=== thumper-gym is now known as thumper === mimico___ is now known as mimico === Zachary_DuBois is now known as Zachary_DuBois|A === Zachary_DuBois|A is now known as Zachary_DuBois [02:35] I have 13.10... How can I clear the list of "recent servers" when going through the files manager? === Zachary_DuBois is now known as Zachary_DuBois|A [02:37] cyphermox, has there been any though given to implementing WPS in NetworkManager / the indicators? [02:42] robert_ancell: this is that push-to-autoconfig stuff? [02:44] desrt, yes [02:45] I started using it a few weeks ago and it's actually very handy, especially when using a phone [02:45] seems like it would be more useful for devices lacking a proper keyboard [02:45] never thought about applying it to a laptop [02:46] still seems like it would be nice, though [02:47] desrt, it's particularly handy in the case you go to someones house where they're technically illiterate and they can't remember their router password [02:47] ya.... i actually experience this a lot [02:48] "oh... let me get out this book i wrote it in... no.. it's on this other piece... oh no... that was with my old provider.... hmmm... it's around here somewhere" [02:48] desrt, btw, still having that dbus unown name issue. I've cleaned up everything around the issue but haven't found what's causing it [02:48] remind me? [02:48] it seems the call is timing out (takes exactly 25s) but no errors are printed which makes me think glib thinks it has succeeded [02:49] in one of the test cases in lightdm, g_bus_unown_name blocks for 25s on shutdown [02:49] you thought it might be due to multiprocess screw ups, so I made really really sure I wasn't doing anything bad there [02:50] ah right. i remember now. [02:50] desrt, are you still hacking at the moment? [02:50] ish... [02:50] tending to builds, etc. [02:50] trying to test some touch issues in gtk [02:50] feel like giving it a whirl locally? [02:50] do i need ubuntu? :) [02:51] desrt, no [02:51] you need bzr though [02:51] i think i can manage that :) [02:51] it's just get lp:lightdm, build then run ./test-unity-compositor-fallback from tests/ [02:51] it fails for me most of the time [02:52] you can cut down that test to make it simpler to see what lightdm is doing, but perhaps you'll see more looking from the glib end [02:58] so that didn't work very well [02:58] gimme a sec [02:58] your autogen doesn't bail out when it's missing stuff for some reason [02:59] k. here goes. [02:59] so the test failed... immediately [03:00] lightdm: symbol lookup error: lightdm: undefined symbol: g_hash_table_get_keys_as_array [03:00] neat... [03:01] are you mucking with LD_LIBRARY_PATH in your test setup? [03:01] yes... you are [03:01] this is not jhbuild-compatible :) [03:02] okay... now the test exits immediately with no error and status 0 [03:05] robert_ancell: i guess you missed that... [03:07] desrt, try running it a few times, or with DEBUG=1 in the env [03:08] the amazing never-failing testcase :) [03:08] I wish === robert_ancell_ is now known as robert_ancell [03:08] it's going in a tight while true; ... loop [03:08] no failures after what must have been 1000 runs [03:09] okay... maybe more like 100 :) [03:10] hah! failure! [03:10] but uh... not the one you were expecting === Zachary_DuBois|A is now known as Zachary_DuBois [03:10] UNITY-SYSTEM-COMPOSITOR EXIT CODE=1 [03:10] RUNNER DAEMON-TERMINATE SIGNAL=11 [03:10] took a few hundred runs to get that though [03:10] no, that's not the one I was expecting :/ [03:12] desrt, did you try with DEBUG=1? [03:13] yes [03:13] no change except that it took longer to run with all the output [03:14] i ran maybe 100 tries like that and then tried with it off [03:15] maybe it is specific to something in Ubuntu [03:15] desrt, what version of glib? [03:15] git master as of a couple hours ago [03:15] i don't think we've made substantial gdbus changes lately, though === m_conley_away is now known as m_conley [03:16] robert_ancell: it's in the TODO file in NM [03:17] cyphermox, yeah, that's what I found. I was wondering if there's been any more thought on it [03:17] I'd kind of like to implement that for 14.10 if no-one else would [03:17] not really that I know of [03:17] I have lots of plans for 14.10 re: network-manager and all === duflu_ is now known as duflu === Zachary_DuBois is now known as Zachary_DuBois|A [06:03] Good morning === RAOF_ is now known as RAOF [08:20] do you guys know what is the responsible of showing me the dialogs to enter my ssh passphrase? it's broken since yesterday [08:20] the buttons show over the line edit === soren_ is now known as soren [09:02] wayland-egl [09:05] seb128: it should be part of libegl1-mesa-dev [09:05] hey ho [09:08] good morning desktopers [09:08] Laney, hey, how are you? [09:08] mlankhorst, well, g-c-c fails to build, so something went wrong in your upload [09:10] oh right [09:10] must be some incomplete reverts [09:10] hm it's in libegl1-mesa-drivers [09:12] bonjour seb128, hey Laney [09:12] pitti, salut, ça va ? [09:12] alright thanks seb128, enjoyed some tasty pancakes yesterday ;-) [09:12] pitti, how as the Alps? [09:12] oh right [09:12] how are you seb128 & mlankhorst & pitti? [09:12] Laney, I enjoyed "beignets", we don't do pancakes here ;-) [09:12] I'm good thanks! [09:13] * Laney goodles that [09:13] seb128: ça va bien, merci ! nous avons eu un bon weekend longue dans les Alps [09:13] https://en.wikipedia.org/wiki/File:BeignetsPowderdSugarCDM.jpg [09:13] oh my god [09:13] I bet yours looked better than that [09:14] lol, they did [09:14] pitti, "les Alpes" in french ;-) [09:14] pitti, tu as fait du ski ? [09:14] seb128: ah, merci [09:15] seb128: oui, la première (et peut-être la derniére) fois à cet an [09:16] pitti, oh, pourquoi peut-être la dernière, l'hiver est pas fini et il y en a de nouveau un hiver à la fin de l'année ! [09:16] seb128: en effet, je voulais "ce hiver" :) [09:17] ah ;-) [09:17] seb128: mais à Augsburg nous n'avons pas un vrai hiver [09:17] nous non plus [09:17] 10°c de prévu aujourd'hui :/ [09:17] nous avons eu de la neige pendant une ou deux semaines ici [09:18] Laney, that looks like the beignets we get, http://wherewearinthecity.com/wp-content/uploads/2013/09/beignets.jpg [09:18] yummy! [09:18] pitti, nous pas de neige cette année :-( [09:19] pitti, do you guys do pancakes or beignets this week? [09:19] seb128: we have something called "Krapfen", like this: http://www.maintal-konfitueren.de/fileadmin/data/images/Downloadbereich/Ambiente/krapfen_frei_gross.jpg [09:19] has something changed recently wrt how the desktop initiates shutdown (ck -> logind)? We have a few reports of users being unable to shutdown/reboot from the menu (bugs 1288087, 1286986, 1286698). [09:19] Launchpad bug 1288087 in upstart (Ubuntu) "System menu for Restart & Shutdown no longer works after installing Office 2010 through Wine" [Undecided,New] https://launchpad.net/bugs/1288087 [09:19] Launchpad bug 1286986 in upstart (Ubuntu) "Dropping to commandline is the only way to shutdown or restart" [Undecided,Incomplete] https://launchpad.net/bugs/1286986 [09:20] Launchpad bug 1286698 in upstart (Ubuntu) "Logout, Restart & Shutdown does not work" [Undecided,New] https://launchpad.net/bugs/1286698 [09:20] seb128: not entirely unlike beignets [09:20] pitti, I see ;-) [09:20] we ate 2 of them yesterday for the carnival [09:20] nice [09:20] I don't like all the rest of carnival, except the Krapfen :) [09:20] jodh, hey, no change that I can think about [09:21] hehe [09:21] seb128: hmm, ok. maybe be a hardware issue for those folk, but the timing of those bugs suggests maybe something else might be involved. [09:22] jodh, the bugs 2-3 are from the same user [09:22] oh [09:22] they are all 3 from the same user [09:22] seb128: gah - hadn't noticed that! good catch :) [09:23] ftr, are we using ck or logind atm? [09:23] jodh, that one user has a local problem [09:23] logind [09:23] ok, I'll see if they can invoke PowerOff using dbus-send. [09:24] jodh, the "upstart-event-bridge.log" snipet in the bug summary has [09:24] process 1206: arguments to dbus_pending_call_unref() were incorrect, assertion "pending != NULL" failed in file ../../dbus/dbus-pending-call.c line 608. [09:24] jodh: we've been using logind since saucy; CK is gone [09:24] not sure if that's a sign something is wrong with the bus session [09:25] pitti: I've got /usr/sbin/console-kit-daemon running on my system still. [09:26] pitti: looks like it is a dep for kde still. [09:27] jodh, feel free to dup those 3 bugs and reassign to unity or gnome-session [09:28] seb128: thanks - will do. [09:35] hey did unity get reverted to a version before saucy or something? [09:35] seb128: sorry for the UNRELEASED thing -- thats was of ocurse the line I changed, when I changed the version to ~trusty1 for the PPA :/ [09:35] the super and alt key handling seems to be back to the crappy pre-saucy timeout-based version [09:35] so if i press super+w and release it really quickly, unity opens the dash instead [09:36] and if i hit alt+anything and release both keys really quickly, the HUD opens up [09:36] Sweetshark, no worry [09:37] seb128: do we need a MIR for it? or is "all binary packages from libreoffice-dictionaires were in main before from openoffice.org-dictionaires and are from the same codebase?" enough? [09:37] hyperair, no unity changes to those that I know, check with bregma or ChrisTownsend or Trevinho [09:38] Sweetshark, check with doko or mterry or didrocks, but I think it should be ok/considered as a rename of an existing source [09:38] well, i guess they all got pinged by that comment so i'll wait [09:39] seb128: k [09:39] hyperair, when did the issue start? [09:39] seb128: after i upgraded to trusty [09:40] seb128: so it's between the saucy version and the trusty version =\ [09:40] (and i only upgraded two days ago) [09:40] k [09:41] well, we had some changes in grabbing, compiz is now grabbing the g-s-d keybindings (same way gnome-shell is doing for GNOME) [09:41] i see [09:41] but that shouldn't impact on super/hud, though I'm unsure [09:41] i recall the super/alt key handling was pretty bad in raring or before that [09:41] basically ctrl+super keybindings required you to press ctrl first [09:42] same goes for alt+super keybindings [09:42] I never use super out of for the dash so I can't say [09:42] sigh [09:42] and I map the hud out of alt [09:42] heh i see [09:42] what do you map it to? [09:42] right-win key [09:43] ah blragh i don't have that key [09:43] you use the hud? [09:43] if not you can just unbind it :p [09:43] yeah i do [09:43] k [09:44] adn my laptop doesn't have many spare keys to bind things to [09:44] well, in any case we should fix the regression [09:44] but meanwhile... [09:44] hence why i'm using all sorts of ctrl+super keybindings [09:44] oh yeah I can see what you see [09:44] i think i filed a bug regarding ctrl+super [09:44] super+w, release really quickly → dash [09:44] yeah [09:45] sucketh [09:45] there are also various problems with alt/hud/screenshot etc [09:45] if you pres alt+printscreen fast enough it opens the hud and then takes a screenshot of it [09:45] ali1234, new in trusty? [09:45] it's been almost a year since i dug into unity's code, but at that time the super/alt handling was basically... listen for super, and watch (within a certain timeout) for recognized keys) [09:45] i dunno if they are new or not since i don't use unity and when i did i unbound alt and super anyway [09:46] we need more keys on keyboard :p [09:46] and then saucy came along and holy shit it was fixed [09:46] bug 1283530 [09:46] Launchpad bug 1283530 in unity (Ubuntu) "Using alt-printscreen when locally integrated menus is turned on creates undesirable results" [Undecided,Confirmed] https://launchpad.net/bugs/1283530 [09:46] and now it's back ¬_¬ [09:46] I wonder if it ever got fixes [09:46] * apw notes that he can only disable the HUD binding in settings, if it is buond to _any_ key then Alt responds to it, it is also impossible to rebind it to Alt_L or indeed any modifier [09:46] fixed [09:46] or if that's random behaviour and you just happened to get lucky with timings [09:46] wat, seriously? [09:47] how many things got hardcoded shittily in the unity codebase this time? [09:47] it was a nightmare of hardcoded behaviour the last time i looked [09:47] ? [09:47] apw, I can't confirm that, I just binded it to lctrl, then back to alt than to the win key [09:48] seb128, one of these boxes i updated not 20m ago, and it is definatly exhibiting this behaviour [09:49] didrocks had similar issues but not happening in a guest session [09:49] seb128, and i have two next to each otehr showing it, both upgrades from saucy and likely older [09:49] yeah, well, it's not that I don't believe you [09:49] it's just that I can't confirm/reproduce [09:49] so where are the "unity settings which it doesn't ever upgrade properly" so i can nuke them [09:49] why is this morning turning into "report my issues about unity", we don't even work on unity in this channel :p [09:50] people, please use launchpad [09:50] :P [09:50] whoops, sorry for starting it [09:50] or ping #ubuntu-unity when the unity7 are around [09:50] let's talk about how indicators don't start properly under panel instead :P [09:50] e.g bregma or ChrisTownsend or Trevinho might be able to help you, but I don't think any of them is around yet [09:50] or why my machine died overnight so I don't have webkit debs :( [09:51] Laney, 0_o [09:51] webkit 10 - 3 Laney [09:51] (or I lost the count but I think webkit is still winning this cycle :p) [09:52] seb128, cause filing bugs is for whimps [09:52] haha [09:52] yeah, never take webkit on [09:52] you'll lose [09:52] can't be as bad as building a kernel [09:52] i had to iterate with a PPA because i kept ENOSPC'ing while building it [09:53] seb128, did yo manage to put the key back to Alt-l btw as in a guest i could move it to anywhere like Alt_R ok but not back to Alt_L [09:53] apw, you don't file bugs and then your issues never get resolved :p (I'm still thinking I should file you about my latitude locking for 1 second every time I change the screen level through the keys) [09:53] things were so much nicer in precise when that was not happening [09:53] * seb128 shakes fist at the kernel [09:53] haha [09:54] might be a unity bug for all we know ¬_¬" [09:54] hyperair: https://launchpad.net/ubuntu/+source/linux/3.13.0-16.36/+build/5659613 vs https://launchpad.net/ubuntu/+source/webkitgtk/2.3.90-1ubuntu1/+build/5619705 [09:54] apw, I managed to put it back to alt_r but the ui calls that "mod2+alt_l" for some reason, it works though [09:54] seb128, what about Alt_L [09:54] that says "level3 shift" [09:55] Laney: holy shit [09:55] seb128, mine calls Alt_R something like level3_shift [09:55] it's extreme [09:55] also C++ [09:55] but yeah, that doesn't work (hud doesn't open) [09:55] but C++ is good! [09:55] Laney: https://launchpad.net/~hyperair/+archive/staging/+build/5365404 [09:55] close enough ;-) [09:56] heh [09:56] don't let Sweetshark hear this conversation [09:56] pinging helps, yeah. [09:56] O:-) [09:57] ^_^ [09:57] * Laney tries again... [09:58] in the meantime, lemme see why the ubuntu-sso-client autopkgtests are borked [09:59] Laney, if you want to look at aptdaemon as well, please do ;-) (tried to ping mvo about those without luck) [10:02] seb128: ok, I'll look after [10:02] Laney, thanks [10:02] might be hard though, those mvo tests tend to be complex ;-) [10:05] * Sweetshark hugs mterry (in absence) for apppoving the hunspell-fr and google-fonts MIRs. [10:08] seb128, just for you i have filed a bug and whined at #ubuntu-unity [10:08] Laney, update_output has [10:08] "Trying easy from laney: ubuntu-settings/14.04.4 gnome-session/3.9.90-0ubuntu11 [10:08] leading: ubuntu-settings,gnome-session [10:08] failed: ubuntu-settings" [10:08] apw, thanks [10:08] it's just an old hint [10:09] Laney, right, can you clear it off? ;-) [10:09] yes, but it is harmless [10:09] k, noted for next time [10:09] I didn't know that hints were generating output when they had no matching content ;-) === ogra_` is now known as ogra [10:10] yeah it's telling you that it couldn't be applied [10:10] k [10:10] removed it now, thanks for the pointer [10:10] yw! [10:11] apw, do you know if my issue of "computer 'hangs' for a second every time you change the screen level through the keys" (latitude e6410) is a known kernel issue? [10:11] I should probably report a bug about that as well [10:11] * seb128 looks in launchpad [10:36] seb128, nope not heard of that, you should file a bug indeed [10:41] oh you're already rebuilding gcc? [10:42] yes, why not? [10:44] * Laney hugs pitti for adt-run --apt-source [10:45] alioth is down which breaks run-adt-test here but this with adt-virt-schroot is nice [10:46] ali1234: how does alioth break run-adt-tesZ? [10:46] me [10:46] it tries to check something out [10:46] Laney: yes, run-adt-test and prepare-testbed will go away, in favor of adt-virt-{lxc,qemu} [10:46] ah [10:46] I think autopkgtest itself? [10:47] pitti, btw, just as fyi https://bugzilla.gnome.org/show_bug.cgi?id=725390#c4 [10:47] Gnome bug 725390 in daemon "ipod displayed with an usb key icon" [Normal,Unconfirmed] [10:47] pitti, do you know what would be the right place for a such "udev quirk that tells udisks the icon to use"? [10:48] is udisks shipping some of those? [10:48] seb128: ah, so your ipod doesn't use afc [10:48] no [10:48] it's an old classic-storage one [10:48] seb128: I thought we have some hack which assigns the new icon to devices with ID_MEDIA_PLAYER [10:48] right, we had [10:48] seb128: does your device still have that tag? [10:48] but that stopped working with the new gvfs [10:48] yes [10:48] seb128: ah, so we probably dropped that hack at some point [10:49] it seems like you dropped it because the issue was resolved upstream [10:49] but they did some refactoring for symbolic icons and other stuff then [10:49] ah no, this is supposed to be fixed upstream [10:49] seb128: yes, our patch was against the gdu backend [10:49] they might have dropped the hack, which wouldn't be suprising if e.g bastien thinks that should be done in a quirk and not in code [10:50] seb128: right, the upstream solution wasn't a hack any more [10:50] seb128: I'll have a look in some mins, just finishing something else [10:50] pitti, k, let me know if you need debug infos from me [10:50] * pitti still fighting the backlog from the long weekend, and people broke stuff all over the place :) [11:03] Laney, why do you need a ffe for pango? isn't that a point/bugfix release? [11:03] to add the new package [11:03] oh, I should have read all the description :p [11:03] cool [11:08] just noticed they fail in debian [11:08] glib too :( [11:09] :-( [11:09] just a missing test-dep though it seems [11:42] cyphermox, hey, could you review https://code.launchpad.net/~matttbe/ubuntu/trusty/network-manager-applet/lp1267100/+merge/200825 ? [11:45] happyaron, is https://bugs.launchpad.net/ubuntu/+source/ibus-anthy/+bug/1279845 still on your todolist? [11:45] Launchpad bug 1279845 in ibus-anthy (Ubuntu) "ibus-anthy sets to jp keyboard layout forcibly." [Undecided,New] === MacSlow is now known as MacSlow|lunch === Trevinho__ is now known as Trevinho === Sweetsha1k is now known as Sweetshark === tkamppeter_ is now known as tkamppeter [12:36] Laney: do you know why we have custom encoding profiles for rhythmbox? http://bazaar.launchpad.net/~ubuntu-desktop/rhythmbox/ubuntu/revision/178 [12:36] people say they break cd ripping for them in bug #1086806 [12:36] Launchpad bug 1086806 in rhythmbox (Ubuntu) "Rhythmbox encoder missing plugin" [High,Triaged] https://launchpad.net/bugs/1086806 [12:36] and I get emailed because "I've touched rhythmbox last" [12:38] People were complaining that the defaults were crappy for ripping, so some guy came with these new ones [12:38] larsu, look at the bug referenced in that commit [12:40] ah I overlooked that there was a reference. Thanks! [12:40] I'm not sure it was ever completely right [12:41] it might be that things got fixed/made different upstream since and that those changes are not needed anymore, not sure [12:41] in fact [12:42] nm [12:42] no, it's not upstream [12:43] the problem is that I can't test it [12:43] but having cd ripping broken by default sounds quite ... bad [12:43] no CD drive? ;-) [12:43] ya [12:43] s/ya/no [12:43] did that user open a bug? [12:43] seb128: yes: https://bugs.launchpad.net/ubuntu/+source/rhythmbox/+bug/1086806 [12:43] Launchpad bug 1086806 in rhythmbox (Ubuntu) "Rhythmbox encoder missing plugin" [High,Triaged] [12:43] I understand "ya", no worry ;-) [12:43] their suggestion is to jut drop the ubuntu-presets [12:43] but I'm not sure that's a good idea [12:44] * larsu doesn't know a lot about gstreamer and default encodings [12:44] let me try here [12:44] that bug is there since 12.10, shrug [12:46] I can confirm the issue [12:46] yeah [12:46] Laney, are you debugging it? [12:47] (no need to be both debugging the same thing) [12:47] no I just checked it [12:47] you go ahead [12:47] k [12:47] shrug, rb is not being nice to glib [12:48] quite some warnings on start [12:49] seb128: is it fixed if you comment the "preset = ..." lines in /usr/share/rhythmbox/rhythmbox.gep? [12:49] that's the workaround people are using === alan_g_ is now known as alan_g [12:50] hey seb128, we're trying to wrap up the schedule for UDS and I just noticed a couple of client sessions that are pending approval: http://summit.ubuntu.com/uds-1403/review/ - I just wanted to give you a heads up in case you hadn't seen them [12:50] dpm, hey, I've not even started looking at vUDS scheduling, but thanks [12:51] larsu, yes, that works [12:51] seb128, no worries. If you want, I can approve them for you now and then you can take care of scheduling them [12:52] dpm, sounds good, thanks [12:53] seb128: I'm pretty sure we can drop the patch. Apparently Laney added it because gstreamer had bad default presets [12:53] larsu, did they change their presets? [12:53] but the ones in the rhythmbox patch are identical with the ones in /usr/share/gstreamer-1.0/presets [12:53] there's at least a few to choose from now [12:53] seb128, ok, approved. They should now appear on the "Schedule meetings" page for you to put in the schedule [12:53] I think before you got no choice [12:55] Laney: do you remember where you got the .prs files from? They are identical to the ones we're shipping with gstreamer [12:55] They came from a contributor [12:55] Conor made those [12:57] maybe we patch them into gstreamer? [12:57] larsu, I'm investigating, hold on ;-) [12:57] I don't think they are identical [12:57] I think you are looking at files installed by rhythmbox [12:57] Laney: bah, you're right. I did a dpkg -S on the dir, not the files [12:57] larsu, dpkg -S /usr/share/gstreamer-1.0/presets/* [12:58] right [12:58] it's probably orthogonal to the preset thing [12:58] I don't know/remember what that is supposed to achieve [12:58] I think the profiles just need their elements names updated [12:58] larsu, Laney: let me look at it, I'm going to come back to you, no need to be 3 looking at the same thing [12:58] okay [12:58] thanks! [12:58] yw! [12:59] hehe === alan_g is now known as alan_g|lunch [13:10] ok, I Hate -Bsymbolic [13:10] and -Bsymbolic-functions [13:26] larsu, Laney: seems https://bugs.launchpad.net/ubuntu/+source/rhythmbox/+bug/945987/comments/53 is right (I wonder how that ever worked) [13:26] Launchpad bug 945987 in rhythmbox (Ubuntu) "No Settings are available in "Preferred format", only preset defaults are used" [High,Confirmed] [13:26] changing the preset to be "ubuntu-default" in the .gep makes it work [13:27] not sure how Laney had a version with the right content (next comment on that bug), http://bazaar.launchpad.net/~ubuntu-desktop/rhythmbox/ubuntu/revision/178 suggests the name mismatch was there from the start [13:27] seb128: aha [13:27] so just fix the name [13:28] does that get rid of it for you? [13:28] ah col [13:28] *cool [13:28] it does get ride of the error [13:28] but the preset is not used by default [13:28] it's listed in the preferences combo as "ubuntu-default" [13:28] not very useful [13:29] I'm looking at changing the default rather [13:30] cool, it has a gsettings "media-type-presets" [13:30] org.gnome.rhythmbox.encoding-settings media-type-presets [13:30] which maps type to preset profiles === MacSlow|lunch is now known as MacSlow [13:31] so first fix is to fix the naming inconsistency (and maybe pick a better name, e.g rename it "Ubuntu") [13:31] since that's the name showing in the preferences combo [13:31] nod [13:31] then update ubuntu-settings to set a default for that key [13:31] larsu, Laney: thanks for pointing it out and for the help debugging ;-) [13:32] still disturbing that we have such bugs for several releases (good to fix it for the LTS as well) [13:34] seb128: :) [13:47] Hello seb128 === alan_g|lunch is now known as alan_g [13:48] GunnarHj, hey [13:48] seb128: Wondering about that mythes-sv package (bug 1284701). I got an FFe and have also uploaded to mentors.debian.net. Is there anything else I need to do? [13:48] Launchpad bug 1284701 in Ubuntu "[FFE] Please upload mythes-sv to the archive" [Medium,In progress] https://launchpad.net/bugs/1284701 [13:48] Sweetshark, hi :), is 4.2.2rc1 already ready for publishing? [13:48] GunnarHj, no, you might have noticed but the sponsoring queue is quite full and people are busy... [13:49] GunnarHj: I'm supposed to pilot tomorrow, will look then if nobody does before [13:50] seb128: I know there are quite a few items... Will try to be more patient. ;-) [13:50] Laney: Thanks! [13:50] if you could tell me (in the bug) how to check it works that will probably help [13:51] Laney: Ok, will do. [13:52] I spent some time sponsoring today to help driving the queue down [13:52] let's see where we manage to get [13:52] tack Gunnar! [13:55] Laney: ;-) Yeah, it's good you know some Swedish if you're going to check that thesaurus. [13:56] I do have a copy of The Bridge, maybe I should watch it without subtitles and see how far I get :P === Pici` is now known as Pici [13:58] Might be a good exercise.. [13:58] good morning cats [13:59] haha, there's a British/French remake with the body being found in the Channel Tunnel... [14:00] saluton desrt [14:00] Laney: bonega! vi lernis! [14:00] desrt, woof to you [14:00] or woo to you [14:00] seb128: no 'w' in esperanto :) [14:01] ooo? ;-) [14:01] ŭ [14:01] nur unu vorto [14:01] Laney: kaj tri pli [14:01] oh man, crazy language day again? [14:01] comes around every month :P [14:05] bregma, Trevinho: ChrisTownsend: did you guys notice the new compiz/unity misplacing buttons in some dialogs like the gnome-keyring ssh agent one? [14:05] they got placed over the text until you focus the dialog [14:05] then the dialog resizes and they get placed under it [14:07] seb128, I've noticed some GTK dialogs are distorted on high-DPI systems, like the unlock dialog nearly fills the whole screen and the text and buttons get lost [14:07] I haven't actually tested on an old-fashioned small screen lately [14:08] I'm using a zoom of 1 here [14:09] Laney: so what would you do if i introduced a patch to glib to make it fail ./configure if your C compiler was broken wrt. IEEE floating point behaviour as mandated by the C spec? [14:10] desrt, did you re-review the u-g-m work from attente yesterday? I would like to land that [14:11] ah! fell off my plate during my war with gcc [14:11] broken as in 'this is what gcc does'? [14:11] Laney: yes :) [14:12] Laney: you could fix this in a number of ways... [14:12] like giving some CFLAGS during the build [14:13] -fexcess-precision=standard, as an example, or -mfpmath=sse -march=pentium3 [14:13] you could also patch out the check [14:13] * desrt is hoping to start a discussion... [14:14] truly, -mfpmath=sse -march=pentium3 would be good for absolutely everyone who doesn't own a pentium 2... [14:15] -fexcess-precision=standard would get us working code, at least, though [14:16] couldn't we go further and use -march=pentium4 ? :P [14:16] and drop the i386 arch :x [14:16] we already use i686 [14:18] yeah I don't know if it's something we could do at the level of glib [14:19] i valid answer to this question is also "i would talk to doko" [14:19] * desrt is trying to guage the impact on other distros -- not ubuntu specifically [14:22] I would say "say Ryan, why don't you start a thread on ubuntu-devel about this?" [14:22] that'll smoke the right people out [14:22] sounds like a good plan [14:22] i'll do that after i review attente's patch :) [14:23] Debian would have to act similarly, and it might be harder to make this kind of change there [14:23] it seems that debian is very likely to be concerned with correctness [14:24] so they might want -fexcess-precision=standard anyway [14:24] what's the performance hit like? [14:25] i looked at the code... it's not pretty [14:25] in i686 you have mmx, it seems... and they use that if it's there [14:26] turning on standards means that it stops using mmx and goes back to oldschool fpu instructions [14:26] sse/sse2 got it right so if those are available they use them with preference [14:26] and indeed this is the default for amd64 (since all known amd64 implementations also have sse/sse2) [14:26] but with -march=i686, we don't know if we have sse/sse2 [14:29] OH GOD [14:29] I did schroot -e --all-sessions, which kills all open schroot sessions [14:30] guess what free software web browser engine had been building all day in another session [14:30] $browser_engine_that_shall_not_be_named 11 - 3 Laney [14:30] Laney, just give up and throw it at a ppa builder I say [14:31] yeah, fat fingers can't get in the way there ... [14:31] or dodgy kernels or whatever made my machine die overnight [15:14] seb128, how can I merge the proposed patch on https://code.launchpad.net/~philpem/ubuntu/trusty/ptouch-driver/fix-esc-i-a-for-pt-series/+merge/209351? Or do I simply upload the package with the patch applied? [15:15] tkamppeter, simply uploading with the patch is fine [15:16] seb128, OK, thanks. [15:16] tkamppeter, yw, thanks for looking at it! [15:18] Laney: discussion opened :) [15:19] or at least it will be once the moderator approves the post [15:23] cool, thanks for that [15:23] seb128, package uploaded. [15:24] Laney: i think i'm going to avoid fixing the test meanwhile [15:24] so you can continue to patch it out or you can build glib with some CFLAGS which cause the compiler to produce correct code [15:25] I uploaded the volatile patch [15:28] seb128, were you complaining about something with failing tests on the builders related to gsettings? === alan_g is now known as alan_g|tea [15:35] tkamppeter, thanks [15:35] kenvandine, no, why? [15:36] kenvandine, xnox was complaining about content-hub/gsettings slowness in the emulator yesterday, but that's only recent discussion around gsetting that I remember [15:37] seb128, ok, i thought i overheard you and desrt talking about this [15:37] kenvandine, we were discussing a segfault issue that's hitting u-s-d and something other softwares [15:39] ok, i am seeing segfaults in the test suite in a ppa [15:39] for content-hub [15:40] might not be related to gsettings at all [15:42] gsettings gives you SIGABRT, not SIGSEGV :) [15:43] desrt, of course it does... you love aborts :) === alan_g|tea is now known as alan_g [15:43] kenvandine, do you have a bt? [15:44] no... it's only on the ppa builders [15:44] kenvandine, https://bugs.launchpad.net/ubuntu/+source/glib2.0/+bug/1284647 is the bug I was mentioning [15:44] not sure how to get that [15:44] Launchpad bug 1284647 in glib2.0 (Ubuntu) "gsettings crashed with SIGSEGV in g_settings_backend_dispatch_signal()" [High,Triaged] [15:50] hmm, this compiz window size bug is starting to get annoying :-/ [15:51] which one? [15:52] https://bugs.launchpad.net/ubuntu/+source/compiz/+bug/1287472 i think [15:52] Launchpad bug 1287472 in Compiz "compiz unnecessarily shrinks new windows" [Medium,Triaged] [15:53] seems like Trevinho is on it [15:53] Laney, larsu: https://launchpad.net/ubuntu/+source/rhythmbox/3.0.1-1ubuntu12 and https://launchpad.net/ubuntu/+source/ubuntu-settings/14.04.5 [15:54] seb128: awesome, thanks! [15:54] yw [16:19] seb128: yeah, one of the tons things I'm doing right now :) [16:21] Trevinho, hey, I also assigned you a new segfault for the client side decoration (not sure if noticed), showed up on e.u.c today [16:22] seb128: yeah, I saw the email... then my firefox decided not to load gmail... (and I can't restart it now) :) [16:26] seb128: nice one [16:27] thanks [16:35] seb128, sorry, was in a meeting... the armhf builders for ppas, those are virtual aren't they? only the archive gets real hardware? [16:38] kenvandine, depends of the ppa I think, normal ppas don't have armhf, do they? [16:38] no... but my ppa does [16:39] but i think by default they're virtual [16:39] that's likely yes [16:39] i bet it's hitting this same qemu thing signond hit a while back === gatox is now known as gatox_lunch === dbarth_ is now known as dbarth-onair === gatox_lunch is now known as gatox === dbarth-onair is now known as dbarth [18:01] * Sweetshark hugs mterry! [18:01] mterry: thanks for the MIRs! [18:02] Sweetshark, :) [18:04] ... esp. since I missed the subscriptions ... [18:05] mterry: seb128 and me werent sure: Do I need another one for https://launchpad.net/ubuntu/+source/libreoffice-dictionaries ? Its essentially a source package rename from openoffice.org-dictionaries -- same codebase and all binary packages from it used to be in main before. === alan_g is now known as alan_g|EOD [18:06] Sweetshark, is there new packaging? [18:07] Sweetshark, if packaging changed, it doesn't hurt to get a quick once over, but meh. Not a *huge* deal either way [18:07] Sweetshark, I would just run through the checklist and make sure a team is subscribed to bugs, etc [18:08] mterry: the packaging was updated, based on the old stuff. seb128 had a glace over it. mostly updating debhelper, killing dpatch and some lintian nags. [18:08] Sweetshark, just make sure a bug subscriber exists and you're fine [18:09] mterry: 'The subscription has been created.' [18:09] (30 seconds before this conversation) [18:09] :) [18:10] * Sweetshark is no a quick learner. But on the third try, there is a chance that he gets it. [18:10] heh === FJKong is now known as FJKong_afk === hggdh_ is now known as hggdh === m_conley is now known as m_conley_away === mjohnson15_2 is now known as mjohnson15 === m_conley_away is now known as m_conley