linux-toradex.git/fs/xfs/xfs_fsops.c, branch v3.10.60

xfs: growfs overruns AGFL buffer on V4 filesystems

2013-12-20T15:45:08+00:00

commit f94c44573e7c22860e2c3dfe349c45f72ba35ad3 upstream.

This loop in xfs_growfs_data_private() is incorrect for V4
superblocks filesystems:

		for (bucket = 0; bucket < XFS_AGFL_SIZE(mp); bucket++)
			agfl->agfl_bno[bucket] = cpu_to_be32(NULLAGBLOCK);

For V4 filesystems, we don't have a agfl header structure, and so
XFS_AGFL_SIZE() returns an entire sector's worth of entries, which
we then index from an offset into the sector. Hence: buffer overrun.

This problem was introduced in 3.10 by commit 77c95bba ("xfs: add
CRC checks to the AGFL") which changed the AGFL structure but failed
to update the growfs code to handle the different structures.

Fix it by using the correct offset into the buffer for both V4 and
V5 filesystems.

Signed-off-by: Dave Chinner 
Reviewed-by: Jie Liu 
Signed-off-by: Ben Myers 
Signed-off-by: Greg Kroah-Hartman

xfs: add fsgeom flag for v5 superblock support.

2013-05-30T22:19:45+00:00

Currently userspace has no way of determining that a filesystem is
CRC enabled. Add a flag to the XFS_IOC_FSGEOMETRY ioctl output to
indicate that the filesystem has v5 superblock support enabled.
This will allow xfs_info to correctly report the state of the
filesystem.

Signed-off-by: Dave Chinner 
Reviewed-by: Eric Sandeen 
Reviewed-by: Brian Foster 
Signed-off-by: Ben Myers 

(cherry picked from commit 74137fff067961c9aca1e14d073805c3de8549bd)

xfs: add CRC checks to the AGI

2013-04-21T19:57:43+00:00

Same set of changes made to the AGF need to be made to the AGI.
This patch has a similar history to the AGF, hence a similar
sign-off chain.

Signed-off-by: Dave Chinner 
Signed-off-by: Christoph Hellwig 
Signed-off-by: Dave Chinner 
Reviewed-by: Ben Myers 
Signed-off-by: Ben Myers

xfs: add CRC checks to the AGFL

2013-04-21T19:55:34+00:00

Add CRC checks, location information and a magic number to the AGFL.
Previously the AGFL was just a block containing nothing but the
free block pointers.  The new AGFL has a real header with the usual
boilerplate instead, so that we can verify it's not corrupted and
written into the right place.

[dchinner@redhat.com] Added LSN field, reworked significantly to fit
into new verifier structure and growfs structure, enabled full
verifier functionality now there is a header to verify and we can
guarantee an initialised AGFL.

Signed-off-by: Christoph Hellwig 
Signed-off-by: Dave Chinner 
Reviewed-by: Ben Myers 
Signed-off-by: Ben Myers

xfs: add CRC checks to the AGF

2013-04-21T19:54:46+00:00

The AGF already has some self identifying fields (e.g. the sequence
number) so we only need to add the uuid to it to identify the
filesystem it belongs to. The location is fixed based on the
sequence number, so there's no need to add a block number, either.

Hence the only additional fields are the CRC and LSN fields. These
are unlogged, so place some space between the end of the logged
fields and them so that future expansion of the AGF for logged
fields can be placed adjacent to the existing logged fields and
hence not complicate the field-derived range based logging we
currently have.

Based originally on a patch from myself, modified further by
Christoph Hellwig and then modified again to fit into the
verifier structure with additional fields by myself. The multiple
signed-off-by tags indicate the age and history of this patch.

Signed-off-by: Dave Chinner 
Signed-off-by: Christoph Hellwig 
Signed-off-by: Dave Chinner 
Reviewed-by: Ben Myers 
Signed-off-by: Ben Myers

xfs: add support for large btree blocks

2013-04-21T19:53:46+00:00

Add support for larger btree blocks that contains a CRC32C checksum,
a filesystem uuid and block number for detecting filesystem
consistency and out of place writes.

[dchinner@redhat.com] Also include an owner field to allow reverse
mappings to be implemented for improved repairability and a LSN
field to so that log recovery can easily determine the last
modification that made it to disk for each buffer.

[dchinner@redhat.com] Add buffer log format flags to indicate the
type of buffer to recovery so that we don't have to do blind magic
number tests to determine what the buffer is.

[dchinner@redhat.com] Modified to fit into the verifier structure.

Signed-off-by: Christoph Hellwig 
Signed-off-by: Dave Chinner 
Reviewed-by: Ben Myers 
Signed-off-by: Ben Myers

xfs: make use of XFS_SB_LOG_RES() at xfs_fs_log_dummy()

2013-02-01T20:55:59+00:00

Make use of XFS_SB_LOG_RES() at xfs_fs_log_dummy().

Signed-off-by: Jie Liu 
CC: Dave Chinner 
Reviewed-by: Mark Tinguely 
Signed-off-by: Ben Myers

xfs: convert buffer verifiers to an ops structure.

2012-11-16T03:35:12+00:00

To separate the verifiers from iodone functions and associate read
and write verifiers at the same time, introduce a buffer verifier
operations structure to the xfs_buf.

This avoids the need for assigning the write verifier, clearing the
iodone function and re-running ioend processing in the read
verifier, and gets rid of the nasty "b_pre_io" name for the write
verifier function pointer. If we ever need to, it will also be
easier to add further content specific callbacks to a buffer with an
ops structure in place.

We also avoid needing to export verifier functions, instead we
can simply export the ops structures for those that are needed
outside the function they are defined in.

This patch also fixes a directory block readahead verifier issue
it exposed.

This patch also adds ops callbacks to the inode/alloc btree blocks
initialised by growfs. These will need more work before they will
work with CRCs.

Signed-off-by: Dave Chinner 
Reviewed-by: Phil White 
Signed-off-by: Ben Myers

xfs: connect up write verifiers to new buffers

2012-11-16T03:35:09+00:00

Metadata buffers that are read from disk have write verifiers
already attached to them, but newly allocated buffers do not. Add
appropriate write verifiers to all new metadata buffers.

Signed-off-by: Dave Chinner 
Reviewed-by: Ben Myers 
Signed-off-by: Ben Myers

xfs: verify superblocks as they are read from disk

2012-11-16T03:34:07+00:00

Add a superblock verify callback function and pass it into the
buffer read functions. Remove the now redundant verification code
that is currently in use.

Adding verification shows that secondary superblocks never have
their "sb_inprogress" flag cleared by mkfs.xfs, so when validating
the secondary superblocks during a grow operation we have to avoid
checking this field. Even if we fix mkfs, we will still have to
ignore this field for verification purposes unless a version of mkfs
that does not have this bug was used.

Signed-off-by: Dave Chinner 
Reviewed-by: Phil White 
Signed-off-by: Ben Myers