| [00:00] <Q-FUNK> especially since a lot of people already expect this to work in gutsy | |
| [08:35] <tjaalton> I guess the xresprobe bugs can be marked as wontfix now | |
| [08:36] <bryce> true | |
| [08:36] <bryce> I can take care of that tomorrow if you'd like | |
| [08:37] <Q-FUNK> hm? | |
| [08:37] <Q-FUNK> what about xresprobe? | |
| [08:37] <tjaalton> Q-FUNK: obsolete | |
| [08:37] <bryce> it's deprecated now | |
| [08:37] <Q-FUNK> replaced by? | |
| [08:38] <tjaalton> xserver autoconfiguration | |
| [08:38] <Q-FUNK> ah yes | |
| [08:38] <tjaalton> bryce: there are only 11 of them, no big deal :) | |
| [08:38] <tjaalton> I need more karma :) | |
| [08:38] <bryce> hehe | |
| [08:38] <bryce> ok go for it | |
| [08:38] <Q-FUNK> although this doesn't yet work for all drivers. they need to be ported to randr 1.2 and upgraded to recent api. | |
| [08:39] <tjaalton> hmm no | |
| [08:39] <tjaalton> it doesn't depend on randr 1.2 | |
| [08:39] <tjaalton> although there are some issues with vesa at least | |
| [08:40] <Q-FUNK> and with -amd | |
| [08:40] <Q-FUNK> siliconmotion also | |
| [08:40] <bryce> what issues? | |
| [08:40] <tjaalton> "no modes left" with vesa | |
| [08:41] <bryce> ah | |
| [08:41] <Q-FUNK> for DDC probing, bartman produced patches that should at least fix -amd. | |
| [08:41] <Q-FUNK> that sound familiar from other drivers too. | |
| [08:41] <Q-FUNK> IIRC someone reported that for -amd as well, on some hardware. | |
| [08:41] <tjaalton> trident is buggy too, it hangs in a loop | |
| [08:41] <tjaalton> but it's upstream now with a backtrace | |
| [08:42] <Q-FUNK> ah, btw, amd 2.7.7.5 is in debian now | |
| [08:42] <Q-FUNK> it introduces basic OLPC support | |
| [08:42] <tjaalton> ok, file a sync request again, and tell the bug # | |
| [08:44] <Q-FUNK> grmbl. doesn't yet show in changelog server. | |
| [08:48] <tjaalton> there, 11 bugs closed | |
| [08:48] <Q-FUNK> #182781 | |
| [08:49] <Q-FUNK> hmm | |
| [08:49] <Q-FUNK> this is in main, right? | |
| [08:49] <Q-FUNK> or is it still in universe? | |
| [08:50] <tjaalton> main | |
| [08:50] <ubotu> New bug: #182778 in xserver-xorg-video-openchrome (main) "Wrong resolution on Hardy Alpha 3" [Undecided,New] https://launchpad.net/bugs/182778 | |
| [08:51] <tjaalton> heh, add openchrome to the list | |
| [08:51] <tjaalton> there are also a lot of reports about not being able to change the resolution on livecd | |
| [08:51] <tjaalton> which puzzles me | |
| [08:52] <bryce> hrm | |
| [08:52] <tjaalton> but that's an old issue | |
| [08:52] <bryce> "not change" as through xorg.conf, or via displayconfig-gtk | |
| [08:52] <bryce> ? | |
| [08:52] <tjaalton> xrandr or d-g | |
| [08:52] <bryce> odd | |
| [08:53] <tjaalton> yep | |
| [08:53] <tjaalton> maybe #u-installer could help, I'll ask | |
| [08:55] <ubotu> New bug: #182781 in xserver-xorg-video-amd (main) "Please sync xserver-xorg-video-amd 2.7.7.5-1 (main) from Debian unstable (main)" [Wishlist,Confirmed] https://launchpad.net/bugs/182781 | |
| [09:19] <bryce> tjaalton: heh, we've triaged too many bugs. bdmurray has skipped us in favor of openoffice for bug hug day this month | |
| [09:19] <tjaalton> :) | |
| [09:20] <bryce> looks like between xorg, xserver, and major drivers, we barely have a total of 100 untriaged bugs left ;-) | |
| [09:20] <bryce> actually probably more like 50 | |
| [09:21] <tjaalton> xorg has maybe 6 | |
| [09:21] <bryce> -nv has the most ironically | |
| [09:21] <tjaalton> heh, yeah | |
| [09:21] <tjaalton> I should go through them now that I have one.. | |
| [09:25] <bryce> so I wonder what we do when we hit 0 New bugs :-) | |
| [09:26] <tjaalton> we try to fix those that are confirmed? :) | |
| [09:26] <bryce> maybe focus on getting everything filed upstream? | |
| [09:26] <bryce> hehe | |
| [09:26] <tjaalton> going through b.df.o is one | |
| [09:27] <bryce> I got one of my upstream reported bugs marked invalid because it was one of those bugs that a bunch of users had "me too"ed their issues onto, so I summarized and sent upstream | |
| [09:27] <bryce> however upstream seems to prefer each person's issue be a separate bug report | |
| [09:27] <tjaalton> as do we | |
| [09:27] <bryce> unfortunately I think launchpad lets us down a bit here, since it's set up to only allow one upstream bug report per bug | |
| [09:28] <bryce> hmm, I've not been so picky that people who have roughly the same issue, report it separately (sometimes) | |
| [09:28] <tjaalton> i've closed bugs where people have reported other issues than the original one, and asked them to file new bugs | |
| [09:28] <bryce> maybe I need to be more strict though. | |
| [09:29] <Ng> I'm going to be bugging you X guys again like I was with gutsy. resuming my laptop is causing graphical issues again ;) | |
| [09:29] <bryce> I suppose that's a good practice | |
| [09:29] <bryce> heya Ng | |
| [09:29] <bryce> heh, the membership of #ubuntu-x sure has grown since I started working here.... | |
| [09:29] <tjaalton> bug 180409 | |
| [09:29] <ubotu> Launchpad bug 180409 in linux-restricted-modules-2.6.24 "bugs in fglrx - blank screen - artifacts" [Undecided,Invalid] https://launchpad.net/bugs/180409 | |
| [09:30] <Ng> hey bryce :) | |
| [09:30] <tjaalton> that's one good example of a messy bug report.. | |
| [09:30] <tjaalton> hey Ng! | |
| [09:30] <Ng> tjaalton :) | |
| [09:32] <Ng> erk, this is more broken than I thought. scrolling in firefox => jumbled mess | |
| [09:32] <Ng> bryce: do you know if the fix from 133118 made it upstream? | |
| [09:32] <bryce> I was fairly sure it did | |
| [09:32] <Ng> I don't think it can be the same thing, because Im sans compiz atm and 2d stuff is broken | |
| [09:32] <tjaalton> Ng: switch to XAA, set 'Option "AccelMethod" "XAA"' | |
| [09:33] <Ng> tjaalton: I did that already to get the 2d acceptably fast ;) | |
| [09:33] <bryce> man, that's sad that I recognize bugs off their bug id# | |
| [09:33] <bryce> my memory really isn't this good normally | |
| [09:34] <Ng> hehe, well to be fair i did buy pretty much everyone pretty much all the time about it ;) | |
| [09:34] <bryce> I'm 99.999% certain the fix was taken upstream | |
| [09:34] <bryce> however, what upstream has done with it since then is anyone's guess | |
| [09:35] <bryce> I know they've done additional work on this area since then | |
| [09:35] <bryce> Ng: file another bug with the current state of the issue, ref the prev bug, and I'll follow it up | |
| [09:35] <Ng> ok, ta :) | |
| [09:36] <Ng> fwiw, http://mairukipa.tenshu.net/855gm-resume-broken.png is what I'm seeing | |
| [09:36] <bryce> don't understand... | |
| [09:37] <bryce> I see what looks like some image doubling or something? | |
| [09:37] <Ng> see how the network-manager icon shows all the trails from the spinning, as well as the latest state (LAN connection) | |
| [09:37] <bryce> mm, sort of... | |
| [09:37] <Ng> if I cover that icon it'll force it to redraw, but if I then connect to another network the trails will build up again until I next force it to completely redraw | |
| [09:37] <bryce> hmm | |
| [09:38] <Ng> so scrolling in firefox causes a bit of a mess ;) | |
| [09:38] <bryce> make sure to describe the situation really well (or include screencast) | |
| [09:38] <Ng> oki doki | |
| [09:38] <Ng> I'm going to have a quick trawl of bugs to see if I can find anything similar | |
| [09:38] <bryce> I'm sure we'll here, "Oh yeah, that's 100% fixed. Just switch to TTM." | |
| [09:39] <Ng> heh | |
| [09:39] <Ng> these interregnum periods are a pita for distros ;) | |
| [09:52] <Ng> filed as bug #182791 | |
| [09:52] <ubotu> Launchpad bug 182791 in xserver-xorg-video-intel "855GM resume problems" [Undecided,New] https://launchpad.net/bugs/182791 | |
| [09:56] <bryce> thanks | |
| [09:57] <bryce> Ng, I've been lightly hammering on Intel to get the <= 855 issues sorted out for a while | |
| [09:57] <Ng> good :) | |
| [09:57] <Ng> there are a lot of such laptops still in use and when they work they work fantastically ime, but it is a bit of a shame to have to chase down weird bugs each release ;) | |
| [09:58] <bryce> I've got it on record that they want -i810 deprecated, and thus desire all issues that work in -i810 but not -intel to be resolved in -intel so we can deprecate -i810 | |
| [09:58] <bryce> sort of circular there, but in a good way | |
| [09:58] <Ng> has i810 changed since gutsy? istr it hadn't changed there since feisty? | |
| [09:59] <bryce> no it hasn't | |
| [09:59] <Ng> mmkay, well I probably won't bother even testing that yet, it was never very good at resume anyway ;) | |
| [09:59] <bryce> upstream in fact has just symlinked -i810 to -intel | |
| [09:59] <Ng> I am going to quickly try EXA just in case, since that is at least the default the driver is choosing | |
| [09:59] <bryce> cool | |
| [10:01] <ubotu> New bug: #182791 in xserver-xorg-video-intel (main) "855GM resume problems" [Undecided,New] https://launchpad.net/bugs/182791 | |
| [10:03] <Ng> heh, good work ubotu ;) | |
| [10:10] <Ng> wow, it's even worse with EXA | |
| [10:10] <Ng> it's almost impossible to get anything to render. at best I could see the background of gtk widgets, but absolutely no foregrounds or text | |
| [10:24] <Ng> heh, I found an (I think) unrelated 855gm bug on fd.o's bugzilla and the intel guys seem to be saying that they don't have such hardware to test on | |
| [10:24] <Ng> this is not encouraging ;) | |
| [10:30] <ubotu> New bug: #182802 in xserver-xorg-video-intel (main) "[hardy] Screen freeze on resume" [Undecided,New] https://launchpad.net/bugs/182802 | |
| [10:46] <ubotu> New bug: #153122 in xserver-xorg-video-openchrome (main) "Fatal error while loading Xorg" [Undecided,Incomplete] https://launchpad.net/bugs/153122 | |
| [10:46] <ubotu> New bug: #164342 in xserver-xorg-video-openchrome "Suspend broken" [Undecided,Incomplete] https://launchpad.net/bugs/164342 | |
| [10:53] <ubotu> New bug: #178313 in xserver-xorg-video-via (main) "antspotlight screensaver hangs computer" [Undecided,Incomplete] https://launchpad.net/bugs/178313 | |
| [13:27] <mvo> tjaalton: you mentioned in #179515 that the screensaver might kill X during a upgrade, do you know any details? | |
| [13:28] <mvo> tjaalton: I'm wondering what to do about it, sending a inhibit signal via dbus for gnome-screensaver would be a obvious (but not trivial as the release-upgrader does not have access to the session bus) solution, but that leaves xscreensaver and kde out | |
| [13:30] <tjaalton> mvo: yeah :/ | |
| [13:30] <tjaalton> mvo: no perfect solution there, but I don't know what else could have caused it | |
| [13:35] <mvo> tjaalton: ok, its at least a hint. I will try to reproduce it here with screensaver on and see what comes out of it | |
| [13:38] <tjaalton> mvo: it only happens if your driver is buggy with DRI | |
| [13:38] <mvo> tjaalton: so for every driver ;) ? | |
| [13:38] <tjaalton> hehe | |
| [13:38] <tjaalton> there's always the possibility, sure :) | |
| [13:38] <mvo> tjaalton: do you know more details? | |
| [13:38] <mvo> I mean, what exactly is causing it? the fact that libdrm/libGL changes while the screensaver is runing? | |
| [13:39] <mvo> some incompatibility between the drm/dri interfaces? | |
| [13:39] <tjaalton> well, GL screensavers tend to trigger bugs | |
| [13:40] <tjaalton> but if the stuff is already loaded in the memory it doesn't matter if the libs are updated | |
| [13:40] <mvo> meh, if its in dri, then I can not reproduce it in a VM :( | |
| [13:40] <tjaalton> no, it's just a scenario | |
| [13:40] <tjaalton> that X crashes when you upgrading | |
| [13:41] <tjaalton> and to minimize the possibility it would be wise to disable the screensaver | |
| [13:41] <mvo> its probably a good theory, because I have not seen it crash in a VM yet, but it seems to happen sometimes on real HW - I have seen some reports | |
| [13:41] <tjaalton> yeah, I bet there are dupes | |
| [13:43] <tjaalton> you can disable gnome-screensaver with gconf | |
| [13:43] <tjaalton> idle_activation_enabled | |
| [13:45] <mvo> I look into it, I need to come up with something that works for xubuntu and kubuntu as well | |
| [13:45] <tjaalton> ok, cool | |
| [15:21] <ubotu> New bug: #182865 in xserver-xorg-video-intel (main) "Cannot switch to console after installing "xserver-xorg-video-intel 2:2.1.1-0ubuntu9.1"" [Undecided,New] https://launchpad.net/bugs/182865 | |
| [15:46] <ubotu> New bug: #182878 in xorg (main) "[gutsy] gma 950 no TV-OUT in dispalyconfig" [Undecided,New] https://launchpad.net/bugs/182878 | |
| [15:46] <ubotu> New bug: #182883 in xserver-xorg-video-intel (main) "dual screen support doesn't work" [Undecided,New] https://launchpad.net/bugs/182883 | |
| [16:41] <ubotu> New bug: #182898 in xserver-xorg-video-intel (main) "erratic widescreen intel gma 3000" [Undecided,New] https://launchpad.net/bugs/182898 | |
| [17:11] <Ng> not a good day for -intel and bugs ;o | |
| [17:16] <tjaalton> poor intel | |
| [19:48] <mvo_> bryce I may do a mesa upload within this week, is there anything planed? I need to patch it so that compiz can detect glx window | |
| [19:55] <bryce> mvo, it should be safe. I don't have mesa uploads planned and don't think timo does either | |
| [19:56] <bryce> I'm going to do an xserver upload for Q-FUNK with some new patches, but that'll be for gutsy-proposed, not hardy | |
| [19:56] <bryce> tjaalton might have some other things planned | |
| [19:57] <mvo_> bryce ok, thanks. I'm currently in testing stage, but it looks promising so far and I don't want to interfere with the X team, especially since you use git now (not sure if you also use it for mesa) | |
| [20:07] <bryce> no we don't | |
| [20:09] <mvo_> good, thanks for this info | |
| [20:12] <Q-FUNK> bryce: aye, thanks for that | |
| [20:12] <Q-FUNK> let's see if it improves everyone's results | |
| [20:12] <tjaalton> mvo_: please go ahead | |
| [20:12] <Q-FUNK> if it does, we can always come back to the Hardy patch before the release takes place | |
| [20:18] <bryce> Q-FUNK: I need to do some UME stuff (trying to get it booted on the test hw they've provided) | |
| [20:19] <Q-FUNK> ah ok | |
| [21:04] <Ng> tjaalton: hmm, don't you have an X40? | |
| [21:05] <tjaalton> Ng: nope, X61 | |
| [21:06] <Ng> ah, doh | |
| [21:06] * Q-FUNK is patiently waiting for his Dell D430 to be delivered. | |
| [21:06] <Ng> I noticed that my particular model doesn't have any pm-utils quirks, so figured that might be causing X to fail to resume, but I've tried what I think should be the matching options from the old acpi-support, which made absolutely no difference ;( | |
| [21:09] <tjaalton> you might want to ask mjg59 for some tips | |
| [21:09] <Ng> yeah | |
| [21:09] <tjaalton> or check thinkwiki | |
| [21:10] <tjaalton> bryce: mjg59 confirmed that SHMConfig is not an option for us (pun intended) | |
| [21:10] <tjaalton> so I closed the bug that asked for including it | |
| [21:11] <tjaalton> ps. comedy inc. rules :) | |
| [21:13] <Ng> tjaalton: it was a sad day when mjg59's x40 got ruined and he switched to an HP ;) | |
| [21:15] <tjaalton> aww | |
| [21:16] <tjaalton> that explains the recent blog post then ;) | |
| [21:18] <bryce> tjaalton: ah ok | |
| [21:35] <Q-FUNK> ah. vorlon pushed the new -amd into hardy | |
| [21:37] <bryce> ok, I'm done (stuck) on the ume stuff for the time being. -amd time. | |
| [21:39] <jcristau> ` | |
| [21:40] <Q-FUNK> heh | |
| [21:40] * jcristau blames ssh | |
| [21:41] <jcristau> fwiw, #182252 is invalid. X is started with '-nolisten tcp' by default, so obviously inet connections don't work | |
| [21:50] <bryce> Q-FUNK, what are the bug ID('s) that are fixed with these two xserver patches? | |
| [21:50] <tjaalton> jcristau: good point, I'll close that one | |
| [21:54] <Q-FUNK> bryce: I'm not sure that there was a separate bug open for this freeze issue on some BIOS | |
| [21:55] <bryce> can you open one please? In order to upload SRU candidates they require patches be tied to bug ID's in launchpad | |
| [21:57] <bryce> are these patches fixes for bug 140051? | |
| [21:57] <ubotu> Launchpad bug 140051 in xserver-xorg-video-amd "amd driver fails to autoconfigure" [High,Triaged] https://launchpad.net/bugs/140051 | |
| [21:58] <Q-FUNK> yes and no | |
| [21:58] <bryce> hmm, maybe I'm asking the wrong question | |
| [21:58] <Q-FUNK> it prevents X from freezing at configure time, but doesn't directly resolve DDC failing with -amd yet | |
| [21:59] <bryce> hmm, maybe I don't understand this | |
| [22:00] <bryce> it sounds like these don't fix the issue, as there is another known issue that prevents it from working | |
| [22:01] <bryce> however it adds the risk of changing existing -amd user's systems (presuming there are -amd users able to use the current driver) | |
| [22:03] <bryce> wouldn't it be more appropriate to put this into hardy or a ppa if the purpose is simply to check if it breaks other -amd user's systems? | |
| [22:04] <Q-FUNK> it needs to go into hardy regardless, if it works, but testing via gutsy-proposed feels as the most harmless way of having a wide test range | |
| [22:04] <bryce> hmm | |
| [22:04] <Q-FUNK> well, -amd is essentially borken for too many people, thanks to the broken x86emu | |
| [22:05] <Q-FUNK> I'm stuck with Etch and a souped-up Feisty to they my thincans here, because of that | |
| [22:05] <Q-FUNK> they->test | |
| [22:06] <bryce> there could also be some risk for non -amd users, since it affects code that (I gather) is run by all xserver users, so want to be careful we're not pushing it too aggressively | |
| [22:09] <bryce> not that I don't trust that this is a good fix to have, but it feels like we're skipping over some intermediary testing steps | |
| [22:09] <bryce> hmm | |
| [22:09] <bryce> ok I'm building the xserver with the patches. Let me go grab lunch and think of how we should do this. | |
| [22:11] <ubotu> New bug: #134951 in ubuntu "gutsy too high resolution (dup-of: 182778)" [Undecided,New] https://launchpad.net/bugs/134951 | |
| [22:18] <Q-FUNK> bryce: I just made a call to the X list for people to test this as much as possible | |
| [22:24] <bryce> back | |
| [22:27] <bryce> the bus patch failed to apply | |
| [22:28] <bryce> Applying patch 146_x86emu-correct-bus-dev-fn.patch | |
| [22:28] <bryce> patching file hw/xfree86/int10/helper_exec.c | |
| [22:28] <bryce> Hunk #1 succeeded at 469 (offset -80 lines). | |
| [22:28] <bryce> Hunk #2 FAILED at 480. | |
| [22:28] <bryce> Hunk #3 FAILED at 494. | |
| [22:28] <bryce> patch: **** malformed patch at line 61: PCI_OFFSET(PciCfg1Addr) + offset); | |
| [22:28] <bryce> P | |
| [22:30] <bryce> checking other patch... | |
| [22:30] <bryce> Q-FUNK: I did think about this some. What I'd like to do is before uploading to gutsy-proposed, to put up a deb and get a few people to test it on gutsy systems, to make sure it doesn't cause X to fail to start | |
| [22:31] <bryce> Q-FUNK: if we can get three thumbs up, I say let's stick it in gutsy-proposed | |
| [22:31] <bryce> oops, the other patch also failed to apply | |
| [22:33] <Q-FUNK> :( | |
| [22:33] <Q-FUNK> let's ask bartman | |
| [22:34] <bryce> hang on, I am looking into it deeper | |
| [22:34] <bryce> I can usually fix up patches to apply, if they're not too intricate | |
| [22:43] <bryce> Q-FUNK: hmm, those patches appear to be against current xserver head. Let me check them against xserver 1.4 | |
| [22:49] <bryce> nope | |
| [22:54] <Q-FUNK> bryce: I just told Bart and Gadi about it | |
| [22:54] <jcristau> i'm not sure what patches these are, but if they're for a post-pciaccess server, then they might need porting | |
| [22:54] <bryce> ok, I'm trying to do the patches manually | |
| [22:55] <bryce> jcristau: ah that might be; yes they appear to deal with pci tags | |
| [22:55] <bryce> hmm, that might be more work than I have time for | |
| [22:56] <Q-FUNK> indeed | |
| [22:56] <Q-FUNK> never mind | |
| [22:56] <Q-FUNK> I'll ask bartman to rebase against 1.3 for the test | |
| [23:07] <ubotu> New bug: #182743 in xorg (main) "hardy xorg not detecting for compaq evo n410c" [Undecided,Confirmed] https://launchpad.net/bugs/182743 | |
| [23:09] <bryce> Q-FUNK: ok cool | |