Bug#1033618: kscreen_osd_service segfaults
Package: kscreen Version: 4:5.27.2-1 Severity: normal X-Debbugs-Cc: dmitr...@gmail.com kscreen_osd_service segfault when I hit Win+P and cancel switching between screens. If that matters it happens in Plasma Wayland session. Stack trace from journalctl is below: Mar 28 22:34:09 zinc dbus-daemon[1053]: [session uid=1001 pid=1053] Activating via systemd: service name='org.kde.kscreen.osdService' unit='plasma-kscreen-osd.service' requested by ':1.16' (uid=1001 pid=1252 comm="/usr/bin/kded5") Mar 28 22:34:09 zinc systemd[1034]: Starting plasma-kscreen-osd.service - KScreen OSD service... Mar 28 22:34:09 zinc dbus-daemon[1053]: [session uid=1001 pid=1053] Successfully activated service 'org.kde.kscreen.osdService' Mar 28 22:34:09 zinc systemd[1034]: Started plasma-kscreen-osd.service - KScreen OSD service. Mar 28 22:34:09 zinc kscreen_osd_service[12991]: qrc:/qml/OsdSelector.qml:71:9: QML Heading: Binding loop detected for property "verticalAlignment" Mar 28 22:34:12 zinc kernel: kscreen_osd_ser[12991]: segfault at 3256 ip 7fdeeb34631e sp 7ffe549895f0 error 4 in libQt5Gui.so.5.15.8[7fdeeb2e2000+4d] likely on CPU 1 (core 0, socket 0) Mar 28 22:34:12 zinc kernel: Code: 15 07 1b 00 00 41 55 44 0f b6 ee 41 54 55 53 48 83 ec 20 48 8b 5f 08 64 48 8b 04 25 28 00 00 00 48 89 44 24 18 31 c0 48 8b 03 <48> 8b 40 20 48 39 d0 0f 85 bd 00 00 00 4c 8b 63 08 89 f5 40 3a 73 Mar 28 22:34:12 zinc systemd[1]: Started systemd-coredump@2-13015-0.service - Process Core Dump (PID 13015/UID 0). Mar 28 22:34:12 zinc systemd-coredump[13016]: [LNK] Process 12991 (kscreen_osd_ser) of user 1001 dumped core. Module libudev.so.1 from deb systemd-252.6-1.amd64 Module libsystemd.so.0 from deb systemd-252.6-1.amd64 Stack trace of thread 12991: #0 0x7fdeeb34631e _ZN7QWindow10setVisibleEb (libQt5Gui.so.5 + 0x14631e) #1 0x7fdeeaee8f4f n/a (libQt5Core.so.5 + 0x2e8f4f) #2 0x7fdeeaee229f _ZN7QObject9destroyedEPS_ (libQt5Core.so.5 + 0x2e229f) #3 0x7fdeeaee7254 _ZN7QObjectD2Ev (libQt5Core.so.5 + 0x2e7254) #4 0x7fdeec6494b9 _ZN7KScreen6OutputD0Ev (libKF5Screen.so.8 + 0x354b9) #5 0x562cc7fc7165 n/a (kscreen_osd_service + 0x9165) #6 0x562cc7fc71a9 n/a (kscreen_osd_service + 0x91a9) #7 0x562cc7fc4b9f n/a (kscreen_osd_service + 0x6b9f) #8 0x7fdeeaedd6f0 _ZN7QObject5eventEP6QEvent (libQt5Core.so.5 + 0x2dd6f0) #9 0x7fdeeaeb16f8 _ZN16QCoreApplication15notifyInternal2EP7QObjectP6QEvent (libQt5Core.so.5 + 0x2b16f8) #10 0x7fdeeaeb4681 _ZN23QCoreApplicationPrivate16sendPostedEventsEP7QObjectiP11QThreadData (libQt5Core.so.5 + 0x2b4681) #11 0x7fdeeaf0a153 n/a (libQt5Core.so.5 + 0x30a153) #12 0x7fdee92627a9 g_main_context_dispatch (libglib-2.0.so.0 + 0x547a9) #13 0x7fdee9262a38 n/a (libglib-2.0.so.0 + 0x54a38) #14 0x7fdee9262acc g_main_context_iteration (libglib-2.0.so.0 + 0x54acc) #15 0x7fdeeaf09836 _ZN20QEventDispatcherGlib13processEventsE6QFlagsIN10QEventLoop17ProcessEventsFlagEE (libQt5Core.so.5 + 0x309836) #16 0x7fdeeaeb017b _ZN10QEventLoop4execE6QFlagsINS_17ProcessEventsFlagEE (libQt5Core.so.5 + 0x2b017b) #17 0x7fdeeaeb82d6 _ZN16QCoreApplication4execEv (libQt5Core.so.5 + 0x2b82d6) #18 0x562cc7fc38d2 n/a (kscreen_osd_service + 0x58d2) #19 0x7fdeeaa4618a n/a (libc.so.6 + 0x2718a) #20 0x7fdeeaa46245 __libc_start_main (libc.so.6 + 0x27245) #21 0x562cc7fc3961 n/a (kscreen_osd_service + 0x5961) Stack trace of thread 12992: #0 0x7fdeeab1b0af __poll (libc.so.6 + 0xfc0af) #1 0x7fdee92629ae n/a (libglib-2.0.so.0 + 0x549ae)
Bug#1031027: libprometheus-cpp-core1.0: Library is built w/o optimization enabled
Hi, Gregor. This is strange. I've backported two packages (mimalloc and prometheus-cpp) to ubuntu bionic. After that I've noticed a huge performance degradation compared to local builds in /usr/local. After that I investigated the issue and found out that both packages do not specify CMAKE_BUILD_TYPE during build and after I've added it into debian/rules the performance was restored. Well, maybe debhelper in bookworm has other defaults then it has in ubuntu. Sorry if I'm wrong. сб, 11 февр. 2023 г. в 00:12, Gregor Jasny : > Hello, > > On 10.02.23 16:44, Max Dmitrichenko wrote: > > The CMakeList.txt of the library doesn't contain default > CMAKE_BUILD_TYPE to be set to Release. Neither this variable is not set in > debian/rules file. So the resulting library is build with no optimization > enabled which hits performance by factor of around 2x. > > > Setting CMAKE_BUILD_TYPE to None and using the distribution default > C(XX)FLAGS is what debhelper defaults to [1]. > > In the prometheus-cpp build logs [2] I can see that the "-O2" > optimization flags get applied. > > How did you measure the performance degradation? > > Thanks, > Gregor > > [1] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=701233#27 > [2] > > https://buildd.debian.org/status/fetch.php?pkg=prometheus-cpp=amd64=1.0.1-1=1653219882=0 > > > -- With best regards Maksim Dmitrichenko
Bug#767756:
Sorry guys, may be I'm telling well known things but Ubuntu has libc6-prof package which installs alternative version of libc6 with no-omit-frame-pointer to /lib/libc6-prof/x86_64-linux-gnu/ . Any who need to profile his program can use it via LD_LIBRARY_PATH environment variable while other programs by default use optimized version. Why reinvent the wheel? -- With best regards Maksim Dmitrichenko
Bug#1005063: lxpanel: Keyboard layout switching is broken
Package: lxpanel Version: 0.10.1-2+rpt11 Severity: important Tags: l10n X-Debbugs-Cc: dmitr...@gmail.com Dear Maintainer, When you configure keyboard layouts via taskbar applet, you can't switch to any layout except the first one. You can notice that for a fraction of second layout is switched but then something triggers a switch to the first layout. -- System Information: Distributor ID: Raspbian Description:Raspbian GNU/Linux 11 (bullseye) Release:11 Codename: bullseye Architecture: armv7l Kernel: Linux 5.10.92-v7l+ (SMP w/4 CPU threads) Kernel taint flags: TAINT_CRAP Locale: LANG=ru_RU.UTF-8, LC_CTYPE=ru_RU.UTF-8 (charmap=UTF-8) (ignored: LC_ALL set to ru_RU.UTF-8), LANGUAGE=ru_RU.UTF-8 Shell: /bin/sh linked to /usr/bin/dash Init: systemd (via /run/systemd/system) Versions of packages lxpanel depends on: ii libasound2 1.2.4-1.1+rpt2 ii libc62.31-13+rpt2+rpi1+deb11u2 ii libcairo21.16.0-5+rpt1 ii libcurl3-gnutls 7.74.0-1.3+deb11u1 ii libfm-gtk4 1.3.2-1+rpt3 ii libfm-modules1.3.2-1+rpt3 ii libfm4 1.3.2-1+rpt3 ii libgdk-pixbuf-2.0-0 2.42.2+dfsg-1 ii libglib2.0-0 2.66.8-1 ii libgtk-3-0 3.24.24-4+rpt3 ii libiw30 30~pre9-13.1 ii libkeybinder-3.0-0 0.3.2-1.1 ii libmenu-cache3 1.1.0-1.1 ii libpango-1.0-0 1.46.2-3 ii libwnck-3-0 3.36.0-1 ii libx11-6 2:1.7.2-1 ii libxml2 2.9.10+dfsg-6.7 ii lxmenu-data 0.1.5-2.1 ii lxpanel-data 0.10.1-2+rpt11 Versions of packages lxpanel recommends: ii libnotify-bin 0.7.9-3 ii lxterminal [x-terminal-emulator] 0.4.0-1+rpt2 ii notification-daemon 3.20.0-4 ii pavucontrol 4.0-2 ii xkb-data 2.29-2 Versions of packages lxpanel suggests: ii chromium-browser [www-browser] 95.0.4638.78-rpt6 ii dillo [www-browser] 3.0.5-7+rpt1 ii menu2.1.48 -- no debconf information
Bug#1004266:
It wasn't mentioned in previous bug reports but packages also missing GeoTrust RSA CA 2018 certificate which actually triggered me into investigation of issues with missing certificates in ca-certificates package. -- With best regards Maksim Dmitrichenko
Bug#1002546: RFS: prometheus-cpp/1.0.0-4 [ITP] -- Prometheus Client Library for Modern C++
Done. Reuploaded 1.0.0-1 version with all changes incorporated. пт, 24 дек. 2021 г. в 13:59, Bastian Germann : > > On Fri, 24 Dec 2021 01:23:15 +0300 Max Dmitrichenko wrote: > > Changes for the initial release: > > > > prometheus-cpp (1.0.0-4) unstable; urgency=medium > > . > >* Provide static libraries with -dev package. > > > > prometheus-cpp (1.0.0-1) unstable; urgency=medium > > . > >* Initial release. Closes: #917817 > > Please keep only the 1.0.0-1 entry as long as this is not sponsored. > Report your changes on this issue on new uploads instead. -- With best regards Maksim Dmitrichenko
