On Mon, Dec 14, 2015 at 02:04:56PM +0000, Michael van Elst wrote: > >4294967264 in 0xffffffe0 with suggests some int32 underrun. > The field is 32bit in FFS1 and 64bit in FFS2.
It is not easy to understand what version we actually use: file system: /dev/rraid0e format FFSv1 endian little-endian magic 11954 time Mon Dec 14 15:27:15 2015 superblock location 8192 id [ 4f48391c 5d30adc9 ] cylgrp dynamic inodes 4.4BSD sblock FFSv2 fslevel 4 What is fslevel 4? Does it have something to do with newfs -O ? The man page only document up to 2. Anyway, if the panic is caused by an underrun, we should hit the same problem with any int length: going below zero will always result into a huge value either witn a 32 bit or a 64 bit int. -- Emmanuel Dreyfus m...@netbsd.org
