[PATCH 12/32] xfs: verify AGF blocks as they are read from disk

[Dave Chinner]

On Mon, Nov 12, 2012 at 05:09:00PM -0800, Phil White wrote:
> Dave, you botched a copy & paste here:
> 
> On Mon, Nov 12, 2012 at 10:54:04PM +1100, Dave Chinner wrote:
> > +	agf_ok = agf->agf_magicnum == cpu_to_be32(XFS_AGF_MAGIC) &&
> > +		XFS_AGF_GOOD_VERSION(be32_to_cpu(agf->agf_versionnum)) &&
> > +		be32_to_cpu(agf->agf_freeblks) <= be32_to_cpu(agf->agf_length) &&
> > +		be32_to_cpu(agf->agf_flfirst) < XFS_AGFL_SIZE(mp) &&
> > +		be32_to_cpu(agf->agf_fllast) < XFS_AGFL_SIZE(mp) &&
> > +		be32_to_cpu(agf->agf_flcount) <= XFS_AGFL_SIZE(mp) &&
> > +		be32_to_cpu(agf->agf_flcount) <= XFS_AGFL_SIZE(mp);
> 
> See: 
> > -	agf_ok =
> > -		agf->agf_magicnum == cpu_to_be32(XFS_AGF_MAGIC) &&
> > -		XFS_AGF_GOOD_VERSION(be32_to_cpu(agf->agf_versionnum)) &&
> > -		be32_to_cpu(agf->agf_freeblks) <= be32_to_cpu(agf->agf_length) &&
> > -		be32_to_cpu(agf->agf_flfirst) < XFS_AGFL_SIZE(mp) &&
> > -		be32_to_cpu(agf->agf_fllast) < XFS_AGFL_SIZE(mp) &&
> > -		be32_to_cpu(agf->agf_flcount) <= XFS_AGFL_SIZE(mp) &&
> > -		be32_to_cpu(agf->agf_seqno) == agno;

Good catch. :)

The agno is still checked in the new code, so it's just a double
check of the agf_flcount. i.e. no actual bug. I'll resend an updated
patch.

Cheers,

Dave.
-- 
Dave Chinner
david at fromorbit.com