https://bugzilla.kernel.org/show_bug.cgi?id=216230 --- Comment #15 from madcatx@xxxxxxxx --- I made a small discovery today. After adding pinctrl_amd module to my initramfs I haven't been able to reproduce the problem so far. That may not be indicative of anything because the problem does not appear every time. A bit more interesting is my current output of /proc/interrupts. For IRQ9 it currently shows 9: 0 11461 0 0 0 0 0 0 0 0 0 0 0 0 0 0 IR-IO-APIC 9-fasteoi acpi, pinctrl_amd Notice that the number of recorded interrupts is around 11000. Before I added pinctrl_amd to initramfs, there were two distinct cases. Either over 100000 interrupts with the kernel complaining about IRQ9 not being serviced or about 90 interrupts and no apparent issues. All of these numbers are from /proc/interrupts dump done right after logging into DE after a reboot. I was able to reproduce these figures multiple times. This suggests that some HW starts to blast IRQs and doesn't stop until it gets serviced. Knowing this, is there anything else you'd like me to try besides your suggestions above? My CONFIG_PINCTRL_AMD is set to CONFIG_PINCTRL_AMD=m in my kernel config and the respective module is loaded. -- You may reply to this email to add a comment. You are receiving this mail because: You are watching the assignee of the bug.
