Hi, krita 5.2.3-1 seg faults each time I try to copy and paste a pic from one Krita tab to another. The seg fault happens a little bit delayed shortly after pushing Ctrl+V or using the paste option of the menu. The last time I used Krita for exactly the same work but without this issue was 2024-05-12, so that time I used 5.2.2-9. IOW between 2024-05-12 and 2024-07-21 something changed. The GNU Debugger output is posted at https://gitlab.archlinux.org/archlinux/packaging/packages/krita/-/issues/4 . • rocketmouse@archlinux ~ $ lsb_release -dr Description: Arch Linux Release: rolling • rocketmouse@archlinux ~ $ grep \ Driver /etc/X11/xorg.conf Driver "modesetting" • rocketmouse@archlinux ~ $ pacman -Q krita krita 5.2.3-1 • rocketmouse@archlinux ~ $ wmctrl -m Name: Openbox Class: PID: N/A Window manager's "showing the desktop" mode: OFF Regards, Ralf
$ pacman -Q xf86-video-intel xf86-video-intel 1:2.99.917+923+gb74b67f0-2
I have created a package for your appimage and I'm encountering the same problem when using the appimage. Inserting an image results in a seg fault. • rocketmouse@archlinux /mnt/winos10/share/.archive/krita $ krita ipad2+pro_1228x5_921x3_2024-07-20.kra QObject::startTimer: Timers cannot have negative intervals Could not find platform independent libraries <prefix> Could not find platform dependent libraries <exec_prefix> Consider setting $PYTHONHOME to <prefix>[:<exec_prefix>] /tmp/.mount_kritadktvTy/usr/lib/krita-python-libs/krita added to PYTHONPATH Segmentation fault (core dumped) • rocketmouse@archlinux /mnt/winos10/share/.archive/krita $ pacman -Qi krita | head -8 Name : krit Version : 5.2.3-1.3 Description : Edit and paint images with the Krita Appimage Architecture : x86_64 URL : https://krita.org Licenses : GPL3 Groups : None Provides : kritaI have created a package for your appimage and I'm encountering the same problem when using the appimage. Inserting an image results in a seg error. • rocketmouse@archlinux /mnt/winos10/share/.archive/krita $ krita ipad2+pro_1228x5_921x3_2024-07-20.kra QObject::startTimer: Timers cannot have negative intervals Could not find platform independent libraries <prefix> Could not find platform dependent libraries <exec_prefix> Consider setting $PYTHONHOME to <prefix>[:<exec_prefix>] /tmp/.mount_kritadktvTy/usr/lib/krita-python-libs/krita added to PYTHONPATH Segmentation fault (core dumped) • rocketmouse@archlinux /mnt/winos10/share/.archive/krita $ pacman -Qi krita | head -8 Name : krit Version : 5.2.3-1.3 Description : Edit and paint images with the Krita Appimage Architecture : x86_64 URL : https://krita.org Licenses : GPL3 Groups : None Provides : krita
"Inserting an image results in a seg fault" should read "Pasting an image results in a seg fault". My apologies for the duplicated text by Comment 2. Is it possible to edit comments? I don't see an option to do so.
I cannot reproduce this issue on Windows. It seems like it is something linux-specific. Ralf, could you tell, do you use X11 or Wayland?
Hi Dmitry, I'm still a X user. From comment 1: $ grep \ Driver /etc/X11/xorg.conf Driver "modesetting" A copy of comment 2: $ pacman -Q xf86-video-intel xf86-video-intel 1:2.99.917+923+gb74b67f0-2 And FWIW additional info. CPU with integrated GPU: $ hwinfo --cpu | grep Model | sort -u Model: 6.191.5 "13th Gen Intel(R) Core(TM) i3-13100" Kernel: $ uname -r 6.10.0-arch1-1 Kernel parameters: $ cat /proc/cmdline BOOT_IMAGE=/boot/vmlinuz-linux root=/dev/disk/by-label/m1.archlinux ro threadirqs ibt=off Regards, Ralf
I can reproduce this with the 5.2.3 appimage. I also see krita(3726868)/(default) kis_assert_common: SAFE ASSERT (krita): "!nodes.isEmpty()" in file /home/halla/dev/5.2/libs/image/kis_layer_utils.cpp, line 1599 And lots of krita(3727282)/(krita.general) KisToolRectangleBase::endStroke: Unexpected tool event has come to endStroke while being mode 0 ! krita(3727282)/(krita.general) KisToolRectangleBase::endStroke: Unexpected tool event has come to endStroke while being mode 0 ! I cannot reproduce the crash in gdb, though, so I don't have a backtrace. ctrl-a; ctrl-c; ctrl-tab; ctrl-v doesn't paste anything, even though the paste is enabled. (Oh, and trying to right-click in the layerbox to paste the clipboard doesn't toggle the menu unless I click on an existing layer.)
*** Bug 490995 has been marked as a duplicate of this bug. ***
As Freya pointed out on the forums, this seems to be related to bug #490843. Trying to paste Krita layer contents into a document with no actively selected layer leads to the segfault. I can consistently get the crash so long as I can get to a state with no active layer selected (by switching tabs) The crash never happens so long as I ensure a layer is selected before pasting.
Thanks Freya and Ralek! I can confirm that this bug has the same cause as bug 490843 and is fixed with its fix. *** This bug has been marked as a duplicate of bug 490843 ***
Git commit 0e0bcc3c614220f96747fd30bbc099afaee88291 by Dmitry Kazakov. Committed on 31/07/2024 at 09:09. Pushed by dkazakov into branch 'master'. Make clip-pasting code a bit more robust We should check if the active layer is actually non-null when pasting a mime-data. It is possible that no layer is selected, but it we still shouldn't crash because of that. M +4 -2 libs/ui/kis_mimedata.cpp M +1 -1 libs/ui/kis_node_manager.cpp https://invent.kde.org/graphics/krita/-/commit/0e0bcc3c614220f96747fd30bbc099afaee88291
Git commit bd541c7c28c1e96aff466cc5c691d0e2014d2094 by Dmitry Kazakov. Committed on 31/07/2024 at 09:11. Pushed by dkazakov into branch 'krita/5.2'. Make clip-pasting code a bit more robust We should check if the active layer is actually non-null when pasting a mime-data. It is possible that no layer is selected, but it we still shouldn't crash because of that. M +4 -2 libs/ui/kis_mimedata.cpp M +1 -1 libs/ui/kis_node_manager.cpp https://invent.kde.org/graphics/krita/-/commit/bd541c7c28c1e96aff466cc5c691d0e2014d2094
*** Bug 491408 has been marked as a duplicate of this bug. ***
*** Bug 491414 has been marked as a duplicate of this bug. ***
*** Bug 491418 has been marked as a duplicate of this bug. ***
*** Bug 491793 has been marked as a duplicate of this bug. ***
*** Bug 491877 has been marked as a duplicate of this bug. ***