] one lesson I learned with exa and uxa is that before we switch to a new tech you only hear from people for whom the tech works great. Those who have problems with it will just ignore it... [08:43] sure [08:43] tjaalton, I suppose we could switch it on for a week and evaluate the fallback via bug reports [08:43] yeah [08:43] (assuming apport is still catching crashes and freezes for us) [08:43] it's been rather quiet :) [08:43] suspiciously [08:44] 10 bugs on raring.. [08:44] against our packages [08:44] which is like 1/4 of the bugs against quantal at this point [08:44] yeah. of course we haven't updated many packages... [08:44] anyway, I have raring systems at hand, I'll do some sleuthing tomorrow and make sure it's still working. [08:45] I've tried it on sandybridge, heard good results on i915 & i945 [08:45] for sna... even if we go ahead and do a 1 week eval, I think I'd still like to see it guinea pigged on edgers for a bit, maybe even a call for testing [08:46] edgers pulls all kinds of crap [08:46] rather do it on another ppa [08:47] yeah that'd be fine. [08:48] or even just have folks manually switch it on in their xorg.conf? [08:48] "Copy this file to /etc/X11/xorg.conf. If that leads to any graphics problems, delete the file and let us know." [08:49] give it a week... if we hear nothing serious then flip it on for everyone for a week [08:50] alright [08:52] huh, reading the forums really makes you sad :) [08:56] phoronix? [08:56] ubuntuforums this time [08:58] sheesh: http://www.phoronix.com/scan.php?page=news_item&px=MTI3MTI [08:58] yeah [08:58] that was amusing to read, again [09:23] tseliot: hey, wanted to ask if nvidia-current will go away and instead just use -NNN in the future? [09:27] tjaalton: yes, it's exactly as you say [09:27] tseliot: right, what I had suspected :) [09:27] I'll work on transitional packages today [09:27] pinged #ubuntu-release about the -173 update to quantal, noticed it's been on the queue since nov 22nd :/ [09:28] ah, thanks [09:35] tseliot: also, during your holidays I uploaded the nvidia-96 update, it's now verification-done so should hit precise soon [09:36] tjaalton: ah, good, I'll sync my git repository then [09:37] hope it's the last upload ;) [09:37] hehe, right [09:40] will the nvidia-experimental-* packages stay? [09:43] guess they're obsoleted by this change [09:48] tjaalton: they should be there only in precise [09:48] right === yofel_ is now known as yofel === Quintasan_ is now known as Quintasan [11:14] bryce: oh my ;P [11:15] my personal favorite is the tf2 post currently, he didn't even try and copied that verbatim [14:25] tjaalton: June 29th, 2012: SNA is now on by default on intel. [14:26] its been on by default for a long time now and the only complaints since have been when I accidentally made UXA the default one upload :P [14:28] Sarvatt: --enable-sna is not the same as --with-default-accel=sna [14:28] oh you mean edgers? [14:28] the changelog lies about it then [14:28] override_dh_auto_configure: [14:28] dh_auto_configure -- --enable-sna --enable-uxa --with-default-accel=sna --with-builderstring="$(SOURCE_NAME) $(SOURCE_VERSION) ($(BUILDER))" [14:28] basically doesn't mention that [14:29] and I trusted the changelog, boo :) [14:29] oh the hook got screwed up [14:29] anyway, I'll create a wikipage for people to report success on [14:29] changelog part, i updated the man page last upload and screwed it up :) [14:29] hehe [14:29] 16:21 < cjwatson> This will be painful [14:30] 16:22 < cjwatson> hate you all [14:30] :) [14:30] colin doesn't like the pain the renamed stack causes on the installer [14:30] wait a minute [14:30] relax [14:31] everyone hates the renamed stack, not like we had a choice :) [14:31] right [14:39] true [14:39] guess I'll file a bug for the mesa -dev issue and get it sru'd [14:42] yeah [15:15] https://bugs.launchpad.net/ubuntu/+source/mesa/+bug/1098215 ? [15:15] Ubuntu bug 1098215 in mesa (Ubuntu Precise) "installing libqt4-opengl-dev uninstalls renamed stack" [Critical,In progress] [15:15] will do some more testing before I upload the fix [15:17] yep, looks good [15:17] don't forget to subscribe ubuntu-sru once done uploading [15:19] hm I think I'll leave out libgbm-dev, I think the api on that one might change and nothing rdepends on it anyway [15:20] ok [16:08] https://wiki.ubuntu.com/X/Testing/IntelSNA [16:08] comments welcome [16:09] or edits :) [16:09] dinner -> [16:09] sounds good [16:47] updated the wiki-page again [16:50] I'd like to send the email to ubuntu-devel@ today [16:51] then again, the wiki can be edited later :) [16:56] Sarvatt: mind proofreading the page? [16:57] actually, I'll wait and let bryce comment on it too [16:58] bryce: so I wrote the wiki page, and email ready to be sent [17:00] tjaalton: sure thing, one sec, testing this bcmwl package out [17:21] tjaalton: looks good to me, is SNA backend needed? [17:21] in the results column [17:22] I guess thats one way to make sure people verify they are actually using SNA [17:23] yeah :) [17:24] wouldn't xorg.0.log say so? [17:24] although it comes from the same output, so.. [17:24] that's where it's from [17:27] dropped the backend bit [18:26] tjaalton, reading [18:46] tjaalton, looks good! I made a few minor copy edits. Also, I put the xorg.conf as an actual attached file. [18:46] I suspect people are going to be more comfortable copying an entire file into place, than with editing xorg.conf's. And most Intel users won't have a file to conflict. [18:47] bryce: ah, good point. and thanks! [18:47] the first version just cat'd the file in place, but that seemed a bit risky :) [18:47] right, now to retest on my own hw. :-) [18:47] Sarvatt: fix your lp id there :) [18:48] tjaalton, ah, I assumed those were yours! sorry [18:48] oh, hehe [18:50] yeah I'll test it on X61 too once I'm done with vag-com & XP.. [18:51] (i965GM) [18:59] bryce: i'm reformatting it a bit, done editing? [19:00] yep, go for it [19:04] added example output for uname/lspci [19:05] ok, I'll send the email and post to the forum [19:05] tjaalton, awesome [19:07] just happened to read this earlier too :) http://ubuntuforums.org/showthread.php?t=2103414 [19:09] there [19:10] let's see how long until it's on phoronix frontpage [19:10] heh [19:10] tjaalton: 'lspci -nn -s 0:2.0' [19:10] (all intel hw has the gfx at that slot afaik) [19:11] jcristau: ahh.. [19:11] saves some grepping [19:11] and the chance of running it on a hybrid laptop [19:11] didn't know how to best handle such case [19:12] if it's always on 0:2:0 then that would solve it [19:12] tjaalton, there's also xpci, although I can't guarantee it's going to work properly on the newest gfx cards [19:13] ImportError: No module named taskhelm [19:13] :) [19:13] on quantal [19:13] tjaalton, yeah, need to fix that. should already work on raring though [19:13] ok [19:36] tjaalton: silly question, but could this http://wstaw.org/m/2013/01/10/plasma-desktopS17003.png be caused by enabling SNA? [19:37] * shadeslayer didn't see this earlier [19:37] ( 2 arrows in the taskbar, the one at the bottom is bogus ) [19:37] shadeslayer: no idea, try without to be sure [19:38] yeah will do that to be 100% sure [19:41] raring has a new pixman too [19:43] no idea what that is [19:43] I also don't have proper instructions to reproduce the bug, so will get back if I notice it with UXA [19:44] ah okay [19:44] can reproduce it in UXA as well [19:44] ok, check what version of libpixman-1-0 you have [19:45] 0.26.0-3 [19:46] not that then [19:46] same in quantal [19:46] yep, haven't upgraded in a bit [19:47] there's a new intel driver too, uploaded earlier this week [19:48] yeah, I guess I'll leave it to upgrade tonight [19:48] added a new first step to the page, upgrade to current raring [19:48] :D [19:49] details.. [21:57] first corruption bug with sna :) [22:13] * bryce installs chromium [22:17] huh, repro'd [22:18] what hw? [22:19] 2a42, same gfx as echidnaman [22:20] Dell Latitude 13 [22:21] ok, I'll try on snb [22:22] not happening on snb or ivb here [22:23] not here eitehr [22:23] either [22:23] shadeslayer: which hw did you have? [22:24] 2nd Generation Core Processor Family Integrated Graphics Controller [8086:0116] (rev 09) [22:24] ah, so it's not the same bug [22:24] posted my screenshot @ #1098334 [22:24] To Phoronix's great surprise, sna isn't a flawless beacon of speed and stability? ☺ [22:24] well, not on every hw incarnation ;) [22:24] it seems [22:25] RAOF, I'm sure after we switch to it, they'll post an article lambasting us [22:27] oh oh [22:27] hmm, seems to be peculiar to reddit; other sites don't exhibit the behavior. doesn't seem to affect firefox [22:27] tjaalton: bug 1098334 has been reproduced by JT [22:27] one sec [22:27] bug 1098334 in xserver-xorg-video-intel (Ubuntu) "Font corruption in Chromium tab bar using Intel SNA" [Undecided,New] https://launchpad.net/bugs/1098334 [22:27] ah he reported it @_@ [22:27] right :) [22:28] cannot reproduce on gooogle-chrome btw [22:28] the proprietary one [22:28] but could easily be because of hw differences [22:28] ok, so this one is not a critical one then [22:29] probably happens only on gen4 [22:30] or the ones using "Broadwater/Crestline" backend [22:43] yeah I set it to medium [22:44] it's a pretty common chipset, although a tad old [22:46] hey don't call my laptop old. it's just 4 years old. :) [22:50] ah, gm45 is actually newer than 965 [22:51] yes [22:53] somehow mixed it with 945 [22:55] http://en.wikipedia.org/wiki/Intel_Extreme_Graphics#Early_graphics [23:55] is it just me or is https://launchpad.net/builders private now?