Commit 035e00ac authored by Darrick J. Wong's avatar Darrick J. Wong Committed by Dave Chinner

xfs: define the on-disk rmap btree format

Originally-From: Dave Chinner <dchinner@redhat.com>

Now we have all the surrounding call infrastructure in place, we can
start filling out the rmap btree implementation. Start with the
on-disk btree format; add everything needed to read, write and
manipulate rmap btree blocks. This prepares the way for adding the
btree operations implementation.

[darrick: record owner and offset info in rmap btree]
[darrick: fork, bmbt and unwritten state in rmap btree]
[darrick: flags are a separate field in xfs_rmap_irec]
[darrick: calculate maxlevels separately]
[darrick: move the 'unwritten' bit into unused parts of rm_offset]
Signed-off-by: default avatarDave Chinner <dchinner@redhat.com>
Signed-off-by: default avatarDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: default avatarDave Chinner <dchinner@redhat.com>
Reviewed-by: default avatarBrian Foster <bfoster@redhat.com>
Signed-off-by: default avatarDave Chinner <david@fromorbit.com>
parent 673930c3
......@@ -53,6 +53,7 @@ xfs-y += $(addprefix libxfs/, \
xfs_inode_buf.o \
xfs_log_rlimit.o \
xfs_rmap.o \
xfs_rmap_btree.o \
xfs_sb.o \
xfs_symlink_remote.o \
xfs_trans_resv.o \
......
......@@ -1213,6 +1213,9 @@ xfs_btree_set_refs(
case XFS_BTNUM_BMAP:
xfs_buf_set_ref(bp, XFS_BMAP_BTREE_REF);
break;
case XFS_BTNUM_RMAP:
xfs_buf_set_ref(bp, XFS_RMAP_BTREE_REF);
break;
default:
ASSERT(0);
}
......
......@@ -38,10 +38,11 @@ union xfs_btree_ptr {
};
union xfs_btree_key {
xfs_bmbt_key_t bmbt;
struct xfs_bmbt_key bmbt;
xfs_bmdr_key_t bmbr; /* bmbt root block */
xfs_alloc_key_t alloc;
xfs_inobt_key_t inobt;
struct xfs_inobt_key inobt;
struct xfs_rmap_key rmap;
};
/*
......@@ -56,13 +57,18 @@ union xfs_btree_bigkey {
xfs_bmdr_key_t bmbr; /* bmbt root block */
xfs_alloc_key_t alloc;
struct xfs_inobt_key inobt;
struct {
struct xfs_rmap_key rmap;
struct xfs_rmap_key rmap_hi;
};
};
union xfs_btree_rec {
xfs_bmbt_rec_t bmbt;
struct xfs_bmbt_rec bmbt;
xfs_bmdr_rec_t bmbr; /* bmbt root block */
xfs_alloc_rec_t alloc;
xfs_inobt_rec_t inobt;
struct xfs_alloc_rec alloc;
struct xfs_inobt_rec inobt;
struct xfs_rmap_rec rmap;
};
/*
......
......@@ -1347,6 +1347,79 @@ struct xfs_owner_info {
#define XFS_RMAP_OWN_INODES (-7ULL) /* Inode chunk */
#define XFS_RMAP_OWN_MIN (-8ULL) /* guard */
#define XFS_RMAP_NON_INODE_OWNER(owner) (!!((owner) & (1ULL << 63)))
/*
* Data record structure
*/
struct xfs_rmap_rec {
__be32 rm_startblock; /* extent start block */
__be32 rm_blockcount; /* extent length */
__be64 rm_owner; /* extent owner */
__be64 rm_offset; /* offset within the owner */
};
/*
* rmap btree record
* rm_offset:63 is the attribute fork flag
* rm_offset:62 is the bmbt block flag
* rm_offset:61 is the unwritten extent flag (same as l0:63 in bmbt)
* rm_offset:54-60 aren't used and should be zero
* rm_offset:0-53 is the block offset within the inode
*/
#define XFS_RMAP_OFF_ATTR_FORK ((__uint64_t)1ULL << 63)
#define XFS_RMAP_OFF_BMBT_BLOCK ((__uint64_t)1ULL << 62)
#define XFS_RMAP_OFF_UNWRITTEN ((__uint64_t)1ULL << 61)
#define XFS_RMAP_LEN_MAX ((__uint32_t)~0U)
#define XFS_RMAP_OFF_FLAGS (XFS_RMAP_OFF_ATTR_FORK | \
XFS_RMAP_OFF_BMBT_BLOCK | \
XFS_RMAP_OFF_UNWRITTEN)
#define XFS_RMAP_OFF_MASK ((__uint64_t)0x3FFFFFFFFFFFFFULL)
#define XFS_RMAP_OFF(off) ((off) & XFS_RMAP_OFF_MASK)
#define XFS_RMAP_IS_BMBT_BLOCK(off) (!!((off) & XFS_RMAP_OFF_BMBT_BLOCK))
#define XFS_RMAP_IS_ATTR_FORK(off) (!!((off) & XFS_RMAP_OFF_ATTR_FORK))
#define XFS_RMAP_IS_UNWRITTEN(len) (!!((off) & XFS_RMAP_OFF_UNWRITTEN))
#define RMAPBT_STARTBLOCK_BITLEN 32
#define RMAPBT_BLOCKCOUNT_BITLEN 32
#define RMAPBT_OWNER_BITLEN 64
#define RMAPBT_ATTRFLAG_BITLEN 1
#define RMAPBT_BMBTFLAG_BITLEN 1
#define RMAPBT_EXNTFLAG_BITLEN 1
#define RMAPBT_UNUSED_OFFSET_BITLEN 7
#define RMAPBT_OFFSET_BITLEN 54
#define XFS_RMAP_ATTR_FORK (1 << 0)
#define XFS_RMAP_BMBT_BLOCK (1 << 1)
#define XFS_RMAP_UNWRITTEN (1 << 2)
#define XFS_RMAP_KEY_FLAGS (XFS_RMAP_ATTR_FORK | \
XFS_RMAP_BMBT_BLOCK)
#define XFS_RMAP_REC_FLAGS (XFS_RMAP_UNWRITTEN)
struct xfs_rmap_irec {
xfs_agblock_t rm_startblock; /* extent start block */
xfs_extlen_t rm_blockcount; /* extent length */
__uint64_t rm_owner; /* extent owner */
__uint64_t rm_offset; /* offset within the owner */
unsigned int rm_flags; /* state flags */
};
/*
* Key structure
*
* We don't use the length for lookups
*/
struct xfs_rmap_key {
__be32 rm_startblock; /* extent start block */
__be64 rm_owner; /* extent owner */
__be64 rm_offset; /* offset within the owner */
} __attribute__((packed));
/* btree pointer type */
typedef __be32 xfs_rmap_ptr_t;
#define XFS_RMAP_BLOCK(mp) \
(xfs_sb_version_hasfinobt(&((mp)->m_sb)) ? \
XFS_FIBT_BLOCK(mp) + 1 : \
......
......@@ -68,6 +68,73 @@ xfs_rmap_skip_owner_update(
struct xfs_buf;
static inline __u64
xfs_rmap_irec_offset_pack(
const struct xfs_rmap_irec *irec)
{
__u64 x;
x = XFS_RMAP_OFF(irec->rm_offset);
if (irec->rm_flags & XFS_RMAP_ATTR_FORK)
x |= XFS_RMAP_OFF_ATTR_FORK;
if (irec->rm_flags & XFS_RMAP_BMBT_BLOCK)
x |= XFS_RMAP_OFF_BMBT_BLOCK;
if (irec->rm_flags & XFS_RMAP_UNWRITTEN)
x |= XFS_RMAP_OFF_UNWRITTEN;
return x;
}
static inline int
xfs_rmap_irec_offset_unpack(
__u64 offset,
struct xfs_rmap_irec *irec)
{
if (offset & ~(XFS_RMAP_OFF_MASK | XFS_RMAP_OFF_FLAGS))
return -EFSCORRUPTED;
irec->rm_offset = XFS_RMAP_OFF(offset);
if (offset & XFS_RMAP_OFF_ATTR_FORK)
irec->rm_flags |= XFS_RMAP_ATTR_FORK;
if (offset & XFS_RMAP_OFF_BMBT_BLOCK)
irec->rm_flags |= XFS_RMAP_BMBT_BLOCK;
if (offset & XFS_RMAP_OFF_UNWRITTEN)
irec->rm_flags |= XFS_RMAP_UNWRITTEN;
return 0;
}
static inline void
xfs_owner_info_unpack(
struct xfs_owner_info *oinfo,
uint64_t *owner,
uint64_t *offset,
unsigned int *flags)
{
unsigned int r = 0;
*owner = oinfo->oi_owner;
*offset = oinfo->oi_offset;
if (oinfo->oi_flags & XFS_OWNER_INFO_ATTR_FORK)
r |= XFS_RMAP_ATTR_FORK;
if (oinfo->oi_flags & XFS_OWNER_INFO_BMBT_BLOCK)
r |= XFS_RMAP_BMBT_BLOCK;
*flags = r;
}
static inline void
xfs_owner_info_pack(
struct xfs_owner_info *oinfo,
uint64_t owner,
uint64_t offset,
unsigned int flags)
{
oinfo->oi_owner = owner;
oinfo->oi_offset = XFS_RMAP_OFF(offset);
oinfo->oi_flags = 0;
if (flags & XFS_RMAP_ATTR_FORK)
oinfo->oi_flags |= XFS_OWNER_INFO_ATTR_FORK;
if (flags & XFS_RMAP_BMBT_BLOCK)
oinfo->oi_flags |= XFS_OWNER_INFO_BMBT_BLOCK;
}
int xfs_rmap_alloc(struct xfs_trans *tp, struct xfs_buf *agbp,
xfs_agnumber_t agno, xfs_agblock_t bno, xfs_extlen_t len,
struct xfs_owner_info *oinfo);
......
/*
* Copyright (c) 2014 Red Hat, Inc.
* All Rights Reserved.
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License as
* published by the Free Software Foundation.
*
* This program is distributed in the hope that it would be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write the Free Software Foundation,
* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
*/
#include "xfs.h"
#include "xfs_fs.h"
#include "xfs_shared.h"
#include "xfs_format.h"
#include "xfs_log_format.h"
#include "xfs_trans_resv.h"
#include "xfs_bit.h"
#include "xfs_sb.h"
#include "xfs_mount.h"
#include "xfs_defer.h"
#include "xfs_inode.h"
#include "xfs_trans.h"
#include "xfs_alloc.h"
#include "xfs_btree.h"
#include "xfs_rmap_btree.h"
#include "xfs_trace.h"
#include "xfs_cksum.h"
#include "xfs_error.h"
#include "xfs_extent_busy.h"
static struct xfs_btree_cur *
xfs_rmapbt_dup_cursor(
struct xfs_btree_cur *cur)
{
return xfs_rmapbt_init_cursor(cur->bc_mp, cur->bc_tp,
cur->bc_private.a.agbp, cur->bc_private.a.agno);
}
static bool
xfs_rmapbt_verify(
struct xfs_buf *bp)
{
struct xfs_mount *mp = bp->b_target->bt_mount;
struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp);
struct xfs_perag *pag = bp->b_pag;
unsigned int level;
/*
* magic number and level verification
*
* During growfs operations, we can't verify the exact level or owner as
* the perag is not fully initialised and hence not attached to the
* buffer. In this case, check against the maximum tree depth.
*
* Similarly, during log recovery we will have a perag structure
* attached, but the agf information will not yet have been initialised
* from the on disk AGF. Again, we can only check against maximum limits
* in this case.
*/
if (block->bb_magic != cpu_to_be32(XFS_RMAP_CRC_MAGIC))
return false;
if (!xfs_sb_version_hasrmapbt(&mp->m_sb))
return false;
if (!xfs_btree_sblock_v5hdr_verify(bp))
return false;
level = be16_to_cpu(block->bb_level);
if (pag && pag->pagf_init) {
if (level >= pag->pagf_levels[XFS_BTNUM_RMAPi])
return false;
} else if (level >= mp->m_rmap_maxlevels)
return false;
return xfs_btree_sblock_verify(bp, mp->m_rmap_mxr[level != 0]);
}
static void
xfs_rmapbt_read_verify(
struct xfs_buf *bp)
{
if (!xfs_btree_sblock_verify_crc(bp))
xfs_buf_ioerror(bp, -EFSBADCRC);
else if (!xfs_rmapbt_verify(bp))
xfs_buf_ioerror(bp, -EFSCORRUPTED);
if (bp->b_error) {
trace_xfs_btree_corrupt(bp, _RET_IP_);
xfs_verifier_error(bp);
}
}
static void
xfs_rmapbt_write_verify(
struct xfs_buf *bp)
{
if (!xfs_rmapbt_verify(bp)) {
trace_xfs_btree_corrupt(bp, _RET_IP_);
xfs_buf_ioerror(bp, -EFSCORRUPTED);
xfs_verifier_error(bp);
return;
}
xfs_btree_sblock_calc_crc(bp);
}
const struct xfs_buf_ops xfs_rmapbt_buf_ops = {
.name = "xfs_rmapbt",
.verify_read = xfs_rmapbt_read_verify,
.verify_write = xfs_rmapbt_write_verify,
};
static const struct xfs_btree_ops xfs_rmapbt_ops = {
.rec_len = sizeof(struct xfs_rmap_rec),
.key_len = 2 * sizeof(struct xfs_rmap_key),
.dup_cursor = xfs_rmapbt_dup_cursor,
.buf_ops = &xfs_rmapbt_buf_ops,
.get_leaf_keys = xfs_btree_get_leaf_keys_overlapped,
.get_node_keys = xfs_btree_get_node_keys_overlapped,
.update_keys = xfs_btree_update_keys_overlapped,
};
/*
* Allocate a new allocation btree cursor.
*/
struct xfs_btree_cur *
xfs_rmapbt_init_cursor(
struct xfs_mount *mp,
struct xfs_trans *tp,
struct xfs_buf *agbp,
xfs_agnumber_t agno)
{
struct xfs_agf *agf = XFS_BUF_TO_AGF(agbp);
struct xfs_btree_cur *cur;
cur = kmem_zone_zalloc(xfs_btree_cur_zone, KM_NOFS);
cur->bc_tp = tp;
cur->bc_mp = mp;
cur->bc_btnum = XFS_BTNUM_RMAP;
cur->bc_flags = XFS_BTREE_CRC_BLOCKS;
cur->bc_blocklog = mp->m_sb.sb_blocklog;
cur->bc_ops = &xfs_rmapbt_ops;
cur->bc_nlevels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAP]);
cur->bc_private.a.agbp = agbp;
cur->bc_private.a.agno = agno;
return cur;
}
/*
* Calculate number of records in an rmap btree block.
*/
int
xfs_rmapbt_maxrecs(
struct xfs_mount *mp,
int blocklen,
int leaf)
{
blocklen -= XFS_RMAP_BLOCK_LEN;
if (leaf)
return blocklen / sizeof(struct xfs_rmap_rec);
return blocklen /
(sizeof(struct xfs_rmap_key) + sizeof(xfs_rmap_ptr_t));
}
/* Compute the maximum height of an rmap btree. */
void
xfs_rmapbt_compute_maxlevels(
struct xfs_mount *mp)
{
mp->m_rmap_maxlevels = xfs_btree_compute_maxlevels(mp,
mp->m_rmap_mnr, mp->m_sb.sb_agblocks);
}
/*
* Copyright (c) 2014 Red Hat, Inc.
* All Rights Reserved.
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License as
* published by the Free Software Foundation.
*
* This program is distributed in the hope that it would be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write the Free Software Foundation,
* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
*/
#ifndef __XFS_RMAP_BTREE_H__
#define __XFS_RMAP_BTREE_H__
struct xfs_buf;
struct xfs_btree_cur;
struct xfs_mount;
/* rmaps only exist on crc enabled filesystems */
#define XFS_RMAP_BLOCK_LEN XFS_BTREE_SBLOCK_CRC_LEN
/*
* Record, key, and pointer address macros for btree blocks.
*
* (note that some of these may appear unused, but they are used in userspace)
*/
#define XFS_RMAP_REC_ADDR(block, index) \
((struct xfs_rmap_rec *) \
((char *)(block) + XFS_RMAP_BLOCK_LEN + \
(((index) - 1) * sizeof(struct xfs_rmap_rec))))
#define XFS_RMAP_KEY_ADDR(block, index) \
((struct xfs_rmap_key *) \
((char *)(block) + XFS_RMAP_BLOCK_LEN + \
((index) - 1) * sizeof(struct xfs_rmap_key)))
#define XFS_RMAP_PTR_ADDR(block, index, maxrecs) \
((xfs_rmap_ptr_t *) \
((char *)(block) + XFS_RMAP_BLOCK_LEN + \
(maxrecs) * sizeof(struct xfs_rmap_key) + \
((index) - 1) * sizeof(xfs_rmap_ptr_t)))
struct xfs_btree_cur *xfs_rmapbt_init_cursor(struct xfs_mount *mp,
struct xfs_trans *tp, struct xfs_buf *bp,
xfs_agnumber_t agno);
int xfs_rmapbt_maxrecs(struct xfs_mount *mp, int blocklen, int leaf);
extern void xfs_rmapbt_compute_maxlevels(struct xfs_mount *mp);
#endif /* __XFS_RMAP_BTREE_H__ */
......@@ -37,6 +37,7 @@
#include "xfs_alloc_btree.h"
#include "xfs_ialloc_btree.h"
#include "xfs_log.h"
#include "xfs_rmap_btree.h"
/*
* Physical superblock buffer manipulations. Shared with libxfs in userspace.
......@@ -730,6 +731,11 @@ xfs_sb_mount_common(
mp->m_bmap_dmnr[0] = mp->m_bmap_dmxr[0] / 2;
mp->m_bmap_dmnr[1] = mp->m_bmap_dmxr[1] / 2;
mp->m_rmap_mxr[0] = xfs_rmapbt_maxrecs(mp, sbp->sb_blocksize, 1);
mp->m_rmap_mxr[1] = xfs_rmapbt_maxrecs(mp, sbp->sb_blocksize, 0);
mp->m_rmap_mnr[0] = mp->m_rmap_mxr[0] / 2;
mp->m_rmap_mnr[1] = mp->m_rmap_mxr[1] / 2;
mp->m_bsize = XFS_FSB_TO_BB(mp, 1);
mp->m_ialloc_inos = (int)MAX((__uint16_t)XFS_INODES_PER_CHUNK,
sbp->sb_inopblock);
......
......@@ -38,6 +38,7 @@ extern const struct xfs_buf_ops xfs_agi_buf_ops;
extern const struct xfs_buf_ops xfs_agf_buf_ops;
extern const struct xfs_buf_ops xfs_agfl_buf_ops;
extern const struct xfs_buf_ops xfs_allocbt_buf_ops;
extern const struct xfs_buf_ops xfs_rmapbt_buf_ops;
extern const struct xfs_buf_ops xfs_attr3_leaf_buf_ops;
extern const struct xfs_buf_ops xfs_attr3_rmt_buf_ops;
extern const struct xfs_buf_ops xfs_bmbt_buf_ops;
......@@ -116,6 +117,7 @@ int xfs_log_calc_minimum_size(struct xfs_mount *);
#define XFS_INO_BTREE_REF 3
#define XFS_ALLOC_BTREE_REF 2
#define XFS_BMAP_BTREE_REF 2
#define XFS_RMAP_BTREE_REF 2
#define XFS_DIR_BTREE_REF 2
#define XFS_INO_REF 2
#define XFS_ATTR_BTREE_REF 1
......
......@@ -42,6 +42,7 @@
#include "xfs_trace.h"
#include "xfs_icache.h"
#include "xfs_sysfs.h"
#include "xfs_rmap_btree.h"
static DEFINE_MUTEX(xfs_uuid_table_mutex);
......@@ -682,6 +683,7 @@ xfs_mountfs(
xfs_bmap_compute_maxlevels(mp, XFS_DATA_FORK);
xfs_bmap_compute_maxlevels(mp, XFS_ATTR_FORK);
xfs_ialloc_compute_maxlevels(mp);
xfs_rmapbt_compute_maxlevels(mp);
xfs_set_maxicount(mp);
......
......@@ -116,9 +116,12 @@ typedef struct xfs_mount {
uint m_bmap_dmnr[2]; /* min bmap btree records */
uint m_inobt_mxr[2]; /* max inobt btree records */
uint m_inobt_mnr[2]; /* min inobt btree records */
uint m_rmap_mxr[2]; /* max rmap btree records */
uint m_rmap_mnr[2]; /* min rmap btree records */
uint m_ag_maxlevels; /* XFS_AG_MAXLEVELS */
uint m_bm_maxlevels[2]; /* XFS_BM_MAXLEVELS */
uint m_in_maxlevels; /* max inobt btree levels. */
uint m_rmap_maxlevels; /* max rmap btree levels */
xfs_extlen_t m_ag_prealloc_blocks; /* reserved ag blocks */
struct radix_tree_root m_perag_tree; /* per-ag accounting info */
spinlock_t m_perag_lock; /* lock for m_perag_tree */
......
......@@ -49,11 +49,14 @@ xfs_check_ondisk_structs(void)
XFS_CHECK_STRUCT_SIZE(struct xfs_dsymlink_hdr, 56);
XFS_CHECK_STRUCT_SIZE(struct xfs_inobt_key, 4);
XFS_CHECK_STRUCT_SIZE(struct xfs_inobt_rec, 16);
XFS_CHECK_STRUCT_SIZE(struct xfs_rmap_key, 20);
XFS_CHECK_STRUCT_SIZE(struct xfs_rmap_rec, 24);
XFS_CHECK_STRUCT_SIZE(struct xfs_timestamp, 8);
XFS_CHECK_STRUCT_SIZE(xfs_alloc_key_t, 8);
XFS_CHECK_STRUCT_SIZE(xfs_alloc_ptr_t, 4);
XFS_CHECK_STRUCT_SIZE(xfs_alloc_rec_t, 8);
XFS_CHECK_STRUCT_SIZE(xfs_inobt_ptr_t, 4);
XFS_CHECK_STRUCT_SIZE(xfs_rmap_ptr_t, 4);
/* dir/attr trees */
XFS_CHECK_STRUCT_SIZE(struct xfs_attr3_leaf_hdr, 80);
......
......@@ -2446,6 +2446,8 @@ DECLARE_EVENT_CLASS(xfs_rmap_class,
__entry->owner = oinfo->oi_owner;
__entry->offset = oinfo->oi_offset;
__entry->flags = oinfo->oi_flags;
if (unwritten)
__entry->flags |= XFS_RMAP_UNWRITTEN;
),
TP_printk("dev %d:%d agno %u agbno %u len %u owner %lld offset %llu flags 0x%lx",
MAJOR(__entry->dev), MINOR(__entry->dev),
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment