[00:03] Hello. I have some questions about the Canonical partner repository (just that repository). I'm wondering what the new package acceptance process is, and what the acceptance criteria are. So far I have not found any web pages about this. === dendro-afk is now known as dendrobates [00:35] kees: please check with chrisccoulson .... if it involves patches we need to talk to upstream as usual. if there is something serious let me know [01:42] there. gtk+2.0 now building on one of the new-n-good armel buildds [01:45] \o/ === ctrlsoft_ is now known as ctrlsoft === ctrlsoft is now known as Guest44441 [01:47] pitti: what do you think of a nautilus work-around for bug 543617 ? (or in umount itself...) [01:47] Launchpad bug 543617 in linux "very slow filesystem I/O" [High,In progress] https://launchpad.net/bugs/543617 === Guest44441 is now known as ctrlsoft [02:03] lamont, any word on bug #533601 [02:03] Launchpad bug 533601 in bind9 "Apport hook for bind9" [Wishlist,Triaged] https://launchpad.net/bugs/533601 === vorian is now known as v === v is now known as vorian [03:05] Hi ubuntu devel [03:05] what is the name of package of text mode ubuntu installer? like ubiquity but in text mode [03:16] seyacat: debian-installer, I would assume. [03:38] pitti: cups> debian 572940 is about CVE-2010-0302, not CVE-2010-0393. [03:38] Debian bug 572940 in cups "CVE-2010-0302: Incomplete security fix" [Important,Fixed] http://bugs.debian.org/572940 [03:38] Use-after-free vulnerability in the abstract file-descriptor handling interface in the cupsdDoSelect function in scheduler/select.c in the scheduler in cupsd in CUPS 1.3.7, 1.3.9, 1.3.10, and 1.4.1, when kqueue or epoll is used, allows remote attackers to cause a denial of service (daemon crash or hang) via a client disconnection during listing of a large number of print jobs, related to improperly maintaining a reference count. NOTE: some of thes [03:38] The _cupsGetlang function, as used by lppasswd.c in lppasswd in CUPS 1.2.2, 1.3.7, 1.3.9, and 1.4.1, relies on an environment variable to determine the file that provides localized message strings, which allows local users to gain privileges via a file that contains crafted localization data with format string specifiers. (http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2010-0393) [05:22] Hi! the other day, I updated Lucid, and a udev file in which I loaded an experimental driver for multitouch trackpads, changed and I am not able to load that driver again [05:24] the file is "/lib/udev/rules.d/66-xorg-synaptics.rules", the line I used was: ENV{x11_driver}="multitouch" === JanC_ is now known as JanC [05:28] NinoScript: Help for Lucid is in #ubuntu+1 [05:31] ScottK: I asked there yesterday and the day before, but nobody knew (and I'm asking again, now)… as it has more to do with the actual development of the driver, I thought I could ask here :P [05:32] You might ask in #ubuntu-x [05:34] Ok, that's useful, thanks! :D [05:52] * kirkland just gives up and assigns all of his bugs to pitti :-) [06:01] I maintain a Debian package, grace, that fixed an annoying UI bug (can trap the cursor, have to kill from console; #567323, LP 552681) in the most recent Debian upload. Any way I can nudge that past the freeze? [06:01] Launchpad bug 552681 in grace "Mouse cursor trapped with right-click in menu, lose input control" [Undecided,New] https://launchpad.net/bugs/552681 [06:01] (Apologies if I overlooked some documentation; my ubuntu-fu is weak, I only found the freeze exception instructions for new *upstream* versions, which this is not.) [06:04] nbreen: That's because bug fix updates don't need a freeze exception. You can request a sync using the requestcync script in ubuntu-dev-tools (which is in Debian now). [06:05] Will do, thanks for the pointer. [08:45] what are the -dbg for? [08:49] how to use the xserver-xorg-video-ati-dbg package to check the internals of the driver [10:34] when will 10.04.1 be released normally? [10:35] ScottK: oh, I wasn't aware that you are blocking on me for the sync script; please add an appropriate license header, I'm really not legally attached to it at all :) [10:36] kirkland: give up? when you are always constantly beating me? :-) [11:47] tseliot, fyi, jokey failed on a fresh install of lucid while installing nvidia-current: http://paste.ubuntu.com/408038/ i had to do it manually [11:47] * tseliot has a look [11:48] fta: well, it's more correct to say that jockey interpreted a warning as an error and claimed to have failed [11:49] tseliot, yep, but after a restart of X, it failed, the kernel module wasn't there [11:50] fta: are you sure about that? the log says that the module was built and installed. Maybe it didn't set the correct alternative. Either way a reboot is required [11:50] i had to --reinstall nvidia-current and reboot (as restart kept on using nouveau, probably due to DKMS [11:52] fta: the problem is that nouveau was preventing nvidia from being loaded [11:53] updating the initramfs would have been enough [11:54] tseliot, maybe. it's not a problem for me, i quickly managed to sort this out, but i assume lots of people will face this [11:54] pitti: logging.error(err) in oslib.py (line 248) shouldn't cause jockey to fail, right? Is there anything else that could be causing this? [11:56] fta: I think we have a bug report about it and it's definitely something I'll have to fix [11:56] ok, good [11:59] fta: I think it's bug #552653 [11:59] Launchpad bug 552653 in jockey "[Lucid] Jockey fails to install nvidia hardware driver" [Medium,Triaged] https://launchpad.net/bugs/552653 [12:00] tseliot, yep, that's it [12:00] ok [12:00] tseliot, btw, why isn't libvdpau1 a dep of nvidia-current? [12:02] fta: programs that use vdpau should have it as a dependency. It's not really specific to nvidia [12:02] i thought vdpau was nvidia specific.. [12:03] no [12:03] other implementations exist now [12:04] the nvidia specific bits are in nvidia-current [12:04] ok :) === dendrobates is now known as dendro-afk [14:04] hi [14:04] while ago I started to wonder. [14:04] why does ubuntu have duplicate keyboard layouts? [14:04] ones for xorg and others for terminal [14:05] it doesn't [14:05] the terminal layouts are automatically generated from the X layouts, and have been for over three years [14:06] any differences are simply because the Linux console's support for some things isn't quite as rich as X's [14:07] so you already generate terminal layouts from X layouts. [14:07] what prevents from using X layouts in terminal directly? [14:07] different file formats [14:07] why bother doing all the work to make the kernel support XKB, when we can just as well convert things on the fly in userspace? [14:08] some things are better done outside the kernel [14:08] and, honestly, what difference does it make? [14:09] some apps ignore the X layouts. [14:09] and switching layout from X doesn't switch the layout kernel is using [14:09] um. what do applications have to do with it? keyboard layouts are at a lower level [14:10] 'dpkg-reconfigure console-setup' changes both of them system-wide [14:10] Would a buildd admin please rescore qscintilla2 and kde4libs. The fact that they are lagging on ports is going to cause a number of other problems. [14:10] it's intentional that it's possible to have a different per-user layout - it's entirely possible on multi-user systems that different users might prefer different layouts [14:11] aah. yet per-user layouts. [14:11] ScottK: done (insofar as it'll make much difference on ia64 :-/) [14:12] cjwatson: thanks. [14:13] cjwatson: though X is ancient, how come it has 'better' keyboard layout stuff than kernel? [14:13] XKB is not all that ancient; it's of approximately the same vintage as the kernel's keyboard layout handling, possibly even a little later [14:14] likely older than unicode as well, right? [14:14] and one might expect XKB to support a wider variety of layouts, given that there are some complex scripts the kernel can't cope with displaying on the console [14:14] Cheery: incorrect [14:15] both XKB and the Linux kernel's keyboard layout support date from the early-to-mid-1990s, after Unicode started [14:15] first version of the Unicode standard was 1991, predating both of them [14:15] but I'm going out so I'm afraid you'll have to do the rest of the research yourself :-) [14:17] cjwatson: thank you anyway. it was fun and interesting topic. === barry` is now known as barry_ === barry_ is now known as barry === primes2h1 is now known as primes2h === sconklin-gone is now known as sconklin [14:35] tseliot: Hello, I reported a bug about plymouth bug #553954 [14:35] Launchpad bug 553954 in plymouth "[Lucid Beta 1] plymouth has a non translatable string coming up on screen" [Undecided,New] https://launchpad.net/bugs/553954 [14:35] tseliot: you wrote that script, didn't you? :-) [14:36] Is bug description correct? [14:37] * tseliot has a look [14:39] primes2h: yes, it's correct. Plymouth isn't localised. Is mountall localised? [14:40] tseliot: it has just been fixed bug 390740 [14:40] Launchpad bug 390740 in mountall ""Routine check of drives"message is not localized" [Low,Fix released] https://launchpad.net/bugs/390740 [14:46] tseliot: what do you think about the bug? [14:47] primes2h: each plymouth theme e.g. the one for ubuntu, kubuntu, etc. will have to be modified so as to allow localisation [14:48] Sounds like step on is a bug on the relevant packages then [14:48] tseliot: it would be better to put that string in an outer package, wouldn't it? [14:48] like all others [14:52] primes2h: I'm not really sure about how we should do this. I'll think about it [14:54] tseliot: Ok, thank you. :-) unfortunately it's a annoying bug because the string is very visible (and quite big) [14:57] I know, I'll see what I can do [14:58] ScottK: what do you mean? are you talking about this bug? [15:00] bug #553954 [15:00] Launchpad bug 553954 in plymouth "[Lucid Beta 1] plymouth has a non translatable string coming up on screen" [Undecided,New] https://launchpad.net/bugs/553954 [15:01] Yes. [15:01] If each plymouth theme needs updatiing, then there ought to be a bug against each one === dendro-afk is now known as dendrobates [15:07] ScottK: Sure, or take the string out to another common package, like others. [15:08] Why is Ubuntu's plymouth theme seeded in standard? [15:24] please sponsor bug 421684 [15:24] Launchpad bug 421684 in obexd "[SRU] bluetooth send malformed files" [Undecided,Confirmed] https://launchpad.net/bugs/421684 [15:30] ari-tczew: k [15:41] ari-tczew: done [15:50] pitti: around? [15:55] https://bugs.launchpad.net/ubuntu/+source/chromium-browser/+bug/513745 [15:55] Ubuntu bug 513745 in chromium-browser "pdf documents (probably other mimetypes too) not opened by chromium-browser" [Undecided,Confirmed] [15:55] that's my problem. anyone? :) [15:57] damnit :/ [16:54] nice, "maverick", i dig it! [16:54] http://www.markshuttleworth.com/archives/336 [17:05] kirkland: although I have a hard time imagining a maverick meerkat [17:05] they seem almost borg-like on TV [17:06] warm and fuzzy borg, but still :-) === yofel_ is now known as yofel [17:26] Since my last dist-upgrade, plymouth now displays in text mode even though I have kms on my thinkpad T30 [17:26] how would I go about figuring out why? === deryck is now known as deryck[lunch] [17:29] mdeslaur: are you missing the plymouth-theme-ubuntu-logo package? [17:30] mdeslaur: themes were split out to a) discourage users from selecting themes that don't play well with mountall, b) let server team drop the graphical splash they don't want; ubuntu-standard currently Recommends: -theme-ubuntu-logo [17:31] slangasek: let me check, hold on a sec [17:31] slangasek: it's installed [17:32] oh, I think I'm getting a kernel oops [17:32] does plymouth fall back if there was en error? [17:32] oh, splendid [17:32] if it's an error it's able to catch, yes [17:32] ok, let me investigate a bit [17:35] I noticed I've got a fair amount of boot funkiness after the last kernel [17:40] yeah, I've got a kernel oops from an updated thinkpad bios warning...so I guess plymouth is falling back to display the error [18:04] slangasek: heh, I see the text mode on my T42 too. I don't have an 'WARNING's in my dmesg [18:04] s/heh/hey/ [18:05] slangasek: I have plymouth-theme-ubuntu-logo installed [18:05] jdstrand: the only time text mode is ever used is if you 1) boot without 'splash' (and then it's the 'details' plugin), 2) plymouth can't find a working drm or fb interface, 3) you don't have graphical themes installed/configured [18:09] slangasek: well, this is a fresh install from yesterday's daily. 1) [ 0.000000] Kernel command line: ... quiet splash. 2) radeon/kms/compiz is working now with 19.28, so I drm is ok (at least for X) '[drm] radeon kernel modesetting enabled' 3) plymouth-theme-ubuntu-logo, plymouth-theme-ubuntu-text are installed [18:09] slangasek: so, I have "splash"...my /proc/fb contains "0 radeondrmfb" [18:09] slangasek: I am unfamiliar with the 'details plugin' [18:09] slangasek: how do I know if the graphical themes are configured? [18:10] $ cat /proc/fb [18:10] 0 radeondrmfb [18:11] mdeslaur: does it feel to you like we are ganging up on slangasek? [18:11] hmm, maverick meerkat? [18:11] jdstrand: the details plugin is the plymouth plugin that shows all the console text, in addition to any prompts that it needs for interaction [18:11] mdeslaur: ls -l /etc/alternatives/default.plymouth [18:11] jdstrand: yes, but I'm guessing he's gotten used to it by now [18:12] jdstrand: what precisely do you see at boot time? [18:12] jdstrand: do you see the "Ubuntu 10.04" text + *s below? [18:12] /etc/alternatives/default.plymouth -> /lib/plymouth/themes/ubuntu-logo/ubuntu-logo.plymouth [18:12] slangasek: that points to /lib/plymouth/themes/ubuntu-logo/ubuntu-logo.plymouth for me as well [18:12] * slangasek nods [18:13] slangasek: yes, but they seem like throbbing dots more than '*'s [18:13] (they are teeny though, so maybe I am blind-- I didn't get much sleep last night ;) [18:13] close enough, we'll attribute the difference to my fallible memory :) [18:14] slangasek: does it help if when I see you name in irc all I see are stars? [18:14] jdstrand, mdeslaur: is cryptsetup installed on these systems? [18:14] heh [18:14] that *may* be unrelated [18:15] slangasek: not here, I am currently just using ecryptfs on this laptop [18:15] slangasek: no cryptsetup...I'm not using any of that security crack :) [18:16] slangasek: I *do* see other text at the bottom of the screen, and it's line breaks aren't right (familiar stair-stepping), but they fade to almost gone just before gdm comes up [18:16] s/it's/its/ [18:16] jdstrand, mdeslaur: please try running: echo FRAMEBUFFER=y | sudo tee /etc/initramfs-tools/conf.d/plymouth && sudo update-initramfs -u, and tell me if the behavior changes [18:16] I hate security. It's *always* in the way :P [18:17] hmm. what version of plymouth, and what version of plymouth-theme-ubuntu-logo, do you have installed? [18:17] slangasek: 0.8.1-3 for all [18:17] ok [18:18] * jdstrand reboots laptop [18:18] I've got 0.8.1-4 for all [18:18] slangasek: no effect. shall I upgrade and try again? [18:18] slangasek: no effect for me either, still text mode [18:19] don't see that an upgrade would help [18:19] is this with the -19 kernel? [18:19] 19.28, yes [18:19] $ cat /proc/version_signature [18:19] Ubuntu 2.6.32-19.28-generic 2.6.32.10+drm33.1 [18:19] * mdeslaur kicks plymouth in the face [18:19] do you see the same problem if you boot an earlier kernel? [18:19] I suspect radeondrm regressions [18:20] slangasek: is -18 early enough? (probably) [18:20] maybe? I don't know, I don't have radeon here to compare with [18:20] 19 had the drm radeon fixes for us [18:20] * jdstrand tries [18:20] this is where you go straight from grub to gdm without getting the boot splash? [18:21] LaserJock: no logo, but have some ascii [18:21] slangasek: with -17 kernel, I still have text mode plymouth [18:21] slangasek: -18 too [18:21] (unsurprisingly) [18:22] I got something like that with -intel but only recently [18:23] -16 gets text mode also... [18:23] slangasek: I know it did work for me a couple of weeks ago, but I can't be more specific-- like I said, I reinstalled yesterday [18:24] LaserJock: yeah, this is recent here too [18:25] mine flickers a few times like it's trying [18:25] but I end up with text stuff -> gdm [18:26] slangasek: fyi, the installer is the same-- text after the first (what I assume to be non-plymouth) bit === deryck[lunch] is now known as deryck [18:34] LaserJock: the question is, what text stuff? The bug they're having is that they see 'Ubuntu 10.04' in text when they should see the graphical splash; there are also known cases where the splash just never gets around to getting started before X starts [18:34] slangasek: ah, ok, that's what I was trying to figure out. Mine is the latter [18:35] LaserJock: known bug, won't be fixed for lucid because the fix available in that timeframe is at odds with boot performance goals [18:35] I see [18:36] one of those "if we boot fast enough people won't notice" :-) [18:36] LaserJock: you're too fast for plymouth dude [18:36] jcastro: the problem is booting keeps getting slower :( [18:37] earlier in Lucid it wasn't noticable much, but now it's taking a bit more time and I start to wonder if something happened [18:37] for maverick, we'll tell *grub* to put up the splash, and if plymouth doesn't start before X, no big loss [18:38] Maverick is going to take me a while. I keep seeing Tom Cruise flying jets. [18:38] hahaha [18:40] ScottK: I keep thinking John McCain [18:40] at least it wasn't Rogue ;-) [18:40] with a silent M? [18:41] I'm also old enough to remember the TV show with James Garner. [18:41] ScottK: oh heck yeah [18:41] one of the best ever [18:41] james_w: oh, is that how you pronounce 'mbrola'? [18:41] Yep. === radoe_ is now known as radoe [18:50] ScottK: because of you, I now have "Danger Zone" stuck in my head [18:50] \o/ [18:50] mdeslaur: thanks :/ [18:51] LaserJock: welcome :) [18:52] mdeslaur: Is that why you fly the way you do? Trying to prove something? [18:52] ICEMAN, CHECK SIX! [18:52] jcastro: What's your problem, Kazanski? [18:53] * ScottK notes: productivity destruction goal for today met. [18:53] heh, it's friday. [18:53] mdeslaur: http://bit.ly/iQYS [18:54] lol [18:54] * mneptok invents a new Ubuntu Community rickroll [19:09] mdeslaur, slangasek: is there a bug on the radeon/text boot issue? [19:09] jdstrand: not yet (from me) [19:09] I don't see a reference in backscroll... [19:09] jdstrand: I was attempting to downgrade plymouth to see where it started [19:09] jdstrand: not a good idea, btw :) [19:10] mdeslaur: heh, I was just going to say 'good idea' :P [19:10] jdstrand: newer mountall dies with older plymouth, so be warned :) [19:11] mdeslaur: I'm not looking at it atm [19:11] mdeslaur: so go with God [19:11] :) [19:11] of course, I'll complain if it isn't fixed, so don't worry about my feedback ;) [19:12] jdstrand: would you mind opening the bug? my laptop is unbootable to do the ubuntu-bug [19:12] mdeslaur: sure [19:30] mdeslaur, slangasek: fyi file bug #554143 [19:30] Launchpad bug 554143 in plymouth "text logo theme used instead of graphical with radeon 7500" [Undecided,New] https://launchpad.net/bugs/554143 [19:31] james_w, hey got a minute for jcastro and me? [19:31] jdstrand: thanks, I just confirmed it [19:31] s/file/filed/ [19:31] thanks [19:36] jdstrand: I had a similar issue with the text theme showing always on a native res inteldrmfb in the past week and removing all of the extra themes and just reinstalling those two fixed it [19:37] Sarvatt: which ones are "those two"? [19:37] Sarvatt: plymouth-theme-ubuntu-logo and plymouth-theme-ubuntu-text? [19:37] Sarvatt: which extra themes? I've got plymouth-theme-ubuntu-logo and plymouth-theme-ubuntu-text. are there others I should be looking for, or are you suggesting I remove plymouth-theme-ubuntu-text? [19:37] yep [19:38] Sarvatt: nm, I see the others, but the are already not installed [19:39] in response to mdeslaur that is, I had solar and spinfinity installed as well and all I did was remove all of the themes and just install those two again and it was fixed afterwards [19:39] I could try to reinstall... [19:40] it only happened when I packed plymouth in the initrd though so it might not be related :D [19:41] yeah, I already tried to update the initramfs [19:41] I'm getting error messages in lucid from /etc/acpi/sleep.sh, missing files in /var/lib/acpi-support. [19:41] In karmic those files were part of the acpi-support package, but they aren't in lucid any more (e.g. /var/lib/acpi-support/system-manufacturer) [19:42] I'm having trouble figuring out what script is looking for those files. [19:42] Sarvatt: didn't work [19:44] do you see the text renderer over vgacon for a split second before the screen changes resolution and then it's offset in the drmfb as well? [19:48] I see the underscore flash a few times before the text mode changes. there are 3 distinct flickers, then I see the underscore flash again (still unchanged), then text mode plymouth kicks in [19:51] Hmm, is it too late to be trying to get a new debmirror into Lucid? [19:51] I note that what's currently there predates all of the new work done to it [19:56] Caesar: you might want to talk to kees since he has been doing the bulk of the work for ubuntu (and I believe upstreamed almost all of it to Debian) [19:56] jdstrand: thanks, kees, you around? [19:57] nigelbabu: of course [19:58] jcastro, looks likes james is around after all :D [19:58] james_w: we want to start scheduling patch days, where we collect people to look at pending patches in lp. [19:59] james_w: when in the cycle do people generally start doing merges? [19:59] james_w: we'd like to time it around that time [19:59] right now, the date I've come up with in wed, may 5th [19:59] (the day before toolchain release) [19:59] jcastro: pretty much as soon as lucid releases [20:00] it ramps up from when the archive opens [20:00] archive opens on 6th right? [20:00] haven't looked [20:01] sounds about right [20:01] it's never an exact date, but it's pretty quick [20:01] the next thu after release day ... [20:01] it's not that rigid [20:01] it's whenever we can get enough of the stuff on NewReleaseCycleProcess done [20:02] don't believe the release schedule pages on this, they're just a guideline/target this early [20:02] cjwatson: yeah we just want to do it around the general time [20:02] jcastro, but we dont want the archive open if you ask me [20:02] right, I'm just trying to help nigelbabu understand that it isn't a fixed thing [20:02] ok [20:02] cjwatson, I understand... now :) [20:03] anyway, our first priority would be to send those patches upstream so they can integrate in [20:03] and we can either just backport from the git or just get the new release some time later in the release cycle [20:06] jcastro, so 5th sounds good enough or you want to move it? [20:06] I was hoping to have something done before lucid, so we can brainstorm on either we're we need to change exiting workflow during lucid [20:07] what was the default version of kde in hardy? [20:07] Caesar: hi! yeah, it's behind. [20:08] Caesar: upstream didn't take any of the rsync work I did, unfortunately. [20:11] Caesar: was there a specific feature you needed from the more recent version? [20:12] Caesar: this what needs merging: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=455082 [20:12] Debian bug 455082 in debmirror "enable more efficient batching" [Normal,Open] [20:16] bryceh, might want to take a look at http://pastebin.ubuntu.com/408255/ [20:17] kees: my concerns are that there was a lot of upstream work done last year, and none of that is going to be in Lucid [20:17] In particular what I'm interested in is the ability to mirror the installer [20:18] Caesar: I welcome a merge :) [20:19] tjaalton: Yay for the backported InputClass stuff. :-) [20:19] kees: I just wonder what's better, an already 3 year old debmirror, or a newer one sans your batching changes [20:19] My xorg.conf with Wacom calibration. http://gist.github.com/353570 [20:19] It sounds like merging is going to be difficult (but I haven't looked yet) [20:20] Caesar: the main issue is that the batching merge isn't easy. the rest of the fixes I already broke out (though upstream seems to be ignoring) in that above debian bug [20:20] Caesar: I haven't had time to get the rsync batching done, but I'd love it if someone could help. [20:20] nigelbabu, ok what am I seeing? [20:21] Caesar: the basic issue is: don't fetch one file at a time with rsync. :) [20:21] bryceh, list of packages with patches attached [20:21] the number is the total of open bugs with patches per package [20:21] kees: I'll see what I can do [20:21] nigelbabu, ok, interesting [20:21] bryceh, jcastro just got the good folks in #launchpad generate it [20:22] Caesar: if I didn't have a stack of other work, I'd commit to spending time on it, but that seems unlikely at this point. :P [20:22] nigelbabu, fwiw the list of patches I'm working through is at http://www2.bryceharrington.org:8080/X/Reports/ubuntu-x-swat/patches.html [20:22] kees: Yeah yeah, join the club :-) [20:22] nigelbabu, it's nicely been growing shorter and shorter :-) [20:23] bryceh, oh great. thats one team on top of their patches [20:23] Caesar: heh :) [20:23] bryceh, so we'll just skip the X bugs in review queue === kklimonda is now known as kklimonda|G1 [20:24] nigelbabu, strangely, it shows more patches than https://edge.launchpad.net/~ubuntu-x-swat/+patches does, but I think deryck has been looking into that [20:24] nigelbabu, yeah the reason xserver-xorg-video-openchrome is at the top of the list is because upstream posts patches to our launchpad queue for folks to test [20:24] bryceh, thats nice :) [20:24] bryceh, I like https://edge.launchpad.net/~ubuntu-reviewers/+patches [20:25] I found that it's not the case that they want those patches *in* ubuntu, they just want users to test them, then they can decide to include them upstream and we get them next time we sync [20:25] they're using as lab rats? :D [20:25] so a totally different patch workflow than everything else. But it works for them I gather [20:25] that's pretty handy though [20:25] nigelbabu, that does look nice [20:26] its the new code I guess since I see that only in edge [20:28] jdstrand, mdeslaur: may be useful to boot with 'plymouth:debug=file:/var/log/plymouth-debug.log' and grab that log file (if plymouth succeeds in outputting it) [20:36] slangasek: thanks, will try once laptop is reinstalled [20:37] * jdstrand will try now [20:37] jdstrand: argh, you're always one step ahead :) [20:38] mdeslaur: heh, will you stepped 'ahead' of me by downgrading plymouth === bjf is now known as bjf-afk === bjf-afk is now known as bjf [20:40] * bjf is taking a break, my dad just walked in [20:40] slangasek: the text theme did not display with that invocation. is that expected? === bjf is now known as bjf-afk [20:41] this may be it: [20:41] ply_open_module:Could not load module "/lib/plymouth/renderers/x11.so": /lib/plymouth/renderers/x11.so: cannot open shared object file: No such file or directory [20:41] slangasek: ^ [20:42] jdstrand: unless you're setting 'export DISPLAY=:0.0' in your initramfs, that shouldn't matter :) [20:42] jdstrand: you could try installing plymouth-x11 though, just to see what changes [20:43] but - the text theme didn't display? what did you get instead, just text output? [20:43] slangasek: I got what looks like the output of the plymouth debug log. like it tee'd [20:43] slangasek: I have the debug log, shall I paste or attach to bug? [20:44] jdstrand: attach to bug plz [20:45] slangasek: fwiw, plymouth-x11 is already installed [20:45] jdstrand: oh, well, hum [20:46] jdstrand: do you have /usr on a separate partition? [20:46] slangasek: nope. / and /home [20:46] do you still have the FRAMEBUFFER=y line in /etc/initramfs-tools ? [20:47] $ cat /etc/initramfs-tools/conf.d/plymouth [20:47] FRAMEBUFFER=y [20:47] yeah; so plymouth is in the initramfs, and we don't ever copy x11.so over [20:47] 'cause if you've got an X server in your initramfs, yer doin it rong [20:49] slangasek: log added to bug [20:50] ta [20:51] mdeslaur: what's your radeon chip? [20:51] I'm inclined to believe this is chip-specific, given that we've got a number of positive radeon tests from beta1 [20:52] slangasek: same, but 16mb [20:52] slangasek: it used to work :( [20:54] yeah, it was workign when manoj gave it back to me, so that was packages from a couple weeks ago and a -17 with backported drm that he added (but mdeslaur demonstrated it is not kernel specific) [20:54] actually, the drm renderer for plymouth wont work with a 8bpp framebuffer anyway will it? [20:54] radeon uses a lower color one for cards with <=16MB ram [20:55] jdstrand, mdeslaur: [./plugin.c] query_device:Visual is FB_VISUAL_PSEUDOCOLOR; not using graphics [20:55] kernel done you wrong [20:56] slangasek: what's the difference between the drm.so renderer and the frame-buffer.so renderer? [20:58] the framebuffer uses up a big chunk of the video ram always and its alot bigger if its the normal 24bpp mode, the alternative is not having enough video ram to do just about anything on the desktop if you want a pretty splash just for the few seconds during boot [21:01] mdeslaur: one uses libdrm, the other uses /dev/fb0 [21:01] slangasek: so how come it's failing with drm.so, shouldn't that work? [21:01] mdeslaur: the drm renderer is only used when you have more than one video output [21:01] mdeslaur: plug in an external monitor and see if it works? :) [21:01] ah, ok [21:02] ^on nouveau and ati, intel doesnt have that limitation [21:02] right [21:03] hmm...I wonder what changed [21:04] i think it was the -16 kernel that brought in the change to where drmfb's for low memory radeon GPU's use a lower color mode [21:04] can't you bump the video ram in your bios? [21:05] (or is it not integrated) sorry don't know which gpu you have [21:05] radeon 7500 [21:07] okay, if it's because of low video memory, I guess we'll live with it [21:08] whats the pci id? [21:09] Sarvatt: 1002:4c57 [21:09] jdstrand: what's your pci id? [21:09] ah not an IGP then :( [21:09] nope [21:11] ah, jdstrand's has the same pci id, but he has 32mb [21:15] /* select 8 bpp console on RN50 or 16MB cards */ [21:15] if (ASIC_IS_RN50(rdev) || rdev->mc.real_vram_size <= (32*1024*1024)) [21:15] bpp_sel = 8; [21:15] is it me or is that comment wrong and its using 8bpp for 32mb too? :D [21:17] ah yep http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=47381156a8f0d793bacfa346cc4cc515399525f7 [21:21] yeah, the comment is wrong :) [21:21] so <=32 Mb [21:24] Thanks Sarvatt for the info, I've updated the bug [21:25] so, it's not really an issue [21:36] mdeslaur: it's not a release-critical issue, but this is clearly a suboptimal outcome [21:37] yes [21:37] slangasek: yes, it is, especially since I still remember when it was working :) [21:38] jdstrand: memories fade :) [21:38] slangasek: is there a workaround? do we expect it to get fixed? [21:39] honestly, I'm not sure why Keybuk implemented the VGA16fb support as a separate renderer, instead of adding generic support to the fb renderer for downstepping the bit depth [21:39] well, radeon 7500 users (of which there are many), will have a nice usplash in all previous releases and then drop to text [21:39] jdstrand: not likely to get fixed for release [21:39] :\ [21:39]