On 2019-01-08 6:49 a.m., Jeff Moyer wrote: >> The traceback seems to indicate the problem is on the bip_get_seed() >> line in t10_pi_complete(). Which suggests that bio_integrity() is >> returning NULL. > > Does your hardware actually support protection information? As far as I know, I do not. If I add a printk to t10_pi_complete(), it doesn't print on a successful boot and does print on a broken boot. So something is causing it to be called erroneously. Logan
