On Mon Sep 24 2007 at 18:20:03 +0000, Antti Kantee wrote:
>  This is because the Solaris superblock format is kind enough to have a
>  few fields in different places.  You can mount and use your file system
>  image with this change:
>  
>  Index: fs.h
>  ===================================================================
>  RCS file: /cvsroot/src/sys/ufs/ffs/fs.h,v
>  retrieving revision 1.47
>  diff -p -u -r1.47 fs.h
>  --- fs.h	24 Sep 2007 16:20:50 -0000	1.47
>  +++ fs.h	24 Sep 2007 18:13:45 -0000
>  @@ -348,9 +348,9 @@ struct fs {
>   	int32_t	 fs_maxsymlinklen;	/* max length of an internal symlink */
>   	int32_t	 fs_old_inodefmt;	/* format of on-disk inodes */
>   	u_int64_t fs_maxfilesize;	/* maximum representable file size */
>  +	int32_t	 fs_state;		/* validate fs_clean field (UNUSED) */
>   	int64_t	 fs_qbmask;		/* ~fs_bmask for use with 64-bit size */
>   	int64_t	 fs_qfmask;		/* ~fs_fmask for use with 64-bit size */
>  -	int32_t	 fs_state;		/* validate fs_clean field (UNUSED) */
>   	int32_t	 fs_old_postblformat;	/* format of positional layout tables */
>   	int32_t	 fs_old_nrpos;		/* number of rotational positions */
>   	int32_t  fs_spare5[2];		/* old fs_postbloff */
>  
>  For as how to detect and fix this, I don't know.

Actually, coming to think of it, why don't we just always form fs_qbmask
from fs_bmask (same for fmask) at runtime?  Is there any case where they
are not expected to be consistent?  Why's qbmask on-disk anyway?

-- 
Antti Kantee <pooka@iki.fi>                     Of course he runs NetBSD
http://www.iki.fi/pooka/                          http://www.NetBSD.org/
    "la qualité la plus indispensable du cuisinier est l'exactitude"