After updating to kernel 6.13 I’ve noticed that my system will occasionally freeze. The issue may have also happened on kernel 6.12 but I skipped by it since I thought it might have been a fluke. The screen will stay up but will be unresponsive until I hold down the power button. I’ve collected the dmesg log from the last run and there are multiple messages like the one below from around the time it freezes.
watchdog: BUG: soft lockup - CPU#4 stuck for 27s! [gnome-shell:3063]
Please note that the 6.13 kernel is still being tested (both vanilla and rawhide).
Anything you install from rawhide is non-standard for fedora 41 so most fedora users will not be able to assist. The same applies to the 6.13.0-361.vanilla.fc41.x86_64 vanilla kernel that you display here. Non-standard kernels are outside the normal support on this forum.
I applaud your efforts but suggest you ask for support from the upstream sources when using non-standard software.
Hey thanks for the response and thanks for pointing that out. I forgot I had enabled the vanilla kernels awhile ago so that I had have access to amd PM fix that stopped my laptop from freezing when recovering from sleep.
I switched back to the normal fedora kernels and I’ll see if the freezing still happens.
I downgraded to 6.12.10-200.fc41.x86_64 and I haven’t seen the freeze yet. But I have been on vacation so not using my laptop as much at the moment. Are you using the fedora kernel or the vanilla kernel?
I’m seeing hard OS freezes when manipulating specific files (for ex. dropping a large image into Firefox) on both 6.13 and recent 6.13.1 using the CachyOS kernel on Silverblue. I was hopeful when the new patch mentioned some possibly related commits but nu u uh
libfs: Return ENOSPC when the directory offset range is exhausted
Revert "libfs: Add simple_offset_empty()"
Revert "libfs: fix infinite directory reads for offset dir"
libfs: Replace simple_offset end-of-directory detection
libfs: Use d_children list to iterate simple_offset directories