You will need to generate some kind of call stack/graph info to start making sense of the mess your running kernel gets itself into in the scenario you described. I imagine that looking at `perf` (`perf top` in particular; it should suffice to paste a few instances of its topmost lines when the involved kernel threads are spinning) would yield some useful data to kick off a potentially productive debate about what could actually be going on. On Fedora, I think `sudo dnf install perf` should set up everything you need to get you going; then, look at `sudo perf top -F 999` for a while, and let the list know what you see. (You can use your terminal emulator's flow control features, available via Ctrl+S to enable and Ctrl+Q to disable, to "freeze" the output to make it easier to copy it into a clipboard buffer.) -- with best regards: - Johannes Truschnigg ( johannes@xxxxxxxxxxxxxxx ) www: https://johannes.truschnigg.info/ phone: +436502133337 xmpp: johannes@xxxxxxxxxxxxxxx
Attachment:
signature.asc
Description: PGP signature
