xfs: fix _xfs_buf_find oops on blocks beyond the filesystem end
authorDave Chinner <dchinner@redhat.com>
Mon, 21 Jan 2013 12:53:52 +0000 (23:53 +1100)
committerBen Myers <bpm@sgi.com>
Mon, 28 Jan 2013 18:49:21 +0000 (12:49 -0600)
commiteb178619f930fa2ba2348de332a1ff1c66a31424
treeb245ff61cafffb860e682c4194dbfd882589304f
parentd26978dd866dbb3b3a9690f3655a5e735055de89
xfs: fix _xfs_buf_find oops on blocks beyond the filesystem end

When _xfs_buf_find is passed an out of range address, it will fail
to find a relevant struct xfs_perag and oops with a null
dereference. This can happen when trying to walk a filesystem with a
metadata inode that has a partially corrupted extent map (i.e. the
block number returned is corrupt, but is otherwise intact) and we
try to read from the corrupted block address.

In this case, just fail the lookup. If it is readahead being issued,
it will simply not be done, but if it is real read that fails we
will get an error being reported.  Ideally this case should result
in an EFSCORRUPTED error being reported, but we cannot return an
error through xfs_buf_read() or xfs_buf_get() so this lookup failure
may result in ENOMEM or EIO errors being reported instead.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Ben Myers <bpm@sgi.com>
Signed-off-by: Ben Myers <bpm@sgi.com>
fs/xfs/xfs_buf.c
This page took 0.025431 seconds and 5 git commands to generate.