Greg Steuck <gne...@openbsd.org> writes: > Since I don't understand it, I decided to look in some well-lit > places.
I forced a trace in the case when the system fails with that protection fault. It's mostly the same as the more complete trace I noted in the other case: kernel: protection fault trap, code=0 Stopped at sd_buf_done+0x18: movq 0x38(%rax),%12 sd_buf_done(fffffd880daccbd0) at sd_buf_done+0x18 scsi_done(fffffd880daccbd0) at scsi_done+0x31 nume_q_complete(ffff800000255000, ffff800001184a80) at nume_q_complete+0x134 nume_intr(ffff800000255000) at nume_intr+0x61 intr_handler(ffff800049e24ab0, ffff800000254200) at intr_handler+0x91 Xintr_ioapic_edge28_untramp() at Xintr_ioapic_edge28_untramp+0x18f acpicpu_idle() at acpicpu_idle+0x131 sched_idle(ffffffff82755ff0) at sched_idle+0x298 end trace frame: 0x0, count: 7 OCRd from this: https://photos.app.goo.gl/axYRWYWfP39A7FaY9
