/srv/irclogs.ubuntu.com/2022/03/29/#ubuntu-desktop.txt

KGB-0gnome-shell Daniel van Vugt 303043 * commented merge request !60 * https://deb.li/iU3Dz05:00
=== cpaelzer_ is now known as cpaelzer
oSoMoNgood morning desktoppers05:54
dufluMorning oSoMoN 06:08
jibelgood morning all06:28
oSoMoNhey duflu, salut jibel 06:37
dufluMorning jibel 06:39
Nafallomornign07:25
Nafallos/gn/ng/07:25
lunamorning07:29
dufluHi Nafallo and luna 07:38
oSoMoNgood morning Nafallo, luna 07:48
seb128goood morning desktopers07:59
oSoMoNsalut seb128, ça va?08:00
seb128lut oSoMoN, nickel, et toi ?08:00
oSoMoNça va, malgré une nuit trop courte08:01
seb128:-(08:04
dufluHi seb128 08:13
seb128hey duflu 08:13
KGB-2gnome-shell Daniel van Vugt 303052 * commented merge request !59 * https://deb.li/cDHF08:46
KGB-0gnome-shell Daniel van Vugt 303058 * commented merge request !60 * https://deb.li/Vncs09:12
seb128tjaalton, so since the ppa git build isn't helping, what's next? do you need me to report the issue upstream on gitlab?11:47
tjaaltonseb128: that would be fine, thanks. so far it's just the crocus (g-ext-app, gnome-chess) and iris (gnome-chess) drivers that are causing crashes, on xorg11:48
seb128'just'11:49
tjaaltonswrast (llvmpipe), and i965 are fine11:49
tjaaltonaren't we defaulting to wayland now?11:49
seb128defaulting yes, xorg is still important11:50
tjaaltonupstream had a look at the trace, but there's nothing that would explain why it fails on xorg11:50
seb128also https://bugs.launchpad.net/ubuntu/+source/webkit2gtk/+bug/1966418 is another regression from the mesa update and is under wayland 11:50
ubottuLaunchpad bug 1966418 in webkit2gtk (Ubuntu) "[jammy regression] Evolution does not display message content anymore" [High, Confirmed]11:50
tjaaltonright11:50
seb128tjaalton, k, reported https://gitlab.freedesktop.org/mesa/mesa/-/issues/622911:55
ubottuIssue 6229 in mesa/mesa "segfaults under X11 since the Ubuntu update to 22" [Opened]11:55
seb128tjaalton, should I try to bisect mesa locally or better to wait for upstream?11:58
jbichagood morning12:04
seb128jbicha, hey, how are you?12:05
tjaaltonseb128: best to wait12:09
seb128tjaalton, thanks12:10
seb128tjaalton, though your quote saying they replying having no idea doesn't sound promising12:10
tjaaltonI can try bisect later12:10
seb128k, in any case I can make time to help if needed, let me know12:10
tjaaltonok, if you know how to bisect mesa then by all means12:12
tjaaltonalso12:12
tjaaltonno need to build it as packages12:12
tjaalton3556 commits between 21.3-branchpoint..22.0-branchpoint12:13
seb128tjaalton, I know how to git bisect, I didn't try on mesa specifically. Is https://wiki.ubuntu.com/X/BisectingMesa somewhat still accurate?12:20
tjaaltonthat's >12y old, probably not12:22
seb128:)12:23
tjaaltonat least need to use meson now12:23
seb128I will just copy the arguments from the debian/rules12:23
tjaaltonstracing gnome-chess seems it's barfing when loading the shader cache12:24
oSoMoNricotz, the firefox-beta.bionic branch has unity-menubar.patch disabled ?12:38
ricotzoSoMoN, yes, didn't I mention this?12:41
oSoMoNif you did, I missed it12:41
oSoMoNI'll look into it12:41
ricotzoh, I see, sorry :(12:42
ricotzoSoMoN, https://bugzilla.mozilla.org/show_bug.cgi?id=175740212:42
ubottuMozilla bug 1757402 in Core "Add native context menu support on GTK" [--, Resolved: Fixed]12:42
seb128tjaalton, hum, any idea why using the same meson parameters than the build I don't end up with a iris_dri.so or crocus_dri.so in my build dir?12:48
ricotzmissing conditional build-dep? hey seb128 :)12:49
seb128ricotz, hey12:52
tjaaltonseb128: what do you get?12:54
seb128tjaalton, it's there after a ninja install but not in the builddir12:55
seb128so I guess that's generated in an extra step12:55
tjaaltonit's all in the same blob12:55
seb128$ ls build/src/mesa/drivers/dri12:56
seb128common  i915  i965  libmesa_dri_drivers.so  libmesa_dri_drivers.so.p  nouveau  r200  radeon12:56
tjaaltonthe dri-drivers are hardlinked12:56
seb128so I was unsure what to set LIBGL_DRIVERS_PATH to12:56
tjaaltonright12:56
tjaaltonhm, MESA_LOADER_DRIVER_OVERRIDE tells which driver to use12:57
tjaaltonLIBGL_DRIVERS_PATH is for the driver path12:57
seb128tjaalton, it's ok, I did ninja install and now LIBGL_DRIVERS_PATH=/tmp/install/usr/lib/x86_64-linux-gnu/dri/ works12:57
tjaaltonright, it'll use the default driver for the hw12:58
seb128tjaalton, k, thx, so git good/bad worked, 11 steps to bisect, it's ongoing13:05
seb128k, Ken isn't available but I will review the incoming rls bugs still14:06
seb128http://reqorts.qa.ubuntu.com/reports/rls-mgr/rls-jj-incoming-bug-tasks.html14:06
seb128quite a busy list now14:06
seb128https://bugs.launchpad.net/ubuntu/+source/gnome-shell/+bug/196680814:06
ubottuLaunchpad bug 1966808 in mutter (Ubuntu) "[amdgpu][radeon] gnome-shell Wayland sessions fail to start (Failed to lock front buffer on /dev/dri/card1: drmModeAddFB2WithModifiers failed: Invalid argument)" [High, New]14:06
seb128sounds like Daniel is working on this one14:07
seb128bug #196622114:07
ubottuBug 1966221 in mesa (Ubuntu) "gnome-extensions-app crashes with 'Segmentation fault (core dumped)'" [High, Confirmed] https://launchpad.net/bugs/196622114:07
seb128I'm bisecting that atm14:07
seb128bug #196641814:08
ubottuBug 1966418 in webkit2gtk (Ubuntu) "[jammy regression] Evolution does not display message content anymore" [High, Confirmed] https://launchpad.net/bugs/196641814:08
seb128another one that has to do with the new mesa14:08
seb128jbicha, could you open a new webkitgtk bug upstream? they said it doesn't seem the same issue than the one we commented on14:08
seb128bug #196657114:09
ubottuBug 1966571 in mesa (Ubuntu) "libva error: /usr/lib/x86_64-linux-gnu/dri/i965_drv_video.so init failed" [High, Confirmed] https://launchpad.net/bugs/196657114:09
seb128tjaalton, ^ states that a no change rebuild of intel-media-driver fixes things, is there an unwanted ABI change there? should I upload a rebuild?14:11
seb128bug #194920014:12
ubottuBug 1949200 in libcanberra (Ubuntu) "gnome-shell crashed with SIGTRAP from error 'Can't create a GtkStyleContext without a display connection' in gtk_style_context_init" [High, Triaged] https://launchpad.net/bugs/194920014:12
seb128I'm going to investigate a bit that one14:12
seb128I'm ignoring some of the assigned or fix commited reports14:12
seb128oSoMoN, bug #1964036 seems one we should investigate14:13
ubottuBug 1964036 in firefox (Ubuntu) "Firefox icon missing from panel following Kubuntu upgrade 20.04->22.04" [Undecided, New] https://launchpad.net/bugs/196403614:13
seb128bug #1966167, seems an annoyance worth fixing but probably not rls?14:14
ubottuBug 1966167 in gnome-shell-extension-ubuntu-dock (Ubuntu) "Show Applications button is not clickable in the corner of the screen" [Undecided, Confirmed] https://launchpad.net/bugs/196616714:14
seb128bug #1966317 seems annoying, unsure how we can handle that better though14:15
ubottuBug 1966317 in gvfs (Ubuntu) "gvfsd-fuse crashed with SIGABRT in __pthread_setname_np()" [Undecided, New] https://launchpad.net/bugs/196631714:15
oSoMoNseb128, yes, I was looking at that one yesterday, but I ran into issues dist-upgrading my focal kubuntu VM, and put it on hold for more pressing matters14:16
seb128oSoMoN, ack, should I assign to you?14:16
oSoMoNI want to get back to it14:16
oSoMoNplease do14:16
seb128oSoMoN, thanks14:16
seb128tjaalton, git bisect ended up on blaming https://gitlab.freedesktop.org/mesa/mesa/-/commit/60d95c5d0fee14:18
ubottuCommit 60d95c5 in mesa/mesa "Auto-enable TLSDESC support"14:18
seb128that was it for jj incoming, sorry got sidetracked14:27
seb128I will only do that list today since we focus on beta this week14:27
tjaaltonseb128: re libva, huh.. go for it14:40
seb128tjaalton, ack14:40
tjaaltonseb128: let me revert that from mesa and see if it fixes this for me too14:41
seb128tjaalton, I confirmed that a revert of the commit fixes it for me, now unsure what's the issue with our toolchain and that change but seems pretty low risk to revert14:42
tjaaltonright14:42
tjaaltonI'll revert it from the archive version, not the ppa one14:44
tjaalton17:45 < ajax> that's wild14:46
tjaaltongood work ;)14:46
seb128:)14:48
seb128tjaalton, on what channel is that btw?14:51
seb128I should maybe have joined directly14:51
tjaalton#dri-devel14:51
tjaaltonthis was part of a bigger merge request14:51
tjaaltonso dropping the enabling bits made it to not hit those anymore14:51
seb128there is only one other commit in that merge request it seems?14:52
tjaaltonright14:53
KGB-0atk tags ce90ab8 Samuel Thibault upstream/2.38.0 * Upstream version 2.38.0 * https://deb.li/web718:45
KGB-2atk upstream/latest fc9e4e8 Samuel Thibault * pushed 40 commits (first 5 follow) * https://deb.li/LeLy18:46
KGB-2atk upstream/latest 5b4a0bd Źmicier Turok po/be.po * Update Belarusian translation * https://deb.li/31pg418:46
KGB-2atk upstream/latest d5a602b Danial Behzadi po/fa.po * Update Persian translation * https://deb.li/3is8M18:46
KGB-2atk upstream/latest 4dc2638 Rūdolfs Mazurs po/lv.po * Update Latvian translation * https://deb.li/iQtlj18:46
KGB-2atk upstream/latest 094a280 Daniel Mustieles po/es.po * Updated Spanish translation * https://deb.li/0p2018:46
KGB-2atk upstream/latest 1ba0aa4 Kukuh Syafaat po/id.po * Update Indonesian translation * https://deb.li/iiS9218:46
jbichaseb128: filed https://bugs.webkit.org/show_bug.cgi?id=23851318:52
ubottubugs.webkit.org bug 238513 in WebKit "[GTK] EGLDisplay Initialization failed on Ubuntu 22.04" [Normal, New]18:52
KGB-0adwaita-icon-theme upstream/latest bb1e575 Jeremy Bicha (2574 files in 57 dirs) * New upstream version 42~really41.0 * https://deb.li/38VFn19:11
KGB-0adwaita-icon-theme pristine-tar 759e763 Jeremy Bicha adwaita-icon-theme_42~really41.0.orig.tar.xz.delta adwaita-icon-theme_42~really41.0.orig.tar.xz.id * pristine-tar data for adwaita-icon-theme_42~really41.0.orig.tar.xz * https://deb.li/HVve19:11

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!