xfs: add support for large btree blocks
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 <hch@lst.de>
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Ben Myers <bpm@sgi.com>
Signed-off-by: Ben Myers <bpm@sgi.com>
diff --git a/fs/xfs/xfs_bmap_btree.c b/fs/xfs/xfs_bmap_btree.c
index 061b45c..3a86c3f 100644
--- a/fs/xfs/xfs_bmap_btree.c
+++ b/fs/xfs/xfs_bmap_btree.c
@@ -37,6 +37,7 @@
#include "xfs_error.h"
#include "xfs_quota.h"
#include "xfs_trace.h"
+#include "xfs_cksum.h"
/*
* Determine the extent state.
@@ -59,24 +60,31 @@
*/
void
xfs_bmdr_to_bmbt(
- struct xfs_mount *mp,
+ struct xfs_inode *ip,
xfs_bmdr_block_t *dblock,
int dblocklen,
struct xfs_btree_block *rblock,
int rblocklen)
{
+ struct xfs_mount *mp = ip->i_mount;
int dmxr;
xfs_bmbt_key_t *fkp;
__be64 *fpp;
xfs_bmbt_key_t *tkp;
__be64 *tpp;
- rblock->bb_magic = cpu_to_be32(XFS_BMAP_MAGIC);
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ xfs_btree_init_block_int(mp, rblock, XFS_BUF_DADDR_NULL,
+ XFS_BMAP_CRC_MAGIC, 0, 0, ip->i_ino,
+ XFS_BTREE_LONG_PTRS | XFS_BTREE_CRC_BLOCKS);
+ else
+ xfs_btree_init_block_int(mp, rblock, XFS_BUF_DADDR_NULL,
+ XFS_BMAP_MAGIC, 0, 0, ip->i_ino,
+ XFS_BTREE_LONG_PTRS);
+
rblock->bb_level = dblock->bb_level;
ASSERT(be16_to_cpu(rblock->bb_level) > 0);
rblock->bb_numrecs = dblock->bb_numrecs;
- rblock->bb_u.l.bb_leftsib = cpu_to_be64(NULLDFSBNO);
- rblock->bb_u.l.bb_rightsib = cpu_to_be64(NULLDFSBNO);
dmxr = xfs_bmdr_maxrecs(mp, dblocklen, 0);
fkp = XFS_BMDR_KEY_ADDR(dblock, 1);
tkp = XFS_BMBT_KEY_ADDR(mp, rblock, 1);
@@ -424,7 +432,13 @@
xfs_bmbt_key_t *tkp;
__be64 *tpp;
- ASSERT(rblock->bb_magic == cpu_to_be32(XFS_BMAP_MAGIC));
+ if (xfs_sb_version_hascrc(&mp->m_sb)) {
+ ASSERT(rblock->bb_magic == cpu_to_be32(XFS_BMAP_CRC_MAGIC));
+ ASSERT(uuid_equal(&rblock->bb_u.l.bb_uuid, &mp->m_sb.sb_uuid));
+ ASSERT(rblock->bb_u.l.bb_blkno ==
+ cpu_to_be64(XFS_BUF_DADDR_NULL));
+ } else
+ ASSERT(rblock->bb_magic == cpu_to_be32(XFS_BMAP_MAGIC));
ASSERT(rblock->bb_u.l.bb_leftsib == cpu_to_be64(NULLDFSBNO));
ASSERT(rblock->bb_u.l.bb_rightsib == cpu_to_be64(NULLDFSBNO));
ASSERT(rblock->bb_level != 0);
@@ -708,59 +722,89 @@
cur->bc_rec.b.br_startoff;
}
-static void
+static int
xfs_bmbt_verify(
struct xfs_buf *bp)
{
struct xfs_mount *mp = bp->b_target->bt_mount;
struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp);
unsigned int level;
- int lblock_ok; /* block passes checks */
- /* magic number and level verification.
+ switch (block->bb_magic) {
+ case cpu_to_be32(XFS_BMAP_CRC_MAGIC):
+ if (!xfs_sb_version_hascrc(&mp->m_sb))
+ return false;
+ if (!uuid_equal(&block->bb_u.l.bb_uuid, &mp->m_sb.sb_uuid))
+ return false;
+ if (be64_to_cpu(block->bb_u.l.bb_blkno) != bp->b_bn)
+ return false;
+ /*
+ * XXX: need a better way of verifying the owner here. Right now
+ * just make sure there has been one set.
+ */
+ if (be64_to_cpu(block->bb_u.l.bb_owner) == 0)
+ return false;
+ /* fall through */
+ case cpu_to_be32(XFS_BMAP_MAGIC):
+ break;
+ default:
+ return false;
+ }
+
+ /*
+ * numrecs and level verification.
*
- * We don't know waht fork we belong to, so just verify that the level
+ * We don't know what fork we belong to, so just verify that the level
* is less than the maximum of the two. Later checks will be more
* precise.
*/
level = be16_to_cpu(block->bb_level);
- lblock_ok = block->bb_magic == cpu_to_be32(XFS_BMAP_MAGIC) &&
- level < max(mp->m_bm_maxlevels[0], mp->m_bm_maxlevels[1]);
-
- /* numrecs verification */
- lblock_ok = lblock_ok &&
- be16_to_cpu(block->bb_numrecs) <= mp->m_bmap_dmxr[level != 0];
+ if (level > max(mp->m_bm_maxlevels[0], mp->m_bm_maxlevels[1]))
+ return false;
+ if (be16_to_cpu(block->bb_numrecs) > mp->m_bmap_dmxr[level != 0])
+ return false;
/* sibling pointer verification */
- lblock_ok = lblock_ok &&
- block->bb_u.l.bb_leftsib &&
- (block->bb_u.l.bb_leftsib == cpu_to_be64(NULLDFSBNO) ||
- XFS_FSB_SANITY_CHECK(mp,
- be64_to_cpu(block->bb_u.l.bb_leftsib))) &&
- block->bb_u.l.bb_rightsib &&
- (block->bb_u.l.bb_rightsib == cpu_to_be64(NULLDFSBNO) ||
- XFS_FSB_SANITY_CHECK(mp,
- be64_to_cpu(block->bb_u.l.bb_rightsib)));
+ if (!block->bb_u.l.bb_leftsib ||
+ (block->bb_u.l.bb_leftsib != cpu_to_be64(NULLDFSBNO) &&
+ !XFS_FSB_SANITY_CHECK(mp, be64_to_cpu(block->bb_u.l.bb_leftsib))))
+ return false;
+ if (!block->bb_u.l.bb_rightsib ||
+ (block->bb_u.l.bb_rightsib != cpu_to_be64(NULLDFSBNO) &&
+ !XFS_FSB_SANITY_CHECK(mp, be64_to_cpu(block->bb_u.l.bb_rightsib))))
+ return false;
- if (!lblock_ok) {
- trace_xfs_btree_corrupt(bp, _RET_IP_);
- XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, block);
- xfs_buf_ioerror(bp, EFSCORRUPTED);
- }
+ return true;
+
}
static void
xfs_bmbt_read_verify(
struct xfs_buf *bp)
{
- xfs_bmbt_verify(bp);
+ if (!(xfs_btree_lblock_verify_crc(bp) &&
+ xfs_bmbt_verify(bp))) {
+ trace_xfs_btree_corrupt(bp, _RET_IP_);
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
+ bp->b_target->bt_mount, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ }
+
}
static void
xfs_bmbt_write_verify(
struct xfs_buf *bp)
{
- xfs_bmbt_verify(bp);
+ if (!xfs_bmbt_verify(bp)) {
+ xfs_warn(bp->b_target->bt_mount, "bmbt daddr 0x%llx failed", bp->b_bn);
+ trace_xfs_btree_corrupt(bp, _RET_IP_);
+ XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
+ bp->b_target->bt_mount, bp->b_addr);
+ xfs_buf_ioerror(bp, EFSCORRUPTED);
+ return;
+ }
+ xfs_btree_lblock_calc_crc(bp);
}
const struct xfs_buf_ops xfs_bmbt_buf_ops = {
@@ -838,6 +882,8 @@
cur->bc_ops = &xfs_bmbt_ops;
cur->bc_flags = XFS_BTREE_LONG_PTRS | XFS_BTREE_ROOT_IN_INODE;
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
cur->bc_private.b.forksize = XFS_IFORK_SIZE(ip, whichfork);
cur->bc_private.b.ip = ip;