I’ve got a segmentation fault thrice on this beta, macOS qt6. I’m not able to consistently reproduce the error, but it happened while I was manipulating and copy pasting the images in editor.
That’s infinitely more seg fault crashes than I got on previous versions. It’s probably a qt bug though.
Did you use 2.1.52 for long before that? The Qt 6.3.0 change in 2.1.52 seems more likely to have caused such problems than any changes made in the beta. If the crashes don’t occur in 2.1.52, can you reproduce them on latest main when running with tools/runopt? If so, would you be able to bisect the changes to figure out where the problem was introduced?
There´s something wrong with 2.1.52 and 2.1.53. Anki crashes without no reason and then it´s imposible to start Anki without restart the mac. I´m using latest beta qt6. This is not happening with 2.1.50.
I am having the same issue (random crashes) with 2.1.52. I’m on main release qt6 for apple silicon. For me it is getting bothersome, but does not require restart. If I force quit from activity monitor I can reload Anki and at least it usually has saved up until the very last thing I did (so no functional impairments)
I don’t know if that is a related bug, but I have a long-standing similar issue, that I reported here. As I explained on that thread, I cannot simply download binaries and run them (even if they are mostly self-contained), so I can only try the latest packaged version of Anki. Thankfully, my distribution is keeping up the pace, so the latest version packaged is current 2.1.52 (and basically all it does is fetch the binaries and setup them in a way that they can be ran).
My issue, however, is much more sever than just crashing: Anki sometimes crash, but sometimes literally kills my GPU. I can still connect to my computer via ssh, but the display will just freeze until hard reboot.
I think my issue is different. I’m not closing a webview window when this error happens, just editing/focusing in an edit window. Also, unlike this crash, I can reliably trigger the crash using the method in the linked issue, and when I do the error log I get is different.
2.1.53 has now been released as stable, and it includes a fix that seems to address the main cause of crashes people were experiencing. If you’re still experiencing crashes after updating, please let us know on a new thread.